GMT
5
:
19
/
8
:
19

What science predicts earthquakes

What science predicts earthquakes

The science that studies earthquakes, their consequences, and the ability to predict their occurrence is known as seismology. There are several branches of seismology:

  • instrumental seismology, that studies and measures ground motion and detects, localizes, and studies earthquakes around the world. Research is aimed at studying the movement of plates, the stresses between plates, the nature of their boundaries and the morphology of internal faults;
  • historical seismology, that searches for evidence and traces of earthquakes in old documents, as well as in the architectural heritage;
  • paleoseismology studies very ancient earthquakes based on the traces found in the landscape or the planet’s structure.

Understanding both current and past catastrophes allows seismologists to identify risk zones and predict the consequences of earthquakes globally and in a particular region.

Why earthquakes are difficult to predict

Seismology, like any science, is based on the processing of information obtained as a result of observations. Thus, meteorology focused on weather prediction, uses data such as temperature, pressure, humidity, and air flows. The reliable and accurate weather forecast is compiled based on collected information.

The dynamics of lithospheric plate movement and processes occurring in the Earth’s crust are hidden from scientists. Therefore, any data obtained from seismographs or GPS stations represent only a part of the overall picture of what is happening in the depths of our planet. Geophysical instruments measure surface phenomena that occur beneath several kilometers of solid rock. The results obtained from these studies are indirect and remote – seismologists do not have direct access to fault zones where earthquakes occur.

For this reason, despite research covering an extremely wide range of areas, seismologists still cannot provide a short-term forecast (a few hours, days or weeks in advance) for the occurrence of earthquakes. The lack of direct data does not allow for precise determination of the location, date, and probable magnitude of the event, making it difficult to prepare or evacuate the population in advance. For this, scientists utilize reliable information obtained from precursor signals – phenomena that systematically precede the occurrence of an earthquake.

New earthquake prediction tools

Seismologists propose numerous theories that are still the subject of research and scientific debates. Thus, experts involved in earthquake prediction, study:

  • the increase (or decrease) in the number of small tremors immediately before a major earthquake;
  • the emergence of a zone of slow and low-amplitude tectonic plate movement near the main shock;
  • the changes in the electromagnetic field;
  • the changes in groundwater circulation or gas concentration, and so on.

Scientists are actively studying the possibility of earthquake prediction based on animal behavior. Abnormal animal behavior is often considered a potential precursor of earthquakes. But the results of all studies cannot provide a sufficiently accurate forecast of events. In various cases, certain precursor phenomena were detected before some earthquakes, but were absent during other events. For this reason, none of them can be integrated into operational warning systems. For this reason, none of them can be integrated into early warning systems.

In 2017, seismologists discovered signals called PEGS (Prompt Elasto-Gravity Signals). With their help, new data can be obtained for faster and more reliable assessment of the magnitude of strong tremors. When an earthquake occurs, a huge mass of rock layers suddenly begins to move, causing a disruption in the Earth’s gravitational field. This extremely weak disturbance propagates as a gravitational wave accelerating to the speed of light.

Digital seismometers can record PEGS signals, making them an ideal tool in an earthquake warning system. But detection of gravitational signals is difficult due to their very low amplitude. The peaks of their oscillations are 1 million times smaller than those of P-waves, which are sensed by animals and some people. Is it possible to extract such weak signals from a huge range of acoustic noise? This problem can be solved by new technology such as Artificial Intelligence (AI).

Methods of protecting the population

Unable to predict earthquakes in the short term, seismology has developed a new effective strategy. It involves alerting the population immediately after an earthquake occurs, but before seismic waves reach inhabited areas.

In several regions of the globe, earthquakes occur relatively far from populated areas (several tens of kilometers). This is particularly relevant to so-called subduction shocks that form in faults within coastal zones. Since destructive seismic waves propagate at speeds of several kilometers per second, it is theoretically possible to sound an alarm a few seconds before an earthquake begins.

This time is enough for emergency shutdown systems (of trains, gas networks, elevators, etc.) to activate, as well as to instruct the population to get someplace safe. Such earthquake early warning systems have already been deployed and are being tested in the most seismically active regions, such as the West Coast of the United States and Japan.