M7.1 earthquake strikes southern Japan; megaquake advisory issued

An indication of elevated risk - but not a prediction

Earthquake Insights is an ad-free newsletter written by two earthquake scientists. Our posts are written for a general audience, with some advanced science thrown in! To get these posts delivered by email, become a free subscriber. If you would like to support our work here, please also consider a paid subscription.

この投稿を日本語（Google による自動翻訳）で読むには、ここをクリックしてください。

---

Citation: Bradley, K., Hubbard, J., 2024. M7.1 earthquake strikes southern Japan; megaquake advisory issued. Earthquake Insights, https://doi.org/10.62481/cea4a692

---

A magnitude 7.1 earthquake struck southern Japan at 16:42 local time on August 8, 2024 (07:42 UTC). The USGS and New York Times originally reported two events of similar magnitudes, but the USGS later corrected the catalog to show only a single event. The USGS PAGER estimates that >800,000 people experienced very strong shaking, but because of Japan’s strong building codes, fatalities are not likely. A 1 meter tsunami (peak to peak) has been measured at a tide gauge station near Miyazaki. News agencies have not reported major damage. The USGS forecasts that the shaking may have caused significant liquefaction across southern Kyushu.

The quake occurred below the eastern coast of Kyushu, within the Nankai subduction zone. We previously wrote about earthquakes in this general region in May 2023 (M5.9 offshore Okinawa) and April 2024 (M6.4 below the Bungo Channel).

The earthquake was felt across southern Japan, with strong shaking near the epicenter. On the map below, felt intensities are shown on the shindo intensity scale. The strongest shaking reached 6- (approximately equal to a Modified Mercalli Intensity of VIII, the scale more commonly used in the US and Europe). Shaking at this level can cause some structural damage, especially for houses with low earthquake resistance. We note that the vast majority of areas that experienced shaking reported intensities of 5- or lower, which is not expected to cause structural damage.

Figure 1: Map of ground shaking. Source: Japanese Meteorological Agency. Note that the intensities here are reported on the shindo intensity scale. (Edited per the comment below from Dr. Wald.)

Megaquake advisory

This is a particularly interesting moment in the history of earthquake preparedness. For the first time since a new protocol was established in 2019, the Japanese Meteorological Agency has issued a “megaquake advisory”: a warning that the likelihood of a M8-9 earthquake on the plate boundary near Nankai is higher than usual. If you can read Japanese, you may want to browse this document provided by the JMA, which appears to be packed with interesting figures.

The JMA is mostly advising people to be prepared in general by taking common-sense measures like securing furniture and confirming the locations of evacuation shelters. These sorts of measures are good to do at any time in a seismically active location. However, they are also advising voluntary evacuation for some vulnerable groups of people, like people with physical disabilities, elderly people, and children. It is unclear how long such a recommendation will remain in place.

This is the first time that we have seen such an advisory, and it will be interesting to see how people react to it. Government institutions will surely react — as with the slowing of the bullet trains — but will regular people change their behavior? We will have to wait and see.

Why the advisory was issued

Those who have been following earthquake news for some time may recall that the 2011 M9.1 Tohoku-Oki earthquake was also preceded by a magnitude 7.2 foreshock. That earlier earthquake was not seen as being particularly notable, and produced little immediate commentary — and no advisory — over the two days leading up to the deadly mainshock. Had an advisory been issued then, it is likely that a significant number of lives would have been saved, because people would have been more prepared to react when the mainshock happened. Furthermore, some of the infrastructural catastrophes that followed the 2011 M9.1 earthquake might have been avoided with two days of preparation.

That example clearly motivates the recent advisory. The earthquake setting, though is slightly different.

The Nankai Trough is a subduction zone with a known history of large (M8+, up to M9+) earthquakes. Historical and geological records indicate that, every century or so, part of the Nankai Trough will probably experience a great megathrust earthquake. A future great Nankai earthquake is surely the most long-anticipated earthquake in history — it is the original definition of the “Big One”. Previous estimates of potential fatalities from such an event — without any warning or evacuation — sometimes exceed 240,000. So, it’s a big deal.

While earthquake prediction is impossible, the occurrence of one earthquake usually does raise the likelihood of another. One of the challenges is that even when the risk of a second earthquake is elevated, it is still always low — for instance, in California the rule of thumb is that any given earthquake has ~5% chance of being a foreshock. So, the advisory was issued because there is a small probability that the M7.1 earthquake is a foreshock to a much larger and more dangerous earthquake.

The history of great earthquakes at Nankai is convincingly scary. Here is a nice figure that we have modified from Figure 1 of Garrett et al. (2016). That paper summarized (in great detail) the geological evidence for the size of historical Nankai-Suruga ruptures.

Figure 2: Map of the Nankai Trough showing locations where geological evidence of great earthquakes has been identified. Right: list of known great earthquakes on the Nankai Trough. Source: Figure 1 of Garrett et al. (2016).

The map at left shows the different segments of the subduction zone, along with the places where different geological observations have been made. The August 8 earthquake occurred at the western edge of Segment Z - the Hyūga-nada segment.

The panel at right shows the segments that appear to have ruptured together, during different historical earthquakes. Over the last 1,340 years, there have been at least 12 great earthquakes. That gives an average recurrence of about 111 years between events, with the last event 78 years ago. But this calculation isn’t really useful, because there are obviously some very large gaps between events (more than 250 years), and some very small gaps (1 day, in 1854). Looking at individual segments, the gaps can be much longer, well over 500 years.

This temporal and spatial irregularity is a common feature of large earthquake recurrence, and while it is frustrating, it is also one of the things that keeps seismologists and earthquake geologists in business. Paleo-earthquake studies are never definitive — it is very difficult to study ancient earthquakes, because the scars they leave on the landscape are difficult to discover, date, and interpret. But the main message is that the Nankai Trough can actually produce huge (M9+) ruptures that extend over more than 600 km, and will do so sometime in the unknown future.

With that in mind, the setting of the M7.1 earthquake is actually quite interesting, and relevant to the seismic hazard. So, let’s take a closer look.

The tectonic context matters

The M7.1 earthquake occurred right at the boundary of the Nankai Trough and Ryukyu Trench. Along both of these subduction zones, the Philippine Sea Plate is subducting to the northwest beneath Japan. However, near the location of the August 8 earthquake, there is a tremendous change in the nature of the subduction zone from west to east. A plot of seismicity immediately highlights one of the differences: the shallow part of the Ryukyu Trench is seismically noisy, while that of the southern Nankai Trough is notably silent.

Figure 3: Seismicity around the M7.1 earthquake in map view and cross-section. Curved line in the cross-section is subduction interface model from slab2.0. Earthquakes are from the JMA catalog; focal mechanisms are from EMSC/GCMT/GFZ/ISC.

This change in seismic behavior along the subduction zone is probably controlled by the geology of the subducting plate. The Philippine Sea Plate is a bit like Frankenstein’s monster — it has been stitched together out of various pieces.

The eastern Philippine Sea Plate — the part that is subducting at the Nankai Trough — is smooth and flat. This region is made up of oceanic crust that formed after ~22 million years ago, due to rapid roll-back of the Pacific slab to the east. On the map below, this young oceanic crust is shaded yellow, with a gray stripe every one million years. The relatively flat seafloor in this area is called the Shikoku Basin.

Figure 4: Left: Map showing the age of the ocean floor. Arrows show plate motion relative to southern Japan, in mm/yr. Black rectangle is location of figure on the right; black circles show earthquakes <100 km deep and red star is the recent M7.2.

In contrast, the Philippine Sea Plate to the west — where it is subducting at the Ryukyu Trench — is much more mountainous and rugged. This crust is much older, having formed in Mesozoic and early Tertiary time (i.e. ~50+ million years ago). This area is called the West Philippine Basin.

The boundary between the Shikoku Basin and the West Philippine Basin is a long underwater mountain range, called the Kyushu-Palau Ridge. This ridge marks a spectacular, 2,500 kilometer-long tear in the Earth’s crust. When the Pacific Plate to the east rolled backward, the overriding plate broke, creating a new spreading ridge and progressively forming the Shikoku Basin. That spreading ridge is now dead, with active spreading only happening far to the south, behind the Mariana Trench.

Near Kyushu, the northernmost part of this ridge is being subducted beneath Japan. This marks the dividing line between the Nankai Trough and the Ryukyu Trench.

This basic geological context is clearly fundamental to the seismicity of the region, but the interpretation isn’t simple.

At first glance, it might seem like the rougher terrain of the West Philippine Basin could be causing greater friction when subducted, thus producing more earthquakes. In contrast, the smoother subducted surface of the Shikoku Basin might seem to be sliding more easily and with less friction, thus producing fewer earthquakes. However, this simple idea is dangerously wrong. In fact, we often see that rough topography on a subduction zone actually decreases the potential for very large earthquakes, because large areas of the fault can’t stay frictionally locked for long periods of time. In contrast, smooth topography can allow widespread locking, suppressing the small earthquakes but eventually producing mega-ruptures. This is the case at Nankai, which is a real mega-quake machine - but is otherwise almost silent!

Magnitude 7-7.5 earthquakes have occurred at Hyūga-nada before — about half a dozen times over the last century — and those earthquakes did not trigger megaquakes. This specific location between the Ryukyu Trench and Nankai Trough seems to have actually been more productive than many other areas along the two subduction zones. While most of those events are too old to have focal mechanisms, their depths make them likely megathrust events. (We note that the depth of the recent M7.1 places it above the slab2.0 model for the subduction zone. However, given the uncertainty in the megathrust depth, and the fact that the low-angle nodal plane of the earthquake would match with a megathrust, we are assuming that this earthquake occurred on the megathrust.) In 1996, a slightly smaller (M6.7) earthquake was recorded very close to the location of the recent M7.1.

Figure 5: M6+ earthquakes in the western Nankai Trough. Earthquakes above M7 are labelled.

To our knowledge, there is no evidence that previous Hyūga-nada earthquakes have ever triggered large, multi-segment Nankai ruptures. It makes sense to us that the subducted ridge offshore Hyūga-nada, or some aspect of the megathrust arising from that geology, might limit the communication between the Hyūga-nada earthquakes and the Nankai megathrust. So, if no subsequent foreshock emerges, it won’t be that surprising.

However, there is also no evidence that smaller Hyūga-nada earthquakes have never preceded large ruptures, in pre-instrumental times (excuse the double negative). The Earth constantly surprises us, probably because science has only operated long enough to see a fraction of the possibilities. We have to humbly accept that our models and expectations of how earthquakes start are simply guesses, incomplete and unreliable for predicting the future.

Thus, the earthquake advisory seems extremely reasonable to us — while the chance of an imminent megarupture is probably still objectively low, it is probably also higher than before this M7.1 occurred.

References:

Garrett, E., Fujiwara, O., Garrett, P., Heyvaert, V.M., Shishikura, M., Yokoyama, Y., Hubert-Ferrari, A., Brückner, H., Nakamura, A., De Batist, M. and QuakeRecNankai Team, 2016. A systematic review of geological evidence for Holocene earthquakes and tsunamis along the Nankai-Suruga Trough, Japan. Earth-Science Reviews, 159, pp.337-357. https://doi.org/10.1016/j.earscirev.2016.06.011

Hubbard, J. and Bradley, K., 2023. M5.9 earthquake offshore Okinawa, Japan preceded by flurry of foreshocks. Earthquake Insights, https://doi.org/10.62481/119e865b

Hubbard, J. and Bradley, K., 2024. Southern Japan shaken by M6.4 earthquake. Earthquake Insights, https://doi.org/10.62481/134d40ae