Tracking pebbles. From painted rocks (inset), a researcher deduced that Owyhee harvester ant colonies help construct desert pavement.

Underground Crew Paves the Desert

SAN FRANCISCO--Compared to floods and earthquakes, insects might not seem capable of reshaping the Earth. But quite a lot is possible given enough legs: As demonstrated by a new study, colonies of ants can pave desert dunes by hauling pebbles to the surface.

Desert pavement is a mantle of loose rocks lying atop a thick layer of fine soils. It is thought to form as windblown dust collects beneath exposed rocks. Combined with forces such as freezing and thawing, this dust ratchets pebbles and cobbles upward from beneath ground to cover a growing base of sediment. But geologist Katherine Leonard of Portland State University in Oregon wondered whether a much smaller scale operation, ant colonies, was contributing too. So she traveled to the Summer Lake dunes in southeastern Oregon, thick with colonies of Owyhee harvester ants (Pogonomyrmex owyheei). Their conical anthills are armored with pebbles of the same size and shape as those in surrounding desert pavement.

As the first step in determining whether ants contribute to desert pavement, Leonard demolished an anthill. Then she scattered the pebbles from the anthill around the former entrance and spray painted several patches of pebbles various colors to track where they were moved. To Leonard's surprise, painted pebbles accounted for less than 25% of the new hill that was built in 16 months. Instead, the ants dug fresh pebbles from the soil, probably while excavating their living quarters.

Leonard estimates that a single ant colony could accumulate enough pebbles in 2 years to cover a square meter with a 1-centimeter-thick layer of desert pavement. Wind and rain would take thousands of years to form an equivalent amount of pavement. So in the 8000 years the sand dunes have existed, Leonard reported 14 December at a meeting of the American Geophysical Union, ants could have supplied most of the pebbles that pave the desert dunes at Summer Lake. Small animals, wind, and other forces scatter the pebbles to create an even layer.

"[Leonard's theory] is a wonderful idea that is in no way related to the original theory" of how desert pavements form, says Robert Anderson, a geologist at the University of California, Santa Cruz. Animals could explain the surfaces found in several western U.S. deserts, he explains. "From a plane, a good chunk of the Great Basin looks like it's got chicken pox," Anderson says. "And the pocks are anthills."

Related site

Background on desert pavement