Plumes of Hot Material near Earth’s Core Grease Way for Moving Slabs of Earth

Article Content

The way long, thin streams of hot semi-molten rocks move far below the Earth’s surface may partially solve a major, decades-long mystery of geology.

A scientific revolution in the mid-1960s installed plate tectonics as the central organizing concept of geology. Plate tectonics describes Earth’s surface as a mosaic of 15 or more giant slabs of rock in slow, perpetual motion.

“Plate tectonics is unique to Earth as far as we know and was crucial to the evolution of life on our planet, but we still don’t know how it works,” said Daniel Blatter, a geophysicist at UC San Diego’s Scripps Institution of Oceanography. For plate tectonics to work, a low viscosity layer must exist at the base of the plates, “much like a thin layer of butter on a tabletop. Yet the cause of the low viscosity is still unclear, decades after the discovery of plate tectonics.”

Blatter and three co-authors report in the April 20 issue of Nature that streams of heated rocks called mantle plumes probably play a role in creating a slippery base for tectonic plates. The study suggests that the plumes could be putting into place long-lived, water-rich melt in a network of thin channels at the base of Earth’s rigid outer shell. This melt would help plates slide by reducing viscosity at the base of the tectonic plates. 

Geophysicists conceived of mantle plumes as spouts of heated material rising from deep below Earth’s crust. Forming on a regional scale, plumes create volcanoes if they break the surface, giving birth to island chains such as Hawaii. 

The Nature study is based on a surprise discovery in geophysical imaging data collected along the Cocos Plate off the coast of Nicaragua in Central America. The study’s conclusions would require a nearby but undiscovered plume to bring in a low viscosity layer. Or perhaps, the Galápagos plume 1,000 kilometers (621 miles) to the south is the source.

The Cocos Plate data were collected about 12 years ago for a different purpose, but that data yielded a surprise discovery that led to the current research.

Co-author Samer Naif, an assistant professor at of Georgia Institute of Technology, originally used the data for his PhD research at Scripps Oceanography. The magnetotelluric (MT) imaging method used on the Cocos Plate measures electrical conductivity beneath Earth’s surface. A mantle with partially melted rocks shows more conductivity than would the same portion of solid mantle.

The surprise discovery was a section at the base of the Cocos Plate displaying unusually high conductivity. This was a likely sign of melted rocks, as Naif and his co-authors reported in a 2013 issue of Nature.

But recently, along with co-author and fellow Scripps Oceanography alumnus Anandaroop Ray of Geoscience Australia, Blatter developed computer code that allowed them to analyze the MT data in a new way.

The new code explores a large family of models that could satisfy the data in accord with various assumptions about subsurface conditions. In doing so they would be able to identify the most likely results even with incomplete knowledge of the system.

Blatter and Naif also wrote codes that would simulate the physics of subsurface melt processes for the study. But even with several days of modeling on Columbia University’s Habanero supercomputer, the results produced a trade-off. 

The scientists could explain some of their data as indicating a mantle segment consisting of more melt but fewer dissolved volatiles—in this case, water and carbon dioxide. Or they could explain the same portion of the mantle as containing less melt but more dissolved volatiles. 

“You can’t just explain it with a normal mantle composition,” Naif said. “You need something else that is anomalous and that’s where the volatiles come in.”

A lack of volatiles would indicate the presence of so much melt that it would be unable to remain in the mantle for long. The melt would rise and erupt at the surface. But the simulations indicate that a water-rich melt would remain in the mantle and greatly reduce its viscosity.

“It becomes much more like honey and much less like rock,” said Blatter, the John W. Miles Postdoctoral Fellow at Scripps Oceanography.

These melt channels could be a byproduct of mantle plumes globally. If an oceanic plate creeps past a plume, it could inherit melt channels and the unusually high volatile concentration needed to sustain its existence.

“We’re not claiming that this is the entire answer, but plumes could be part of it,” Blatter said.

 In addition to Blatter, Naif, and Ray, the research team included Scripps Oceanography alumnus Kerry Key of Columbia University. The team received support from the National Science Foundation and Columbia University’s Electromagnetic Methods Research Consortium.