A new study in Nature Communications reports that subducting slabs can become effectively “stalled” for long periods—not simply due to mantle temperature or plate motion, but because distinct rock assemblages within the downgoing plate alter how the slab interacts with the surrounding mantle. Using geophysical constraints combined with thermodynamic modeling, the researchers argue that slab dynamics depend on whether the slab interior is dominated by hydrated ultramafic lithologies or by comparatively dry basaltic components.

The central finding is that wet harzburgite—an olivine-rich, ultramafic rock—can sustain slab stagnation by promoting hydration-related weakening processes and controlling the timing and location of mineral transformation. In parallel, the team shows that dry basaltic slab assemblages can also contribute to stagnation, but through a different pathway: by limiting certain water-driven reactions while still maintaining a mechanically coherent, slow-moving interface at depth.

According to the authors, the fate of volatiles is the key variable. As the slab descends, pressure and temperature drive phase changes and dehydration reactions. These reactions can either increase or decrease the effective viscosity and friction along plate boundaries, shifting the balance between slab pull and mantle resistance. Where hydration is sustained, water remains available to weaken lithospheric portions that would otherwise become too rigid to decouple.

In the wet harzburgite scenario, dehydration and metamorphic reactions are predicted to sustain zones of reduced strength within the slab, preventing it from sinking efficiently into the lower mantle. This creates a feedback loop: slower descent limits thermal loss, which in turn preserves conditions that keep the weakening processes active.

In the dry basaltic scenario, the model emphasizes that even without abundant water, basaltic mineralogy can maintain mechanical coupling over a broader depth range. The study suggests that dry assemblages may delay the development of decoupling horizons, thereby prolonging stagnation until the slab eventually transitions to a different rheological regime.

The researchers also connect their results to the observed diversity of subduction-zone behavior, where some slabs advance smoothly while others appear to linger at mantle depths. By treating rock composition explicitly rather than assuming uniform slab properties, the work offers a framework for interpreting why similar plate boundary geometries can yield markedly different slab lifetimes.

Crucially, the analysis links mineral stability, dehydration partitioning, and viscosity contrasts into a single physical narrative. This moves slab “stalling” away from a purely geometric explanation and toward a lithology-driven process that can be tested against seismic signatures.

The findings will likely influence how future models incorporate water budgets and rock types in subduction simulations. If validated by geophysical observations, the wet-versus-dry assemblage distinction could become a practical lever for predicting which slabs will stagnate and for how long.

In short, the study portrays slab stagnation as a sustained state maintained by the contrasting rheological effects of hydrated harzburgite and comparatively dry basaltic materials, reshaping our understanding of how deep Earth water and rock chemistry govern plate evolution.

Subject of Research: Subduction zone dynamics and slab stagnation mechanisms

Article Title: Slab stagnation sustained by wet harzburgite and dry basaltic slab assemblages.

Article References: Shi, D., Shen, Y., Zhao, R. et al. Nature Communications (2026). https://doi.org/10.1038/s41467-026-75639-9

Image Credits: AI Generated

DOI: 10.1038/s41467-026-75639-9

Keywords: slab stagnation; subduction; wet harzburgite; dry basaltic assemblage; dehydration; mantle rheology; volatiles; mineral transformations

A seismic shockwave traveling within the Earth moved the entire Japanese archipelago by 5 to 6 millimeters, posing a previously unknown Earthquake danger, according to a University of Chicago researcher.

Revealed in a recent paper published in the journal Science, Sunyoung Park and colleagues identified a shear wave bouncing off the Earth’s core after the Tohoku-Oki earthquake in 2011. According to the researchers, this specific type of rebound effect in the mainshock area after an earthquake has never been recognized before, offering new insights into the residual dangers of seismic events and how the layers of our planet interact.

Japan Shifts

As the best-recorded megathrust event in history, the 2011 Tohoku-Oki earthquake has already been well characterized in past studies. However, the team decided to take another look at the event, as monitoring stations throughout Japan detected the ScS shear wave’s return after hitting the core, as well as the archipelago’s eastward movement.

“When we examined the GPS time series in detail, we noticed a small, coherent eastward motion across much of Japan that followed seismic waves reflected from the core (the ScS phase) return to Japan,” Park told The Debrief. “This motion was puzzling because there was no known earthquake (aftershocks) associated with it at that timing, and so we wanted to understand what was really happening.”

According to the team, a previously unknown phenomenon produced the notable ground shift. In just those 13 minutes, a shear wave from the Tohoku-Oki earthquake raced through the planet, bounced off the core, and reactivated the tectonic plate beneath Japan. 

Analyzing the Earthquake Data

In their analysis, they found the rebound shear wave aligned with the tectonic plate boundaries, which, combined with weakened frictional forces after the mainshock, produced the shear wave reflection.

“We only gradually became convinced that this returning wave could actually trigger slip on multiple plate boundaries,” Park said. “It was particularly surprising that such a long-traveling wave (traveling ~5800 km, comparable to the radius of the Earth) could both trigger an event and do so over such a broad region, comparable to the length of Japan, involving several different plate interfaces.”

The movement is described as “steplike,” due to the small but sudden nature of the shift. In the data, the movement resembles a staircase: a flat position, followed by a sudden shift, then another flattening at a different level. Resolving that tiny step proved a challenge.

“In this case, the steplike motion is only a few to several millimeters, so it is relatively small compared to the noise and various processing artifacts that can appear in GPS records,” Park explained. “We did not process the GPS data ourselves; instead, we relied on a carefully processed data product. Even so, we had to be very cautious, because subtle signals can sometimes be introduced by the way the data are processed.”

The research reexamined the main shock source for any evidence of large underwater landslides to rule out other potential causes and found no other correlations.

A New Type of Earthquake Events

According to the team, it is no surprise that such events have previously gone unrecognized. For this rebound effect to occur, an extremely powerful earthquake is required to produce a noticeable effect, and dense, high-precision monitoring networks are needed to detect tiny motions over a large area. Additionally, most seismic studies focus on localized effects rather than viewing the aftermath on such a broad scale.

The slip is six to seven times that of the main shock rupture, making it the broadest single event rupture area ever recorded, doubling that of the great Sumatra Earthquake in 2004. According to the researchers, the Sumatra earthquake may be another example of this behavior, although no work has yet been done to investigate this.

From their work, the team says they cannot predict the exact amount of slip from such an event in the future, similar to the inability to predict normal earthquakes; yet, they can predict when the rebound effect will occur after the main thrust.

“I want to understand more generally how processes deep inside the Earth interact with shallow parts of the Earth,” Park concluded. “I am interested in why deep earthquakes happen and how they may be linked to the whole subduction system and the dynamics, from shallow to deep depth ranges.”

Park added that he is also finding “that these deep events can produce measurable deformation at the Earth’s surface, just like the step-like motions we see in this study,” all of which is helping researchers like him “better understand both these earthquakes and the structure of the Earth.”

The paper, “ScS-Triggered Slip on Megathrust Interfaces after the 2011 Mw 9.0 Tohoku-Oki Earthquake,” appeared in Science on June 18, 2026.

Ryan Whalen covers science and technology for The Debrief. He holds an MA in History and a Master of Library and Information Science with a certificate in Data Science. He can be contacted at ryan@thedebrief.org, and follow him on Twitter @mdntwvlf.