Almost imperceptible ground movements can announce a giant earthquake.
The principle starts with a simple phenomenon. In some regions, one tectonic plate sinks beneath another: this is a subduction zone. The two plates do not always slide freely. Some sections remain locked by friction, while the overall movement of the plates continues.
Diagram of the subduction process, in which an oceanic plate plunges beneath another tectonic plate.
Credit: K. D. Schroeder, CC BY-SA 4.0, via Wikimedia Commons.
This situation slowly deforms the Earth's crust around the locked zone. Highly precise position measurements, made using satellite navigation systems, make it possible to track these movements. Researchers can thus identify the parts of the fault that resist movement and accumulate energy.
Axel Periollat and Gareth Funning, from the University of California, Riverside, have developed a method for turning these measurements into probability maps. Their algorithm tests a large number of possible distributions of locked zones. It retains those that best reproduce the movements actually measured at the surface.
Kamchatka, in Russia's Far East, provided a real-world test. Data had been recorded before the magnitude 8.8 earthquake of July 30, 2025. The resulting map reveals a large locked region that closely corresponds to the area that subsequently ruptured. A giant earthquake in 1952 also began in this same deep region.
The calculated accumulation is consistent with the energy released in 2025. In this region, the plates converge at about 80 mm per year. Over the roughly 73 years since the great 1952 earthquake, the locked sections could therefore have accumulated several meters of delay relative to the expected movement.
Yet this method cannot predict when the next earthquake will occur. It locates areas where a major rupture appears possible, without providing a date. At most, it can predict that an earthquake's magnitude increases as time passes.
The researchers are now applying their approach to other subduction zones, notably in Japan, Mexico, New Zealand and the northwest Pacific. One difficulty is making measurements offshore, since many major faults lie beneath the ocean. Acoustic instruments installed on the seafloor could provide the missing data and make these maps more precise.