UT researcher better understands earthquakes through computer modeling

Lauren Rahman

A researcher at UT is developing a new way of computer modeling to better understand earthquakes using findings from a 2011 earthquake in Japan.

Thorsten Becker, the lead author of the paper and a professor in the Jackson School of Geosciences, worked in collaboration with a team of researchers from Japan and Purdue University to publish their study in Earth and Planetary Science Letters
in December.

Earthquakes display precursory phenomena, such as foreshocks, which are smaller earthquakes that occur before the big one, that indicate a buildup before an earthquake, Becker said.

“We’ve known about these for a long time, but the problem is that the occurrence of the precursory phenomena isn’t regular,” Becker said. 

Becker said his research was based on studying the 2011 Tohoku-oki earthquake in Japan, one of the biggest recorded, and that this revealed a lot about the deformations of the Earth and foreshocks. 

The patterns that his team found when studying the Japanese earthquake were similar to the patterns they found while studying earthquakes in the lab and on computer models, Becker said. 

“It has been documented that the changes in the ratio between small and large earthquakes before the event was a trend … that happens to be consistent with laboratory work,” Becker said. 

Becker said normally there’s a certain force within the Earth’s crust along subduction zones where earthquakes are likely to occur; however, prior to an earthquake, they saw this force start to change. 

“In some locations in the last three years before the earthquake, they actually slightly changed, and then they snapped,” Becker said.

Becker’s current work is part of a larger research network called Modeling Collaboratory for Subduction, a research group recently funded by the National Science Foundation, said to Gab Lotto, program manager for the organization.

“Its purpose is to bring together earth scientists who model subduction zones from across the US and around the world to collaborate on computer models of earthquakes, volcanoes and related hazards,” Lotto said. “We’re going to be running three interdisciplinary workshops over the next few years to jump start collaborations and ask big questions about what is known and unknown in the science of subduction zone hazards.”

Developing these models can help see if certain parts of a fault are more likely to rupture than another, Becker said.

“The bigger picture here is to use all the information that we can get from looking at smaller earthquakes, from looking at surface deformations and from numerical modeling to build a better physical description of a subduction zone … to have a tool to see these changes next time,” Becker said. 

Geological sciences professor Stephen P. Grand said current earthquake research is not capable of making predictions.

“I would say with fair certainty that there is no accepted way to ‘forecast earthquakes,’” Grand said.

Becker said although these findings can’t be used to predict an earthquake and may not be useful for evacuations before the earthquake, it could be helpful for allocating money toward certain regions susceptible to earthquakes in order to mitigate risk.

“There is some indication that if we were measuring the right thing and have the right physical models, we could make some headway in decadal-scale forecasting,” Becker said.


Want more content like this in your inbox every morning?

* indicates required