About 1,800 miles beneath the surface, Earth’s internal structure changes abruptly where the solid rock of the mantle meets the swirling molten iron of the outer core. But the boundary between the core and the mantle may not be as sharply defined as scientists once thought. By analyzing earthquake waves that bounce off the core-mantle boundary, researchers have found evidence of a thin zone where the outermost core is more solid than fluid.

The existence of such “core-rigidity zones”–small patches of rigid material within the fluid outer core–has been suggested before, but this report marks the first time scientists have detected one. Researchers Sebastian Rost and Justin Revenaugh of the University of California, Santa Cruz, are publishing their findings in the November 30 issue of the journal Science.

The nature of the core-mantle boundary is important because researchers now think it influences phenomena ranging from the behavior of Earth’s magnetic field to the massive plumes of hot rock that rise through the mantle and erupt on the surface at volcanic hot spots such as Hawaii. The interaction of core-rigidity zones with the magnetic field, for example, may help explain the slow wobbling of Earth’s rotation axis, called nutation, said Revenaugh, an associate professor of Earth sciences at UCSC.

“Studies of Earth’s nutation provided one line of evidence that got people thinking there might be these little patches of rigid material in the outer core,” he said. “So previous evidence was consistent with that idea, but now we have evidence that cannot be explained any other way.”

The picture of the core-mantle boundary has grown increasingly complicated in recent years with advances in seismic tomography, which uses seismic waves from earthquakes to probe the internal structure of the Earth. As seismic waves radiate outward from the epicenter of an earthquake, their speed and other properties are affected by the different materials they pass through.

In the 1990s, seismic tomography showed the existence of “ultra-low velocity zones” at the base of the mantle, which some scientists interpret as evidence of partial melting of the mantle. Rost, a postdoctoral researcher, said an ultra-low velocity zone overlaps the area where he detected a core-rigidity zone, but that doesn’t necessarily mean there is a connection between the two. He said the structure of the core-mantle boundary may turn out to be as complex as Earth’s surface layer.

“I think what we have down there is just as complicated as the crust,” Rost said. “I have a dataset that shows a very sharp core-mantle boundary just a little north of where we detected a core-rigidity zone. As we look at smaller scales, I think we will see more and more variation.”

Rost and Revenaugh studied seismic shear waves, which cannot travel through a fluid and reflect off the core-mantle boundary. They looked at waves generated by earthquakes near the islands of Tonga and Fiji in the South Pacific and recorded by an array of instruments in Australia.

According to Rost, the high quality of the seismic data collected by this array was essential for detecting the rigid zone, which is only a few miles across and about 150 meters (about 500 feet) thick. “It’s very thin and about the size of Santa Cruz,” Revenaugh said.

There are two schools of thought about how this rigid material could occur in the molten metal of the outer core. One idea is that the core and the mantle react with one another to produce a material with intermediate density. But this process seems unlikely to produce a layer more than a few meters thick, Rost said.

The other idea relates to the growth of the solid inner core. As the Earth cools and heat flows out of the core, iron from the molten outer core solidifies onto the inner core. This increases the concentration of lighter elements in the outer core, and if those elements are near the saturation point they will also solidify out. But because they are lighter than iron, they will float to the top of the core and collect at the core-mantle boundary.

“You can think of it as an upside-down puddle formed by material rising up to the top of the core,” Revenaugh said.

Whereas puddles of water form at low points on the land, “puddles” of solidified light elements from the core would form at high points in the core-mantle boundary. The seismic evidence suggests the rigid zone consists of a solid matrix with some molten iron in it, Rost said.

“It fits with the idea of an area where solid material has collected within the liquid outer core,” he said.