Aftershock, triggered or coincidence? The relation between large earthquakes

Published: 02/17/2021

One large earthquake can cause other strong earthquakes. The last few days have shown several examples of earthquake triggering and sequences.

During the past couple of days, the world has seen a small increase of seismic activity. Since February 10th, the United States Geological Survey (USGS) has detected 11 earthquakes with a moment Magnitude of 6.0 or higher. Quakes of magnitude 6 are considered as “strong earthquakes”. They usually occur every two to three days, but they do not occur frequently and are often clustered in time, leading to apparent active periods like now. Often, statistical randomness is the only cause of these clusters. But in some cases, there is a direct relationship between such earthquakes which leads to an occurrence shortly after each other. This has also observed over the past few days.

February 10th has been the most active earthquake day for many months. Seven strong earthquakes were detected, six of them clustered near the Loyalty Islands south of Vanuatu. The largest one with Magnitude 7.7 triggered a tsunami warning for nearby coasts but didn’t cause any significant damage. Neither did the fore- and aftershocks or the other strong earthquake on this day, a magnitude 6.2 event southwest of the Indonesian island Sumatra.

Three days later, a magnitude 7.1 earthquake has hit the Japanese coast. It was widely and strongly felt on the main island Honshu, caused damage to more than a thousand structures and injured 157 people. Although the shaking was severe, intensity 6+, the second highest level on the Shindo intensity scale, this quake remained rather harmless without a major tsunami.

Just 90 minutes after Japan, Papua New Guinea was hit by a magnitude 6.0 earthquake. It originated from an offshore fault in save distance to the nearby islands of New Guinea and New Britain. No damage was registered, and no tsunami observed.

Yesterday, on February 16th, the last large quake of this active week occurred. Vanuatu againn, but this time with magnitude 6.2 and west of its capital Port Vila. No damage was registered but similar to the Loyalty Islands quake, sequences preceded and succeeded this event, starting on February 14th and still continuing.

Except for the Sumatra event, all these earthquakes occurred on the south-eastern edge of the Pacific plate. There seems to be a spatial and temporal correlation to a certain degree. Indeed, there are strong indicators that some of these events did not occur independently from other earthquakes. We know examples for two types of related seismicity that can follow large events: Aftershocks and triggered events. But what do they mean and how can we distinguish them?

First, let us start with the more prominent case: Aftershocks. Most people will have heard about it and some might have already experienced some. Aftershocks are a type of earthquakes which occurs after larger earthquakes. Due to its massive energy release and movement, surrounding rock layers can experience additional stress. Depending on the rock properties, this stress can accumulate for a long time or lead to impending new ruptures and thus new earthquakes. This usually occurs only in the direct vicinity of such a mainshock earthquake, usually within the length of the mainshock’s rupture.

In case of the Mw7.7 Loyality Islands earthquakes, the rupture length is about 150 km, according to the USGS Finite Fault model. This means that all following earthquakes which we see within a 150 km radius, including four Mw6.0+, are aftershocks. Aftershocks do not have a maximum magnitude, but they are usually not stronger than the mainshock magnitude minus one. If they become larger than the mainshock, the nomenclature changes and the initial mainshock becomes a foreshock while the following event turns into the new mainshock. There are no geophysical differences between fore-, main- and aftershocks, therefore it is not possible to call a quake a foreshock before the mainshock has happened.

Aftershocks can also last for a long time, sometimes for years or decades. That is why the Mw7.1 Japan quake is also considered as an aftershock. The mainshock, the infamous Mw9.1 Tohoku earthquake, occurred almost ten years ago. Since the earthquake activity near the former rupture zone is still higher than it was before the mainshock (called “background seismicity”), all seismic activity there is considered to be part of the aftershock sequence. Because for every single event there is a given chance that it would not have happened without the 2011 event. However, this kind of definition is solely basing on statistics. There is currently no way to prove that a large earthquake would not have happened if there was no bigger event years ago. Some scientists say that especially in regions with low seismicity, distinct clusters of seismicity which we see for example in China or Europe, are all the remnants of long-lasting aftershock sequences where the mainshock happened centuries ago, maybe even in prehistoric times. They explain it with the significantly lower background seismicity on nearby faults, which should also be applied as a standard for those clusters of higher activity.

However, distinguishing between aftershocks and independent seismicity might be difficult in some cases. The other kind of earthquake relation is even harder to prove: Triggering.

Large earthquakes are able to cause new seismicity that is more than a fault length away, although this kind of activity is less common. These “distant aftershocks” are called triggered events. There are two ways to trigger other earthquakes: 1. By stress increase on nearby fault due to the rock displacement of the mainshock (static triggering) and 2. By shear stress increase due to the passing of high-energetic seismic waves (dynamic triggering). While the first way is limited to a small area that only slightly increases the aftershock radius (we see this usually in long-term earthquake sequences like the Central Italy sequence 2009 – 2019 or the Landers – Ridgecrest sequence in California 1999 – 2019). Dynamic triggering can occur thousands of miles away, even on the other side of the globe. However, earthquakes of magnitude 7.5 or more are needed for distant triggering, otherwise the energy of the seismic waves, which decreases with travel distance, would not be large enough to affect other faults. Fluids, mineralized waters that are present within the earth’s crust, play an important role in dynamic triggering. Therefore, triggered earthquakes often occur in sequences or swarms and preferrable in volcanic areas. They are mostly limited in size, but depending on the given conditions might reach high magnitudes.

Figure 1: Seismicity in Vanuatu since 2020 with two likely related earthquake clusters west of Port Vila

A likely example of triggered earthquake swarms has been seen in Vanuatu. While the sequence west of Port Vila is too far away to be a direct aftershock of the Mw7.7 Loyality Islands quake, its swarm-like characteristic indicates some kind of dynamic triggering. Another long-lasting and strong earthquake sequence which happened between May and September 2020 immediately north of yesterday’s event might play an additional role: This sequence, including two M6+ events, might have loaded the fault segment to the south so that only a little more stress was needed to cause new earthquakes. This little bit of stress which can be applied by seismic waves of a large, distant quake.

Numerical modelling of triggering or aftershock sequence is difficult, needs a lot of data and is often not possible under given conditions. Therefore, it mostly remains unclear whether an earthquake was triggered or not and its likelihood is often defined by simple criteria. Thus, it seems that the Japan-aftershock happened coincidently a few days after the large Vanuatu quake and there is no indication that the Papua New Guinea earthquake is related to one of these previous events. While these criteria are often a good clue, the proof or disproof of earthquake relationships mostly remains missing.

Recent Reports

A deadly combination of earthquake properties and regional characteristics resulted in one of the worst disasters in Afghan history

News agencies report a megaquake at Lake Baikal on June 8th, earthquake surveys don't. Censorship? Propaganda? Warfare? An explanation:

21 damaging earthquakes with the lowest loss statistics since November 2020.

Low seismicity and damage in April 2022. Casualties were mostly related to induced earthquakes and landslides.

An earthquake early warning on the Balkan peninsula has taken many residents by surprise. Which lessons can be learned for Google's plans of a global earthquake warning system?