Deep Mw7.1 earthquake strikes Sabah, Malaysia
A signal from a deeply subducted slab?
We need your help! We are dedicated to providing rapid scientific analysis following earthquakes, but to make this effort sustainable, we need financial support. If you can, please consider upgrading to a paid subscription. If you use our Substack as part of a group, or for teaching, consider a teaching or institutional subscription. Every paid subscription makes a real difference — and unlocks access to the more than three hundred posts in our archive!
Free subscribers still get all of our new posts by email, and all of our posts are free to read for one month. If you have recently lost your job or are a student or researcher without ability to pay, let us know and we will upgrade your subscription to “paid” at no cost.
A magnitude 7.1 earthquake struck below Sabah, Malaysia on February 23rd at 12:57 AM local time (February 22nd UTC).
Usually when we say “below,” we mean up to ~30 kilometers below the Earth’s surface, which covers most of the Earth’s crust. Most earthquakes occur within this outer layer of the Earth, because this is where the rocks are cold enough, and therefore brittle enough, to break under stress, rather than just squishing around slowly.
In the case of this recent earthquake, however, “below” means a full 620 kilometers below Earth’s surface. Because of this great depth, the earthquake source was located at least 620 kilometers away from the nearest person. A very deep earthquake that might otherwise have caused dangerous shaking (if it was shallower) will probably only be felt as a relatively modest event. The USGS estimates that the maximum shaking in this event reached intensity V (moderate), which is generally not enough to cause damage. Comments to the EMSC describe rattling closet doors; spooky, but not dangerous.
Second, the shaking can be felt over a very wide area, decreasing only gradually with distance. This is because of simple geometry: a site located right above the epicenter is 620 kilometers away from the hypocenter (vertically). A site located two hundred kilometers away from the epicenter is still just 650 kilometers away from the hypocenter. (You can use the Pythagorean theorem to calculate the length of the hypotenuse of the triangle: (620^2+200^2)^(1/2). That is only 4% further away from the earthquake source than the epicenter (the point directly above the earthquake origin), which explains why we don’t expect the intensity to decay very quickly with distance.
In practice, what that means is that people reported feeling this earthquake from as far away as Singapore, Thailand, and the Philippines. Some of the felt reports come from as far away as 1,500 kilometers!
Deep earthquakes
It isn’t unusual to see earthquakes below 30 kilometers. In fact, the discovery of deep earthquakes and their distribution inside Earth was one of the keys that unlocked plate tectonics. Crust on the Earth is not static: it forms at spreading centers, and is recycled back into the mantle at subduction zones. As great slabs of oceanic crust sink into the mantle, they carry cold rock and water down with them. Earthquakes trace out most of these subduction zones, occurring within the slabs as they bend, break, and chemically adjust to the higher temperatures and pressures within the mantle. For that reason, it is generally unsurprising when we see earthquakes at 100, 200, even up to 300 kilometers depth, along one of Earth’s great subduction zones. Below that depth, these earthquakes generally stop.
And yet, there is some funny business that can occur in subducting slabs at even greater depths: earthquakes sometimes start up again, at ~500-700 kilometers depth. These very deep focus earthquakes seem like they shouldn’t be possible, because the temperatures are way too hot for traditional brittle deformation. And yet, seismometers detect these events and undeniably place them at these great depths.
This earthquake occurred near the maximum possible depth of earthquakes on… well, Earth. To understand what these earthquakes mean, we need to understand the rocks that host them.
We often think about only the upper parts of subducting slabs: the basaltic crust and marine sediments that form near the surface, and which cause so much action in the upper part of the subduction system.
However, slab is not just crustal rocks. The crust that is subducting is sitting on top of a hugely thick section of mantle rocks, which is also subducting. This includes the mantle lithosphere (which can host earthquakes), and the more flowable asthenosphere below (which does not). These rocks sink downward together during subduction, as part of great mantle convection cells.
What are those mantle rocks like? Well, about fifty percent of the upper mantle is a very nice mineral called olivine: Mg, Fe)2SiO4. (The parentheses mean that each molecule of olivine can have two atoms of magnesium, two atoms of iron, or one of each.) We can find this same mineral at the surface, which in gem form is called peridot, although it is relatively rare in the crust. But in the mantle, it is much more common. We even sometimes find blocks of olivine-rich rock embedded in basaltic volcanoes. These xenoliths (“foreign rocks”) are pieces of the mantle that were entrained in eruptions and carried up to the surface.
As slabs subduct, they carry this olivine down from the upper mantle into deeper region. And as olivine descends into Earth’s interior, the increasing pressure forces major changes in the crystal structure. Oversimplifying, at about 410 km depth, olivine rearranges its crystal structure and transforms into wadsleyite: still (Mg, Fe)2SiO4. At 520 km, there is a transition into ringwoodite: still (Mg, Fe)2SiO4. Then at 660 km depth, the high pressure finally breaks up the four-oxygen rock band, and ringwoodite decomposes into perovskite: (Mg,Fe)SiO3 and ferropericlase: (Mg,Fe)O. Those last two high-pressure minerals tend to flow instead of breaking under stress, and thus the only earthquakes deeper than ~660 kilometers occur only in rare, thick, cold slab remnants.
It’s hard to find a nice cartoon that shows these features, which doesn’t also contain a lot of other tangential information. We like this one from Ohtani and Ishii (2024):
This figure shows how very deep earthquakes occur in the cores of deeply subducted slabs. But what actually happens during these earthquakes?
At the moment the leading hypothesis for how most deep-focus events occur is transformational faulting. The idea is that sudden run-away phase changes in mineral driven far too deep for their own comfort are capable of triggering earthquakes. When these phase changes happen, the atoms within the mineral reorganize into a more compact geometry. This sudden decrease in volume may trigger similar phase changes in neighboring rocks, creating a sudden aligned region where the resisting stresses are suddenly decreased and allowing lateral slip. That explains why these deep earthquakes have sliding mechanisms, rather than implosive mechanisms. However, we should note that the physical mechanisms of deep focus earthquakes are still much debated, and there is a lot of room for new discoveries.
Armed with that general picture, let’s look at the tectonic setting for this M7.1 earthquake.
Deep structure below Sabah
Southeast Asia is encircled by subduction zones, and as a result is home to many deep-focus earthquakes. (We actually prefer to call this the Southeast Asian Ring of Fire, in contrast to the Pacific Ring of Fire.) The map below shows only earthquakes below 500 kilometers depth. There is a line of deep-focus earthquakes to the east of the recent event, associated with the sinking Halmahera slab; these run from the Philippines to northern Sulawesi. There is another long arc of deep-focus earthquakes farther to the south, tracing the deep part of the Sunda slab as it dives beneath the Lesser Sunda Islands, Java, and southern Sumatra.
None of these earthquakes are anywhere near the recent M7.1. And in fact, the USGS Slab2.0 model shown on the map does not have any slabs near the recent earthquake. We apparently have some rogue deep earthquakes on our hands.
We suspect that there is some kind of deeply subducted slab in this area, but the earthquakes are so rare that they don’t map out any clear structure.
Fortunately, deeply subducted slabs can be mapped by a method called seismic tomography. Tomography is a generic term for imaging the interior of an object using penetrating waves. Seismic tomography involves imaging the interior of the Earth using seismic waves, including those generated by earthquakes. In basic terms, we know that the velocities of seismic waves change with depth (usually moving faster in deeper rocks, which are denser). However, when we look at the actual arrival times of earthquake waves, we can see that they sometimes travel faster or slower than we expect based on a generic velocity-versus-depth model. By combining information from many earthquake-receiver paths, we can build a picture of the seismic velocities inside the Earth.
We know that areas with slower seismic waves usually indicate hot material, and faster waves usually indicate cold material. So, when we look at a 3D model generated with seismic velocities, we can generally interpret where deeply subducted slabs rest: they are colder than expected.
So, what can we see of this earthquake area using tomography? A cross-section across the Indonesian archipelago, through the area of today’s deep earthquake, is very interesting. We have quickly overlayed earthquake locations onto a figure by Wu and Suppe (2019), who drew a profile through the tomographic model MITP08 by (Li et al., 2008), passing quite near today’s hypocenter.
Here is the profile, showing the deep earthquake near its end:
And here is the cross section. We have added approximate depths of relevance as well:
On the left, the modern-day Sunda subduction zone is prominent, descending steeply from its highly seismic upper section, down through the metamorphic transitions. Lurking below is a great blue/purple blob, which sits beneath the Sunda Slab. This is a remnant of the Neotethys, the great ocean basin that once existed between India and Eurasia, before the two collided.
Today’s deep earthquake falls quite close to a fast anomaly pointed out by Wu and Suppe, which they interpreted as the remnant of yet another slab, a small one that basically fell down and landed on top of the much larger, deeper remnants of Neotethyan slabs.
Wu and Suppe suggested that our blue blob of interest could be the remnants of the Proto-South China Sea, a somewhat theoretical ocean basin that might once have separated Borneo from the current South China Sea. This Proto-South China Sea is actually quite controversial, and we decline to adopt a strong stance about it here. That might be intellectual cowardice, but it’s also intellectual convenience. (Kyle fondly recalls any number of energetic conversations about the Proto-South China Sea with the late, great Paul Tapponnier, who was perhaps its staunchest detractor.)
In any case, today’s deep earthquake is surely related to some past subduction, whether of the Proto-South China Sea or some other now-lost ocean basin. Regardless of the exact origin of the blue blobs at depth, there is clearly still at least a little seismic potential remaining at these great depths. Over time, as these deep slabs equilibrate to their surroundings, they will inevitably go seismically quiet. However, by that time, new slabs that are still seismically active will probably have joined that great graveyard in the mantle.
Most earthquakes in and around Sabah occur at shallow depths, as we were reminded by the deadly 2015 M6.0 Sabah earthquake, which occurred very close to the recent M7.1 in map view, but was actually more than 600 kilometers away due to the difference in depth. Fortunately, the extremely deep earthquakes like today’s event are mainly intellectual stimulators, and not societal devastators.
References
Bradley, K., Hubbard, J., 2023. One Ring to rule them all? Maybe not!. Earthquake Insights, https://doi.org/10.62481/dac74c77
Li, C., Van Der Hilst, R.D., Engdahl, E.R. and Burdick, S., 2008. A new global model for P wave speed variations in Earth's mantle. Geochemistry, Geophysics, Geosystems, 9(5). https://doi.org/10.1029/2007GC001806
Ohtani, E. and Ishii, T., 2024. Role of water in dynamics of slabs and surrounding mantle. Progress in Earth and Planetary Science, 11(1), p.65. https://doi.org/10.1186/s40645-024-00670-7
Wu, J. and Suppe, J., 2018. Proto-South China Sea plate tectonics using subducted slab constraints from tomography. Journal of Earth Science, 29(6), pp.1304-1318. https://doi.org/10.1007/s12583-017-0813-x
Zhan, Z., 2019. Mechanisms and implications of deep earthquakes. Annual Review of Earth and Planetary Sciences. 48, 147-174, https://doi.org/10.1146/annurev-earth-053018-060314.
Great educational summary of very deep earthquakes.
Thanks for this analysis. Arguably a different topic but related to disaster-prone Asia: What submarine volcanoes in Asia concern you the most? I'm seeing a lot of focus on the Axial Seamount off the U.S. west coast but are there any in Asia that might result in a "black swan" event?