Now a team of researchers has discovered that this climatic blip at the boundary between the Oligocene and Miocene epochs corresponded with a rare combination of events in the pattern of Earth's orbit around the Sun.

In a paper published in today's issue of Science, the researchers show that the transient glaciation and other climatic variations during a period from about 20 to 25.5 million years ago correspond with variations in Earth's orbit known as Milankovitch cycles.

Although the concept of such relationships is not new, some of the results were surprising, said James Zachos, a professor of Earth sciences at the University of California, Santa Cruz, and lead author of the paper.

"When we began examining the temporal relationship of the orbital oscillations relative to the oscillations in the climate record, we never suspected that the transient glaciation at 23 million years ago had anything to do with orbital anomalies," Zachos said.

The astrophysicist Milutin Milankovitch first proposed that cyclical variations in certain elements of Earth-Sun geometry can cause major changes in Earth's climate. The main variables are eccentricity, obliquity, and precession.

Eccentricity refers to the changing shape of Earth's orbit around the Sun, which varies from nearly circular to elliptical over a cycle of about 100,000 years.

Obliquity refers to the angle at which Earth's axis is tilted with respect to the plane of its orbit, varying between 22.1 degrees and 24.5 degrees over a 41,000-year cycle.

And precession is the gradual change in the direction Earth's axis is pointing, which completes a cycle every 21,000 years.

"Because there are several components of orbital variability, each with lower frequency components of amplitude modulation, there is the potential for unusual interactions between them on long timescales of tens of millions of years," Zachos said. "What we found at 23 million years ago is a rare congruence of a low point in Earth's eccentricity and a period of minimal variation in obliquity."

The result of this rare congruence was a period of about 200,000 years when there was unusually low variability in the planet's climate, with reduced extremes of seasonal warmth and coldness. Earth's orbit was nearly circular, so its distance from the Sun stayed about the same throughout the year.

In addition, the tilt of Earth's axis, which gives rise to the seasons, varied less than usual. In other words, the tilt doesn't always vary between the same extremes in its 41,000-year cycles; the obliquity cycle itself varies in amplitude over a longer period of about 1.25 million years. Similarly, the eccentricity cycle peaks every 400,000 years.

The combination of a low-amplitude "node" in the obliquity cycle and a minimum in eccentricity would have caused only several degrees difference in summer temperatures at the poles, but it was probably enough to allow the Antarctic ice sheet to expand, Zachos said.

Zachos's collaborators on the paper were Nicholas Shackleton and Heiko Pälike of Cambridge University, Justin Revenaugh of UC Santa Cruz, and Benjamin Flower of the University of South Florida.

The researchers obtained detailed climate records for the late Oligocene and early Miocene by analyzing sediment cores drilled out of the ocean floor. Cutting through layers of sediments laid down over millions of years, such cores contain a chronological record of past climates written in the chemistry of fossilized shells left behind by tiny marine organisms. Oxygen isotopes in the shells, for example, reflect ocean water temperatures and the amount of ice trapped in glaciers.

In the 1970s, scientists using these techniques obtained the first good evidence in support of Milankovitch's theory, almost 50 years after he had proposed it.

According to Zachos, researchers are still trying to get a handle on the relationships between climate cycles and orbital variations. Since most of the research has focused on the past 5 million years, the new paper is valuable because it looks at a more distant window in time when conditions on the planet were different.

In the period they examined, the late Oligocene and early Miocene, Zachos and his collaborators found evidence of several climate cycles with frequencies corresponding to the Milankovitch cycles.

But the correspondence of the orbital anomaly with the transient glaciation event at the boundary between the two epochs is especially interesting, Zachos said. The climate system seems to have undergone a fundamental shift at this boundary, which also marks a major break in the paleontologic record.

"I'm not sure everyone will be convinced that the orbital anomaly alone is responsible," Zachos said. "But the congruence of those orbital cycles is a very rare event, and the fact that it exactly corresponds with this rare climatic event is compelling." - By Tim Stephens

[Contact: James Zachos, Tim Stephens]

13-Apr-2001