Aftershocks of ancient earthquakes

Any major earthquake is likely to be followed by aftershocks. Survivors of seismic devastation live in dread of them for weeks, even months. In reality the fault responsible for the initial event continues to move for longer than that. Commonly, aftershock activity dies down in magnitude and frequency over time, sometimes after a few weeks and in other cases much later to reach ‘normal background seismicity’ for the associated tectonic setting. Near a major plate boundary, such as the San Andreas Fault system in coastal California or the mid-Atlantic Ridge in Iceland, there is a continual risk of damaging seismic events, but the area around each major event becomes less risky a few tens of years afterwards. For instance, the Loma Prieta area on the San Andreas became quiescent sixteen years after the October 1989 Magnitude 6.9 earthquake that wrought havoc in San Francisco – and interrupted a Major League baseball match in the city. The December 1954, Magnitude 7.3 Dixie Valley earthquake in the active extensional zone of Nevada had a longer period of instability: 48 years. There is no fixed period for the aftermath, seismicity ‘stops when it stops’.

Earthquakes of greater than Magnitude 2.5 in eastern North America (see key to magnitudes at lower right). Those shown in blue date from 1568 to 1979, those in red between 1980 and 2016. (Credit: Chen & Liu, Fig 1)

Sometimes devastating earthquakes take place in what seem to be the least likely places: in tectonically ‘stable’ continental plate interiors. A Magnitude 7.9 earthquake in Sichuan Province, central China on 12 May 2008 left 86 thousand dead or missing, 374 thousand injured and 4.2 million homeless. It occurred in a region whose ancient fault systems had had little if any historic activity. One of the best studied records of seismic events in the middle of a continent is in the Mississippi River valley at the Missouri-Kentucky border, USA, near the town of New Madrid. This experienced three major earthquakes in 1811 and 1812 at Magnitudes estimated from 7.0 to 7.4. Seismicity there has continued ever since. Others that occurred long ago in the ‘stable’  North American continental crust were in South Carolina (1886) and southern Quebec, Canada (1663). They and the subsequent, lesser earthquakes that define clusters up to 250 km around them have been studied using spatial statistics (Chen, Y. & Liu, M. 2023. Long-Lived Aftershocks in the New Madrid seismic Zone and the Rest of Stable North America. Journal of Geophysics Research: Solid Earth, v. 128; DOI: 10.1029/2023JB026482). Yuxuan Chen and Mian Lui of Wuhan University, China and the University of Missouri, USA considered the dates of historic events, their estimated magnitudes and their proximity to other events in each cluster. The closer two events are the greater the chance that the later one is an aftershock of the first, although the relationship may also indicate a long-lived deformation process responsible for both. The authors suggest that this ‘nearest-neighbour’ approach may reveal that up to 65% of earthquakes in the New Madrid zone between 1980 and 2016 are aftershocks of the 1811-1812 major earthquake cluster, and a significant number of modern events in South Carolina could similarly relate to the 1886 Charleston earthquake. On the other hand, small modern earthquakes in Quebec are more likely to be part of the regional seismic background than to have any relationship to the large 17th century event.

Earthquakes are manifestations of deep-seated processes, most usually the build-up and release of strain in the lithosphere. If such processes persist they can result in long-lived earthquake swarms. So both delayed aftershocks and a high background of seismicity can contribute to the mapped clusters of historic events: a blend of relics of the past and modern deformation. They are yet to be detected in earthquake records associated with tectonic plate boundaries. A long history of movements within continents suggests that it is possible that long-delayed aftershocks may masquerade as foreshocks that presage greater events that are pending. Chen and Liu’s nearest-neighbour approach may therefore distinguish false alarms from real risk of major seismic motions.

See also: Some of today’s earthquakes may be aftershocks from quakes in the 1800s. Eureka|Alert, 13 November 2023