26 November 2025—A satellite deployed to measure ocean surface heights was up to the challenge when a massive earthquake off the Kamchatka Peninsula triggered a Pacific-wide tsunami in late July.
The Surface Water Ocean Topography or SWOT satellite captured the first high-resolution spaceborne track of a great subduction zone tsunami, researchers report in The Seismic Record.
The track shows an unexpectedly complex pattern of waves dispersing and scattering across the ocean basin, one that could help tsunami scientists better understand how these events propagate and how they could threaten coastal communities.
Angel Ruiz-Angulo at the University of Iceland and colleagues also used data from DART (Deep-ocean Assessment and Reporting of Tsunamis) buoys in the path of the tsunami to build a better picture of the magnitude 8.8 earthquake rupture. The 29 July event in the Kuril-Kamchatka subduction zone was the sixth largest earthquake recorded globally since 1900.
“I think of SWOT data as a new pair of glasses,” said Ruiz-Angulo. “Before, with DARTs we could only see the tsunami at specific points in the vastness of the ocean. There have been other satellites before, but they only see a thin line across a tsunami in the best-case scenario. Now, with SWOT, we can capture a swath up to about 120 kilometers wide, with unprecedented high-resolution data of the sea surface.”
SWOT was launched in December 2022, as a joint mission of NASA and the French space agency Centre National d’Etudes Spatiales, to provide the first global survey of Earth’s surface water.
Ruiz-Angulo said he and study co-author Charly de Marez “had been analyzing SWOT data for over two years understanding different processes in the ocean like small eddies, never imagining that we would be fortunate enough to capture a tsunami.”
Since the wavelength of a big tsunami is longer than the ocean’s depth, researchers often consider these tsunamis to be “non-dispersive.” That is, they mostly remain intact as a singular wave shape as they travel, instead of breaking up or “dispersing” into a leading wave and a train of trailing waves.
“The SWOT data for this event has challenged the idea of big tsunamis being non-dispersive,” Ruiz-Angulo explains.
This animation shows the simulated tsunami wave heights generated by the M8.8 earthquake. Around 70 minutes after the earthquake, the path of the SWOT satellite appears, shown in slow motion to illustrate how the fast-moving satellite captured the tsunami and the dispersive waves that followed the main crest. The tsunami amplitudes measured by SWOT match remarkably well with those produced by the model from the TSR authors. | Courtesy of Angel Ruiz-Angulo
Numerical models of tsunami propagation with dispersion were a better match for the satellite observations of the Kamchatka tsunami, he and his colleagues concluded.
“The main impact that this observation has for tsunami modelers is that we are missing something in the models we used to run,” Ruiz-Angulo added. “This ‘extra’ variability could represent that the main wave could be modulated by the trailing waves as it approaches some coast. We would need to quantify this excess of dispersive energy and evaluate if it has an impact that was not considered before.”
The researchers also realized the tsunami predicted by an earlier model based on seismic and land deformation data did not exactly match the tsunami observations collected by two of the DART tide gauges. The tsunami based on the earlier model was predicted to hit one gauge earlier and one gauge later than observed. The researchers used the DART data in an analysis called inversion to reevaluate the tsunami source.
They concluded that the Kamchatka earthquake source extended further to the south and that the earthquake rupture length was 400 kilometers—significantly longer than the 300 kilometers predicted by the other models.
“Ever since the 2011 magnitude 9.0 Tohoku-oki earthquake in Japan, we realized that the tsunami data had really valuable information for constraining shallow slip,” said study co-author Diego Melgar.
Since then, Melgar’s lab and others have been working on ways to include DART data in inversions, “but it is still not always done because the hydrodynamic models needed to model DARTs are very different than the seismic wave propagation ones for modeling the solid Earth data. But, as shown here again, it is really important we mix as many types of data as possible,” Melgar said.
One of the largest recorded Pacific tsunamis was triggered by a massive magnitude 9.0 earthquake in 1952 in the same Kuril–Kamchatka subduction zone. That tsunami led to the creation of the international alert system that led to Pacific-wide warnings during the 2025 event.
“With some luck, maybe one day results like ours can be used to justify why these satellite observations are needed for real or near-real time forecasting,” Ruiz-Angulo said.