Geochemistry, Mineralogy, Volcanology News

What Controls Giant Subduction Earthquakes?

Subduction zones with a low dipping angle and thick sediments can produce giant earthquakes; this finding lets researchers estimate worst-case scenarios for coastlines around the world.

By

Giant earthquakes—those greater than magnitude 8.5—are rare. That’s good news for people living on the coastlines along subduction zones where giant earthquakes occur but bad news for geophysicists who want to understand where and why they strike. Now, a new study that models seismic activity in subduction zones has pinpointed the factors responsible for Earth’s largest earthquakes.

“We have just a few hundred of these very big events over the whole history,” said Andreas Schäfer, a disaster researcher at Karlsruhe Institute of Technology who was not involved in the new study. “Empirically speaking, that’s not a lot of data.”

To sidestep this data problem, Iskander Muldashev, a geophysical modeler at Bremen University, and Stephan Sobolev, a geodynamic modeler at GFZ Helmholtz Centre Potsdam, developed numerical models that simulate seismic cycles for subduction zones. The models showed that a shallow angle of subduction for the sinking oceanic plate and a thick layer of sediments in the trench where it meets the continental plate were the most important factors in creating a large rupture zone, leading to giant subduction earthquakes.

Earthquake Surprises

Two of the largest earthquakes (and subsequent tsunamis) ever observed occurred in the past 2 decades, the 2004 Sumatra earthquake and the 2011 Tohoku earthquake. Both had an estimated magnitude of 9.1, which surprised scientists. “No one expected such large earthquakes at those places,” said Sobolev.

Influential research dating back to 1980 proposed that earthquake magnitude depended on the age of the subducting plate and the rate of subduction. Specifically, a young oceanic plate with a rapid rate of subduction was estimated to produce the biggest earthquakes. But conditions at the Sumatran and Japanese subduction zones didn’t fit into this classical view.

“That idea, which was wonderful in its simplicity, didn’t work,” said Sobolev. “So, the question: What are the controlling factors?”

In their new paper, published last month in Geochemistry, Geophysics, Geosystems, Muldashev and Sobolev used a 2D cross-scale numerical model they developed when Muldashev was a graduate student at Potsdam. It simulates subduction processes on timescales of millions of years but can also zoom in to timescales as small as 40 seconds to capture the activity of earthquakes. They varied multiple factors, including subduction rates, the geometry of the subduction zone, and the amount of friction between the plates, to see which factors led to earthquakes with the greatest magnitude.

A map of the continents surrounding the Pacific Ocean, with highlighted subduction zones where giant earthquakes may occur.
This map displays subduction zones predicted to host earthquakes with maximum magnitudes of 8.8–9.2 and more than 9.2. Circles show the locations of previous earthquakes with magnitudes greater than 8.5 in subduction zones. Red circles indicate compressive upper plate strain (UPS), and green circles indicate neutral UPS. Dotted circles indicate preinstrumental events. Credit: Muldashev and Sobolev, 2020, https://doi.org/10.1029/2020GC009145, CC BY 4.0

The modeling pointed to the angle of subduction of the oceanic plate as the most important factor—the flatter the dipping angle of the slab is, the larger the possible magnitude of the earthquake is. This is because with a shallow angle, the slab will have a longer surface within the temperature range capable of generating earthquakes, creating a wider seismogenic zone. A low level of friction in the subduction zone is also important for creating giant earthquakes, so a less rough ocean bottom or thick sediments that can smooth over a rough subducted seafloor were also critical. These characteristics allowed the rupture to travel deeper, which also increased the rupture’s width.

Although the results contradict the classical view of giant earthquakes from the 1980s, they confirm findings from recent numerical modeling efforts and statistical analyses and point to the overall size of the rupture zone as the key to producing giant earthquakes.

One limitation of the study is that because of the complexity of the models and the constraints of available computing power, the model is 2D, though the researchers extrapolated the results into 3D. “These 3D models, in my view, are still stretched 2D models that don’t really capture the 3D complexity of real subduction zones,” said Wouter Schellart, a geodynamicist at Vrije Universiteit Amsterdam. To take the next step, Schellart thinks researchers should extend the models into three dimensions, taking into account variables that might affect the estimated magnitudes, such as the curvature of the subduction zone or irregularities in the plate boundaries.

What’s the Worst That Could Happen?

Muldashev and Sobolev applied their findings to estimate potential worst-case earthquake scenarios for subduction zones worldwide and developed maps highlighting areas where giant earthquakes have the potential to occur. The areas align well with the locations of giant earthquakes from the 20th and 21st centuries and with similar maps based on statistical analyses of earthquake observations. The agreement suggests that the community may be getting a better handle on what controls the size of giant earthquakes and where they might strike. “From a scientific perspective, it is good to know that we are making progress,” said Schäfer.

Muldashev cautions, however, that we still don’t know enough to predict where future giant earthquakes will occur with any precision. “So far, with the tools and knowledge and the records that we have, we cannot make good predictions,” he said, “but this is one step forward.”

—Patricia Waldron (@PatriciaWaldron), Science Writer

Citation: Waldron, P. (2020), What controls giant subduction earthquakes?, Eos, 101, https://doi.org/10.1029/2020EO150390. Published on 15 October 2020.
Text © 2020. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.