Subduction Zone Segmentation

There are a lot of things that our scientific community will do with these data collected right after the earthquake. Recordings of aftershocks and other earthquakes (nearby and far away) will tell us more about the geometry of the megathrust (the main fault separating the plates at depth), properties of the earthquakes themselves, and about the structure and properties of the material above and below it. We want to know why the earthquake ruptures started and stopped where they did, and how each earthquake has affected the stresses (internal forces) that will ultimately cause the next one. And we hope to gain new insights into how and when slip occurs on the megathrust, during earthquakes and between them, which is the topic of this post.

We have known for years that the slip behavior of the megathrust between earthquakes changed systematically from the northeast to the southwest along the Alaska Peninsula. In the northeast, a very wide section of the fault is stuck together by friction, but in the southwest the fault is mostly creeping along steadily at about the rate of plate motion. There are various ideas about the cause of that, but we still don’t know how much that transition is smooth and gradual as opposed to being abrupt, or if the transitions at all different depths are the same or not. We really need to know this better before we can evaluate ideas for what controls it, and move beyond noting correlation to understanding causes.

A few years ago, one of my former students (Shanshan Li) and I proposed that the changes in slip behavior along the length of the subduction zone were segmented and abrupt (Li and Freymueller, 2018). This was mainly based on the Shumagin Islands, where we had the best data. Later, one of my current students (Connor Drooff) added some data from a nearby volcano, corrected for the deformation caused by the volcano, and produced an updated segmentation model with five segments (Drooff and Freymueller, 2021). You can see the model in the gray shading background in the figure below; the darker areas are more stuck than the lighter areas. He did that work back in 2019, but it took a while to get it written up and published. The figure also shows contours of slip from the Simeonof earthquake (in purple), from a paper currently in review (Xiao et al., in review).

When the Simeonof earthquake occurred in July 2020, we could see that the area that slipped mostly corresponded to one of Connor’s segments. Mostly, because although the edge of the rupture near the area of highest slip stopped very close to the segment boundary, at greater depth (to the north) the 2020 rupture crossed over the boundary and continued on, although with less slip. So the segmented model seems to have gotten it partially right, but the real world is a bit more complicated.

The background gray scale shows the interseismic slip deficit model from Drooff and Freymueller (2021), with darker colors indicating a larger slip deficit and light colors indicating more creep, and aftershocks of the July 2021 Chignik earthquake (from the Alaska Earthquake Center) shown in red. The dashed blue lines are the segment boundaries in the model. The purple outlines are contours of slip in the July 2020 Simeonof earthquake (each 0.5 meters), from Xiao et al. (submitted). The orange outlines show the slip area of the 1938 earthquake. The dotted outline is the aftershock zone as drawn by Davies et al. (1981). The solid and dashed outlines are the preferred and alternate model high slip areas (>2.5 m slip) from Freymueller et al. (2021), based on tsunami modeling. The far-field tsunami arrivals required that the main slip in 1938 was shallow.

What about the July 2021 Chignik earthquake? Getting a detailed slip model for that earthquake will require more data from close to the source, which we will be getting soon. But we can say something based on the aftershock distribution, because the aftershocks usually approximate the area of slip. We can also look at initial rough slip models like the USGS finite fault model. The USGS model suggests that the slip stopped at about the Semidi Islands, which would mean that the 2021 Chignik earthquake more or less ruptured another one of the segments in Connor’s model, and roughly the same range of depths (20-40 km) as the July 2020 Simeonof earthquake. But the aftershocks extend further east, out to about Chirikof Island. We’ll know more after more data are collected and analyzed.

Another interesting question is how this earthquake relates to the 1938 M8.3 earthquake along the same area. The figure shows three orange regions for the 1938 slip area. The big dotted outline is the aftershock zone as drawn by Davies et al. (1981), but that is likely much bigger than the actual slip area. The solid and dashed outlines are the preferred and alternate models (slip > 2.5 m) from Freymueller et al. (2021) based on modeling the tsunami. That study found that the main slip in 1938 was shallow, and more likely in the easternmost part of the Davies outline. It might be that the Chignik earthquake rupture area didn’t overlap at all with the 1938 high slip zone. Answering that question will be another focus of future research with this data. The aftershocks show some clusters that extend south toward the trench into that shallower part of the megathrust. Are those areas of shallower slip, or maybe afterslip – creep that follows the earthquake?

Meanwhile, the seismology team has arrived in Kodiak, and preparing to head out to install the first stations. Expect to hear from them tomorrow!

Jeff Freymueller, Michigan State University


2 thoughts on “Subduction Zone Segmentation

  1. Great post Jeff! This sets the stage very nicely.


  2. Very nice and clear write up Jeff! Good luck with the upcoming detailed investigations!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this:
search previous next tag category expand menu location phone mail time cart zoom edit close