Researchers from the Seismological Society of America utilized a remote seismic station on Upolu, Samoa—part of the Global Seismographic Network—to measure and identify the source of the American Samoa earthquake swarm. Without a local seismic monitoring network, they relied on a single seismic sensor 250 km away. Their findings were published in The Seismic Record.
Map of Manuʻa Islands with Raspberry Shake locations (red triangles) and villages (black squares) contributing felt reports. The swarm origin is localized to where the yellow and red rings 664 intersect: directly under, or offshore either north or south of, Taʻū Island. Image Credit: Yoon et al. (2025) TSR
Residents of the Manu'a Islands in American Samoa experienced multiple daily earthquakes from late July to October 2022, which sparked fears of an impending tsunami or volcanic eruption.
Since there was no seismic monitoring network on the islands to measure the shaking and help seismologists find the source of the earthquake swarm, an earthquake catalog for the region produced no results.
However, Clara Yoon of the US Geological Survey and colleagues discovered another method to fill in the seismic gaps because the people of the Taʻū, Ofu, and Olosega islands needed answers. The scientists analyzed shaking data from a single seismic sensor 250 km from the American Samoa swarm using machine learning and a different method known as template matching.
Yoon and colleagues described how, before the installation of local permanent seismic stations in American Samoa in August and September 2022, the team used these single-station data to track the swarm along with resident reports of shaking.
The non-eruptive volcanic earthquake swarm began in July 2022, approximately 15 km offshore from Taʻū Island. The Samoa volcanic islands form as the Pacific tectonic plate drifts over a hotspot in the South Pacific Ocean.
At first, the only information available about the swarm came from resident reports of the frequent shaking, which happened several times a day for a few seconds at a time.
When the earthquakes started, American Samoa had no instrumental geophysical monitoring, so even basic information about the source of the shaking–with implications for emergency decision-making and public safety–was nonexistent.
Clara Yoon, US Geological Survey
The researchers addressed this using a remote seismic station on Upolu, Samoa, which is part of the Global Seismographic Network. According to Yoon, the EarthScope data center offers data that can be downloaded in almost real time.
However, Yoon and colleagues employed a deep-learning model called EQTransformer in conjunction with a technique called template matching to select these small earthquakes from a noisy seismic background because the seismic signal of the American Samoa earthquake swarm was hard to detect at the distant station.
EQTransformer found many earthquakes with locations consistent with eastern American Samoa, the largest of which matched up with times of felt reports. These felt reports, contributed by local residents of American Samoa to the National Weather Service, were essential sources of data about the earthquakes, and gave us confidence that the EQTransformer-detected events were actually the same earthquakes felt by the local population.
Clara Yoon, US Geological Survey
Using this new earthquake catalog, the researchers could describe the beginning and peak of the swarm activity. The swarm's location was rapidly determined in August 2022 by deploying low-cost, portable Raspberry Shake sensors.
The researchers concluded that although the swarm ended in October 2022 without an eruption, it was probably caused by the movement of volcanic magma.
Yoon pointed out that a method similar to their single-station technique might be helpful in other parts of the world, like offshore areas with the potential for tsunamis or earthquakes within a tectonic plate, where there is little permanent seismic monitoring and little knowledge of seismic hazard.
She continued that the largest earthquake in the American Samoa swarm was magnitude 4.5, so it is unlikely that global seismic networks would have picked it up.
“If no one had lived nearby to report the frequent shaking, this American Samoa swarm may have gone entirely unnoticed. Many unknown seismic sources and phenomena are waiting to be discovered, perhaps by future large-scale comprehensive applications of deep-learning approaches in seismology,” said Yoon.
Journal Reference:
Yoon, C. E., et al. (2025). Remote Single-Station Seismic Monitoring of the July–October 2022 Earthquake Swarm at Ta‘ū Volcano, American Samoa. The Seismic Record. doi.org/10.1785/0320240040.