We had an earthquake here last week. It was a relatively tiny one, only a 4.2 moment magnitude, but it shook the house, rattled the windows, and made my desk chair sway a bit. I don’t panic. I took too much geology to panic. I tend to feel excited, even when it’s a sizable quake. What we get here in the San Francisco Bay Area on our lovely San Andreas transform fault is still nothing compared to what happens in some of the worst quakes along the ring of fire. These occur where one plate subducts under another, essentially diving down below it. Those quakes are tremendous, and they rattle places like Alaska, Japan, and Chile with alarming frequency.

One of the ways we track the results of these quakes, and in fact, most quakes in places where there are usually quakes, is a sophisticated and extensive network of GPS monitoring systems. The ground-based sensors are linked to GPS satellites in orbit, and when a quake happens, even the slightest change in position or elevation is recorded, and we can piece together just how the ground moved as a result.

Usually, scientists work their way forward from the type of rock, its viscosity, the layer’s thickness, and what not to figure out why the GPS data changed the way it did. This is a forward model. Now a team of scientists has applied a new method, an inverse method, what is called “data inversion”. They take all of that post-quake positional data and work backward to measure how the positions changed over time, and what kinds of stress and strain in the rocks would react that way. Instead of predicting earthquake reactions, they work to understand the properties of the plates that led to the deformation.

It turns out that this method requires far less computational power and time than the forward model. The team applied it to an 8.8 moment magnitude quake in Chile back in 2010, and they were able to see how the viscosity of the two plates changed over time. Using their method, they are now able to increase how much we know and understand about subduction zones, and it’s another piece of the puzzle for maybe, possibly, someday being able to determine where earthquakes will occur.

More Information

Eos article