Researchers Investigate Mysteries of the African Rift

December 13, 2001

San Francisco — The formation and evolution of the African Rift Valley are shaded in mystery, but geoscientists at Penn State are mapping the history of the Rift through space and time by analyzing the chemistry of ancient lava from Lake Turkana, northern Kenya.

"We know what the lavas at Turkana look like today and we know what the lavas looked like 35 million years ago in Ethiopia," says Dr. Tanya Furman, associate professor of geosciences. "We also know where the Afar plume is today, but we don't know what it looked like 35 million years ago. The only place we can go back more than 30 million years is in Turkana."

Furman and Kelly Knight, graduate student in geosciences, are looking at the chemical signatures of lavas from the Rift Valley in Turkana to understand the evolution of mantle plumes and how continents split apart.

A mantle plume is a jet of molten lava from the Earth's mantle that reaches the Earth's surface, but is independent of the motion of the tectonic plates. Typically, when mantle plumes rise under oceans, they form islands. The Hawaiian Islands were formed by a plume which remains stationary while the Pacific plate moves above it, creating a string of islands in its wake.

When a plume appears under a continent, a rift may form. Eventually, an important transformation takes place, the rift splits the continent in two, creating an ocean and the rift itself comes a mid ocean ridge. Mid-ocean ridges are the boundaries of two plates and are connected to the Earth's tectonic activity. While geoscientists know the general history of continental rifting, exactly when a rift becomes a ridge and when it becomes part of the tectonic system is unknown.

The African Rift Valley runs in eastern Africa from the junction of the Red Sea and the Gulf of Aden southward to Malawi where it gradually disappears.

"Some of what we found is consistent with the presence of a plume in Turkana about 35 million years ago," Knight told attendees at the fall meeting of the American Geophysical Union today (Dec. 13) here. "We know the plume was in the Ethiopia Sudan area and it appears that lava was channeled 1,000 or more kilometers beneath or through the lithosphere to Turkana."

Turkana is the best place to look at the chemistry of the old lava because three successive rifting events have occurred in the area, stretching and thinning the Earth's crust. While the rest of the Rift has a crust of 100 kilometers with a layer of chemically unusual rocks stuck on the bottom, Turkana has only about 20 kilometers of crust.

The researchers are looking at the chemistry of the lavas including major and minor components, trace elements and isotopes. The recent findings are based on the ratios of trace elements in the lavas, including zirconium to niobium, uranium to thorium, lanthanum to niobium, and barium to niobium. The trace elements in the lavas are related to the source rocks that were melted by the plume. By tracing the lavas back to the rocks, the researchers can trace the paths of the lavas in space. By looking at the old lava in the Turkana area, the researchers can trace the lavas in time.

"We choose trace element ratios that are indicative of the source region of the rocks," says Furman. "These data, plus our preliminary neodymium isotope analyses, suggest that plume lavas have been erupting at Turkana for perhaps 35 million years."

The researchers are also waiting for the results of isotope studies using strontium and lead to clarify the chemical structure of the plume.

The African Rift Valley is a heavily seismic area and there have been volcanic eruptions in the past 50 years. The area is actively changing, but while Furman and Knight know that the rift north of Ethiopia is now an ocean ridge and that the rift south of Turkana is not, they still do not know when, where and how the change from mantle plume to mid ocean ridge occurs.

**aem**

Contacts:
A'ndrea Elyse Messer (814) 865-9481 aem1@psu.edu
Vicki Fong (814) 865-9481 vfong@psu.edu
EDITORS: Dr. Furman may be reached at furman@geosc.psu.edu or by telephone at 814-865-5782. Ms. Knight is at kknight@geosc.psu.edu or by telephone at 814-865-6720.