Israeli researchers have discovered a method to predict earthquakes 48 hours ahead with 80% accuracy, as detailed in a peer-reviewed study published in the scientific journal Remote Sensing in May.
How did they work it out?
By studying changes in the Earth's ionosphere, the sliver of atmosphere which meets the vacuum of space, the Ariel University and Center for Research & Development Eastern Branch research team was able to evaluate potential precursors to several major earthquakes that occurred in the past 20 years.
The researchers defined major earthquakes as those surpassing Mw 6 on the Moment magnitude scale, which measures an earthquake's magnitude based on its seismic movement.
The method developed by the team saw them implement a machine learning support vector machine (SVM) technique, applied with GPS map data of ionospheric total electron content to calculate its electron charge density.
Through this technique, they managed to find that an earthquake can be accurately predicted with 80% accuracy.
Moreover, the researchers were also able to accurately predict when an earthquake will not occur in a specified area with 85.7% accuracy.
The study was led by Dr. Yuval Reuvani, Dr. Li-Ad Gotlieb and Dr. Nimrod Inbar of Ariel University and the Center for Research & Development Eastern Branch and doctoral student Said Asali from the Department of Computer Science at Ariel University. The research was funded by Israel's Energy Ministry and the Israel Science Foundation.