Over 40,000 seamounts—undersea mountains that don't breach the ocean's surface—are scattered across the ocean floor. Some form linear chains, while others occur as dispersed, isolated features that are not part of well-defined volcanic chains.

Despite the huge number of seamounts, however, scientists have been uncertain about their origin.

The Hawaiian Islands are part of a long seamount chain that scientists believe was produced by hot mantle plumes rising from the core–mantle boundary. This is the essence of the classical hotspot hypothesis. According to this interpretation, mantle plumes cause partial melting beneath moving tectonic plates, producing volcanic chains such as the Hawaiian–Emperor chain. However, this hypothesis can't explain the origin of seamounts in general since only about 50 seamount chains conform to the classical hotspot model.

To refine this hypothesis, a research team led by Professor LIU Lijun at the Institute of Geology and Geophysics (IGG) of the Chinese Academy of Sciences (CAS) used unprecedented computational power and model resolution to reconstruct the dynamic evolution of Earth's interior over the past 270 million years. The team identified a unified deep-Earth dynamic mechanism responsible for the formation of widely distributed seamounts, thereby extending and refining the classical mantle plume model.

The study was published in Nature Geoscience on June 10.

Comparison between the modeled present-day 3D mantle plume structure and geological observations. The upper surface shows continental topography, while white regions show continental shelves, seamounts, island arcs, and other geological features. Orange isosurfaces represent mantle plumes, while gray isosurfaces indicate subducting slabs deeper than 300 km. (Image by IGG)

Using computer simulations, the researchers analyzed the spatiotemporal behavior of deep-rooted mantle plumes and their role in generating volcanism within ocean basins. They noted that the formation and evolution of both linear, well-organized seamount chains and isolated scattered seamounts are closely associated with higher temperatures in portions of the asthenosphere—a region of the upper mantle. These areas of higher temperature, called asthenospheric thermal anomalies, are generated by mantle plumes originating from the core-mantle boundary.

Taking the Pacific plate as an example, during the early stages of mantle plume ascent, large amounts of plume-derived heat accumulated beneath the young Pacific lithosphere, forming extensive asthenospheric thermal anomalies. These anomalies exhibit a clear spatiotemporal correspondence with scattered seamounts in the Western Pacific Seamount Province.

The simulations showed that during subsequent evolution, mantle plumes could split either from their lower-mantle roots or within the mantle transition zone, generating secondary mantle plumes. This process further increased the number of modeled hotspots and created favorable conditions for the formation of additional seamount chains.

The hot material associated with these plumes can persist within the asthenosphere over geological timescales, gradually migrating and dispersing under the influence of mantle convection. The modeled temperatures of these residual thermal anomalies show a significant linear correlation with the elevations of seamounts observed at corresponding locations.

These results suggest that asthenospheric thermal anomalies can generate numerous small and scattered seamounts, supporting the idea that such thermally anomalous regions serve as "seamount brewing zones."

Significant spatiotemporal correlations between mantle plume-induced asthenospheric thermal anomalies and observed seamount occurrences. (a–f) Background colors denote asthenospheric temperature anomalies at different evolutionary stages. Green dots mark the paleo-emplacement locations of observed seamounts. (g) Locations and elevations of all identified seamounts. (h) Relationship between observed seamount elevations and modeled present-day thermal anomalies directly beneath each seamount. The red line represents the mean thermal anomaly, and the black line denotes the linear regression fit. (i) Probability distribution of modeled thermal anomalies for different seamount-elevation classes. (Image by IGG)