The vibrating of the surface of the Earth, resulting from the sudden release of some kind of hidden energy in the Earth's lithosphere that creates seismic waves is termed as earthquake. Predicting the earthquakes accurately has always been a challenge for scientists. A fault is a fracture or zone of fractures between two different blocks of rock. It allows the blocks to move towards each other.
That movement may be rapid in the form of an earthquake or slowly in the form of creep. Most of the times scientists fail to monitor the fault’s failure. But now researchers have developed a computer science approach using machine learning that is able to predict the fault’s failure in an early stage. It creates alerts by listening to the acoustic signal discharged by the laboratory-created earthquake. It also identifies new signals forecasting information throughout the earthquake cycle.
Machine learning algorithms can predict fault’s failure at times of laboratory quakes with exceptional accuracy. It is a new physics of failure, by way of examination of the recorded audible signal from the experimental setup in the lab. This research is being considered a milestone in detecting the happening of earthquake that always brings calamity to people, before it’s too late. It may also help in minimizing threats from weapons of mass destruction, and solving issues related to energy, environment, and global security threats.
By: Anita Aishvarya