Researchers in the U.K., Mexico and the U.S. have found a way to forecast an earthquake up to seven days before the ground shakes.
A University of Cambridge professor said a machine known as an earthquake simulator can listen in for a particular sound before the earthquake happens.
“When you rub rocks together, they emit sounds. And (with) the human ear, you can’t hear most of these sounds,” Sir Colin Humphreys said on CKNW’s The Jon McComb Show.
“Animals can hear some of the sounds, there are stories of animals racing away before an earthquake happens and humans think ‘what are you doing?’ We have detectors that can hear all the sounds emitted before an earthquake actually happens.”
LISTEN: Sir Colin Humphreys joins Jon McComb to speak about earthquake-detection machine
Humphreys said researchers have determined the exact frequency before a quake hits, but they haven’t found a way to determine its magnitude.
“We can’t do that at the moment but we believe the same machine learning applied to the signals coming, if we analyze data going properly, that will tell us the size for the earthquake,” he said.
California’s San Andreas Fault will be one of the first areas where the simulator will be tested.
Humphreys noted the signals can also be used to detect avalanches and landslides.