A new study published in Science presents the most extensive global survey ever completed of earthquakes that originate within the continental mantle. These events occur far below the crust but outside traditional subduction zones. For decades they were considered rare and difficult to verify because routine earthquake catalogs often misrepresent depths. The new analysis shows that these deep continental events are not isolated anomalies. The researchers identified 459 mantle earthquakes since 1990 across a wide range of tectonic settings.
The study analyzed more than 46,000 continental earthquakes recorded since 1990. Out of this starting group, 10,770 events met strict quality standards for waveform comparison. This allowed the research team to determine whether each earthquake originated in the crust or the mantle. The authors solved a long standing problem in global seismology. Determining earthquake depth from standard catalogs can be unreliable because many events are assigned conventional depths such as ten, thirty, or thirty three kilometers. These assigned values often mask the true depth of an earthquake and prevent accurate classification.
To bypass this issue, the researchers used a method based on the ratio of two seismic waves known as Sn and Lg. Both waves travel regionally, but Sn moves through the mantle while Lg moves primarily through the crust. Mantle earthquakes tend to produce stronger Sn energy when compared with Lg. Crustal earthquakes produce the opposite pattern. By comparing each earthquake to several nearby earthquakes recorded at five or more of the same stations, the researchers could determine whether its energy pattern matched a mantle source. This method does not require an exact depth measurement. Instead, it uses the physical differences in how seismic waves propagate through different layers of the Earth.
The results show that mantle earthquakes make up three to four percent of continental seismicity. Although they remain less common than crustal events, they occur in more places than previously recognized. The global map produced in the study outlines clusters of mantle earthquakes across Europe, Asia, Africa, North America, and Australia. These clusters often appear in regions with known crustal seismicity, which provides the necessary comparison events for the Sn to Lg method. Some clusters appear in regions previously believed to contain only crustal earthquakes.
One of the most notable regions is the Alpine Himalayan belt. The researchers identified parallel belts of mantle earthquakes beneath the Apennines in Italy and beneath the Dinarides and Hellenides. Earlier interpretations suggested that these areas might host mantle seismicity related to the movement of the Adria microplate. Previous studies dismissed many of these events because depth determinations were uncertain. The new wave amplitude method confirms them as mantle earthquakes and establishes a continuous pattern along this section of the belt.
The Vrancea region of Romania, already known for its deep seismic activity, contains twenty two mantle earthquakes in the upper portion of its deep seismic zone. The researchers also identified mantle earthquakes beneath the Caucasus and along both sides of the Caspian Sea. Earlier studies proposed that only the central and eastern Caucasus hosted mantle seismicity. The new survey confirms that pattern. It also reveals mantle earthquakes beneath the Zagros Mountains in southern Iran, an area where earlier searches did not identify any deep continental events.
In East Asia, the Tibetan Plateau presents one of the strongest contrasts. The interior of the plateau shows very few mantle earthquakes. However, the edges surrounding the plateau host numerous events, forming an almost continuous boundary of mantle seismicity. A newly recognized cluster stretches 350 kilometers along the Longmenshan region on the eastern side of Tibet. These mantle earthquakes appear to be aftershocks of the magnitude eight Wenchuan earthquake that struck in 2008.
The East African Rift also contains mantle earthquakes. Earlier research identified isolated events here with more limited methods. The new global comparison confirms an extended zone of mantle seismicity across Tanzania, Zambia, and the surrounding areas. In this region the upper mantle attenuates seismic waves strongly, but the comparison with nearby crustal earthquakes allows mantle events to be distinguished.
One of the most unexpected findings lies around the Bering Strait region between Russia and Alaska. The crust is thin here and the lithosphere is warm, which would normally limit brittle failure at depth. Despite this, the study reveals a widespread band of mantle earthquakes that are not associated with known volcanic centers. This distribution spans both sides of the strait and continues into northeastern Alaska and the Richardson Mountains of Yukon. Seismic imaging had previously suggested lower crustal or subcrustal activity in parts of this region. The new method confirms mantle seismicity and expands its extent.
Within the continental United States, mantle earthquakes were previously confirmed only in isolated cases in Wyoming. Those events occurred in 2013. The new survey verifies them and identifies additional mantle earthquakes across the western United States. Southern Idaho contains the highest number of mantle earthquakes in the contiguous United States. These events do not correlate with Holocene volcanic centers. More mantle earthquakes appear in Nevada, northern Mexico, Texas, and Arkansas. None appear in the eastern or midwestern states due to low seismicity and limited opportunities for waveform comparison.
In Central America, two mantle earthquakes appear along the downdip projection of the subducting slab but do not lie directly on the slab itself. South America shows no mantle earthquakes in this survey because the region outside the Andes lacks dense seismic coverage.
In Southeast Asia and Australia, the Arafura Sea had been the location of a previously reported mantle earthquake. That event did not meet the full criteria in this study due to limited station coverage. However, the researchers identified a new cluster in the nearby Aru Trough. Additional mantle earthquakes appear along the Australia Woodlark boundary in New Guinea, beneath central Sulawesi, and beneath northwestern Australia.
The global distribution reveals that mantle earthquakes occur in a broader range of environments than early theories suggested. Some earlier explanations required the presence of cold remnants of oceanic slabs beneath continents. While this mechanism may apply in some regions, it does not explain the extensive mantle seismicity in places such as western North America or the Bering Strait. Temperature estimates from several regions exceed the threshold once thought to prevent brittle failure. This indicates that the upper mantle can host seismic rupture under warmer and more variable conditions than expected.
Some mantle earthquakes clearly follow large crustal earthquakes. The Longmenshan aftershocks in China, the mantle earthquakes following the 2018 Papua New Guinea event, and the mantle event following the 2018 Palu earthquake in Sulawesi all appear to result from stress redistribution after major crustal ruptures. The mechanics behind this response remain unresolved.
Several regions still show no mantle earthquakes in the survey. These areas either lack dense seismic networks or have extremely low crustal seismicity, which prevents the comparison required by the Sn to Lg method. Some cratons fall into this category. Sparse data from northern Africa, parts of Australia, and far eastern Russia also limit classification.
The study demonstrates that continental mantle earthquakes are a persistent global phenomenon instead of isolated geological curiosities. Although they represent a small percentage of all earthquakes, their presence across diverse tectonic and thermal environments provides new insight into continental structure and stress distribution. The research published in Science establishes a new framework for identifying these events and offers the first complete view of their worldwide distribution.
Source:
Based on research published in Science:
https://www.science.org/doi/10.1126/science.adz4367
