Baptistina's Terrible Daughters

Some 250 million years ago, a biological catastrophe struck Earth. An estimated 75 percent of all land-living species and up to 95 percent of all ocean-dwelling species disappeared forever in a geologically brief pulse of mass extinction. And it probably occurred in less than 160,000 years--perhaps as little as 10,000 years. That is equivalent to less than a second if Earths entire history were compressed into a single day.

The mass extinction at the end of the Permian geological period was the worst such event Earth has ever endured. It was far worse than the better-known extinction 65 million years ago that ended the dinosaur era. Imagine that 99 out of every 100 people you know suddenly died. Now imagine that entire families of animals, not just individual species, suddenly disappeared: no more rabbits, no more bats, no more horses. The Great Dying, as some call the end-Permian extinction, was the biggest population crash in evolutionary history.

Scientists have long struggled to understand what virtually transformed Earth into a cemetery. The scenarios proposed so far include a global decline in sea-water oxygen levels, greenhouse warming, and massive volcanic eruptions. Evidence has recently emerged for an entirely different kind of disaster that may have contributed to the Permian mass die off: a strike by a 30-mile-wide (50 kilometers) asteroid.

In 2006, Ohio State University geoscientist Ralph von Frese reported possible evidence of a geologic structure resembling an impact crater. The feature is buried in rock miles beneath the East Antarctic ice sheet. It lies below a region called Wilkes Land and may be more than 300 miles wide (500 km). If the feature is found" to be an impact crater--and the evidence is very far from conclusive--it would be large enough to swallow Ohio.

Von Frese has no direct, physical evidence to prove the Antarctic feature is a crater. Nor does he have geological proof that it even dates to the end-Permian time. The evidence now consists only of a warp in the local gravity and surface topography.

However, another team has reported much more substantial backing for a possible impact associated with the end-Permian extinction. It's a structure beneath the ocean floor 16 miles (25 km) off Australia's northwest coast called the Bedout ("Bedoo") High. The scientists, led by Johns Hopkins University geochemist Luann Becker, have marshaled multiple forms of evidence.

Becker and her colleagues say the data available are most consistent with a buried impact structure dating precisely to the end-Permian extinction. The impactor itself might have been a meteorite roughly the size of the one that helped finish off the dinosaurs 65 million years ago. Neither Becker's nor von Frese's scenario is yet universally accepted, but they have kept the extraterrestrial-impact hypothesis simmering.

Von Frese found what he believes to be clues to Earth's "mother of all extinctions" by observing the Moon and planets. The information he commonly works with consists of gravity and magnetism measurements collected by satellites and space probes. He looks for concentration of mass, or "mascons" for short, in the lunar crust. Many mascons exist on the Moon.

Mascons can form when a meteorite smacks into the crust, causing molten rock to well up into the scar. The upwelled material is denser than the surrounding crust, creating a small increase in the local gravity. Spacecraft can detect the gravity boost as they pass over a mascon. Scientists, in turn, work backward from the size of the orbital blips to the variations in local gravity that caused them. Von Frese used this technique to map mascons on the Moon. His impact detector: the lunar-orbiting Clementine probe.

For the last several years, von Frese and his team have studied the sub-ice geology of Antarctica using data from NASA's twin Gravity Recovery and Climate Experiment (GRACE) satellites. The mission launched in 2002. Super-accurate gravity maps contain information about ocean currents, groundwater motions, glacial ice dynamics, and other processes of interest to environmental scientists. The team has used GRACE data to create something akin to a crude X ray of what lies beneath the Antarctic ice.

A newspaper report von Frese saw in 2004 provided the original impetus to seek mascons in Antarctica. The article discussed the possibility that a meteorite strike caused the end-Permian extinction. Similarly, a geologic structure known as Chicxulub crater, buried beneath the northern edge of Mexico's Yucatan Peninsula, marks the location of the impact linked to the dinosaur's decline about 65 million years ago.

The much more severe end-Permian extinction happened 185 million years earlier. Since then, shifts in Earth's dynamic crust have erased most Permian crustal geology--but not all.

"When we looked for it, it just jumped out at us," von Frese recalls. "It's similar to what is seen for mascons on the Moon and Mars. Sometimes it's called a sombrero anomaly."

The Wilkes Land mascon lies nearly centered below a shallow circular pattern previously mapped with ice-penetrating radar. It reminded von Frese of a lunar mascon. Similar structures exist on Mars, Venus, and Mercury. Most of them are geologically "dead." Their interiors cooled long ago into rock, "freezing" the mascons in place. Even the most ancient mascons remain visible on geologically inactive moons and planets.

Earth is different. The mantle below our planet's relatively thin crust remains hot and plastic. Over millions of years, the mantle flows. These motions gradually obscure mascons. Earth's active surface quickly erases surface craters.

"We've seen nothing else yet in the satellite data," von Frese says. "We've looked at other big impact craters, like Vredefort in South Africa, but we don't see anything."

The Vredefort Structure is the highly eroded remnant of a crater about 190 miles (300 km) across and more than 2 billion years old. On the Moon, a crater as large as Vredefort would still have a mascon. Vredefort has none.

The Antarctic mascons very existence suggests to von Frese that the Wilkes Land gravity anomaly is much younger than the Vredefort structure. He roughly estimates it to be less than 1 billion years old, although this is just an educated guess. In another half a billion years, he says, the Wilkes Land anomaly--whatever it is--will disappear.

Von Frese's mascon hypothesis remains on shaky scientific ground. What he needs to prove his case are actual