Seismic swarm near southern tip of April 3 M7.4 in Taiwan
Disquieting seismic sequence raises questions about evolving hazard
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.
Note: since this post was published, several more earthquakes have occurred in this swarm, including a M6.1, M6.0, and M5.8 - i.e. events as large or larger than the ones we discuss below. We are keeping an eye on the evolving situation. (April 23rd, 01:55 UTC)
On April 3, a M7.4 earthquake struck below the eastern coast of Taiwan; the resulting building damage and rockfalls killed 17 people. Since then, hundreds of aftershocks have occurred over an area ~80 km long and 50 km wide, mostly offshore. Although aftershocks were at first rapid-fire - with on average, a M3+ event every four minutes - the events slowly died down to a much lower rate. At two weeks since the mainshocks, only a handful of M3+ earthquakes were occurring each day.
Until today: the Taiwanese Central Weather Administration reports that 85 earthquakes above M3 have occurred on April 22, 12 of them above M5 - these are shown in red on the map below.
Many of these earthquakes are being felt all across Taiwan. For instance, here are the intensity maps for the five earthquakes above M5.4, as reported by the Central Weather Administration. Shaking is of course strongest near the epicenter, but the largest aftershocks are clearly being felt at both the northern and southern tips of the island. Note that the intensity scale used in Taiwan is not the same as that used in the US and Europe, but seems to be the same as or similar to that used in Japan.
It is not surprising for aftershocks to come in clusters, or for larger aftershocks to occur after a period of relatively limited seismicity. What is more surprising - and somewhat disquieting - is the pattern of seismicity in this particular case.
First, the new cluster of seismicity is not a typical mainshock-aftershock sequence. It did not start with a large event and then decay - instead, the earthquakes have stepped up in magnitude over time: first, a M5.5 (17:08 local time), then a M5.7 (18:50 local time), then a M5.9 (22:11 local time). Each of those larger earthquakes was followed by their own sequence of smaller aftershocks.
Second, the new cluster of seismicity is concentrated at the southern end of the April 3 rupture. Until this point, aftershocks have been pretty broadly distributed across a wide area - representing the incremental adjustments of faults around the mainshock to the new stress state, over an area of ~4000 square kilometers. These aftershocks, in contrast, are located in a much smaller area, ~250 square kilometers.
What exactly is going on with the underlying fault system? Swarms of seismicity like this typically reflect some kind of progressive movement of material around or on the faults - for instance, pressurized fluids moving along the fault surface, or some kind of viscoelastic deformation of the surrounding rocks. Viscoelastic simply means a combination of viscous and elastic deformation - the rocks at depth are hot enough, and stressed enough, to slowly flow rather than fracture. This type of flow always happens after large earthquakes, to some extent, and is strongest around areas where the earthquake caused big stress changes in the deep crust. While the flow does slow down over time, it can help transfer stress onto other, nearby parts of a fault system.
As we discussed in our previous post following the mainshock, there are a number of complex faults that lie beneath Taiwan’s Longitudinal Valley. These faults are known to have produced large earthquakes in the past - including a series of four M7+ earthquakes in 1951.
Could the current swarm of seismicity indicate elevated hazard on the faults to the south, which did not rupture on April 3? We think the answer is probably yes, although it is hard to be specific about the reasons. Swarm-like activity in an area where stresses were presumably increased by a recent large earthquake may indicate that these faults are being pushed toward a larger failure.
A recent example of a swarm that transitioned into a large fault rupture was the January 1, 2024 Noto Peninsula earthquake in Japan. There, a highly active series of swarms which grew over several years hosted the nucleation point of the M7.5 mainshock. A lot of research is currently being done to try to understand any potential relationships between the swarms and the eventual mainshock.
So, will another large earthquake occur south of Hualien? We don’t know! It is never really possible to predict whether a large earthquake is going to happen, even when we see unusual earthquake behavior. This is simply because so-called “unusual” behavior is surprisingly common, and large earthquakes are always rare. The chance of a triggered large-magnitude event is probably quite low - but that’s what we always say.
However, this swarm is a certainly good opportunity for people in the region to revisit their recently tested earthquake preparations.
References:
Hubbard, J. and Bradley, K., 2024. Deadly M7.4 earthquake strikes Taiwan. Earthquake Insights, https://doi.org/10.62481/c4a3297f
Hubbard, J. and Bradley, K., 2024. M7.5 earthquake strikes western Japan, triggers tsunami. Earthquake Insights, https://doi.org/10.62481/e8bf9b2e
Heyo, Taiwanese citizen checking in here! Thanks for providing a much needed informed analysis. When all you have is breaking news, it can be hard to actually derive any insights from the chaos.
Just wanted to make a quick note though - I notice you've provided the option to view the article in our native Chinese, as you do with most other articles regarding foreign nations. However the option provided is for Simplified Chinese, which is more commonly used in mainland China, as opposed to the Traditional Chinese we use here in Taiwan. To some, the difference is as minor is using British English spelling vs. American English, but to others its a tad politically sensitive, so just thought I'd raise the issue in case you weren't already aware.