In a discovery that has repercussions for everything from domestic agricultural policy to global food security and plans to mitigate climate change, researchers at the University of Massachusetts have found that the rate of soil erosion in the midwestern U.S. is 10 to 1,000 times greater than pre-agricultural erosion rates.

The newly discovered pre-agricultural rates, which reflect the rate at which soils form, are orders of magnitude lower than the upper allowable limit of erosion set by the U.S. Department of Agriculture.

The U.S. National Science Foundation-supported study, which appears in the journal Geology, makes use of a rare element, beryllium-10, or 10Be, that occurs when stars in the Milky Way explode and send high-energy particles, called cosmic rays, rocketing toward Earth. When this galactic shrapnel slams into the Earth's crust, it splits oxygen in the soil apart, leaving tiny trace amounts of 10Be, which can be used to precisely determine average erosion rates over the span of thousands to millions of years.

"We went to 14 small patches of remnant native prairie that still exist in Iowa, Minnesota, South Dakota, Nebraska and Kansas, and used a hand auger to collect deep soil cores, in material that dates back to the last ice age," says Isaac Larsen, a geoscientist at the University of Massachusetts Amherst and the paper's senior author. "We brought this soil back to our lab, sifted it to isolate individual sand grains, removed everything that wasn't quartz, and then ran these few spoonfuls through a chemical purification process to separate out the 10Be — which was just enough to fit on the head of a pin."

Not only is topsoil crucial for U.S. agriculture — the annual cost of diminished agricultural productivity and environmental degradation due to erosion is estimated at tens of billion dollars each year — and worldwide food security, but climate mitigation plans that rely on storing carbon in the soil. "The key is to reduce our current erosion rates to natural levels," says Larsen.

"This basic research addresses important scientific questions about landscape evolution," says Justin Lawrence, a program director in NSF's Division of Earth Sciences. "The knowledge gained could lead to more sustainable agricultural practices."