In a peer-reviewed study published in the scholarly journal Remote Sensing in May, Israeli researchers have developed a mechanism to predict earthquakes 48 hours in advance with 80% accuracy.
The Ariel University and Center for Research & Development Eastern Branch research team were able to assess potential triggers for several significant earthquakes that occurred in the last 20 years by examining changes in the Earth’s ionosphere which is the thin layer of atmosphere that meets the vacuum of space.
Major earthquakes were identified by the researchers as those that were greater than Mw 6 on the Moment magnitude scale, which gauges an earthquake’s size based on seismic displacement.
The team proposed to use a machine learning support vector machine (SVM) algorithm to determine the electron charge density of the ionospheric total electron content using GPS map data.
They found that an earthquake may be forecasted with an accuracy of 80% using this method.
Additionally, with an accuracy rate of 85.7%, the researchers were also able to forecast when there won’t be an earthquake in a certain location.