A rockfall in Yosemite Valley in October 2010.

Tom Evans

What’s causing cliffs to crumble all over the world?

If there’s anything nearly guaranteed to bring out the rock climbers in California’s Yosemite National Park, it’s a warm sunny day. And even though dry surfaces are more easily gripped than slick ones, there are still dangers at hand: Multiton slabs of rock can pop off the face of a cliff with little or no warning, careening downslope dozens of meters to shatter in a cloud of geological shrapnel. Dozens of such rockfalls occur every year, from Brazil to Japan, yet what causes many of them has remained a mystery. Now, researchers think they’ve cracked it.

California’s Yosemite National Park is a hot spot of rockfalls. Hundreds have occurred in the park, yet 15% didn’t have any obvious trigger: no earthquakes, no heavy precipitation, no freeze-thaw cycles that can cause water trapped in a fissure to freeze, expand, and gradually wedge open a crack. In fact, some occurred in the middle of a sunny summer day.

To try to discover what might cause such events, Brian Collins, a geological engineer with the U.S. Geological Survey in Menlo Park, California, and Yosemite National Park geologist Greg Stock installed strain gauges, which measure overall changes in length, at three spots on a 19-meter-long, 4-meter-wide slab of rock that’s tenuously attached only along its top and bottom edges to a south-facing cliff in the park. The researchers measured the deformation of the near-vertical slab—a 10-centimeter-thick, 20-metric-ton layer of granite—every 5 minutes from May 2010 through October 2013. They also used devices that look like scissor jacks (used by unfortunate drivers to lift their car and change a flat tire) to monitor the movement of the slab toward and away from the cliff from which it was splitting. Finally, they monitored weather conditions at the site, including the intensity of sunlight, the humidity and temperature of the air in the gap behind the slab (a space as much as 12 centimeters wide), and the humidity and air temperature 2 centimeters from the outer surface of the slab. On three separate occasions during the 42-month period, the scientists used a laser-emitting device to scan the slab from a spot 30 meters away, thus gaining an independent measurement of its motions during an 18-hour period.

Over the course of an average day, the slab bulged outward and then shrank back about 8 millimeters (or approximately half the width of a thumbnail), the researchers report online today in Nature Geoscience. Although changes in humidity seemed to have little, if any, effect on the slab’s deformation, changes in temperature had a major influence. The maximum bulge typically occurred between 1 p.m. and 4 p.m., when temperatures were the highest for the day, and minimum bulge occurred between 7 a.m. and 9 a.m., when temperatures were generally at their lowest point of the day. The maximum range of deformation (the biggest difference between morning and afternoon positions) occurred when temperature variations were their greatest—usually in spring and fall, not during the summer.

The bulging of the granite slab stems from its daily cycle of heating and cooling, the researchers surmise. When the material heats up, it expands, Collins says. But because its ends are still attached to the rock and thus pinned, the slab’s only choice is to buckle and bulge. (For the same reason, railroad rails and bridge components in the summer sun often buckle and pose great danger if they can’t expand freely.) That bulging, in turn, tends to open the cracks at the top and bottom of the fissure behind the slab, generating stresses that pull the rock apart and thus drive crack growth. Each day’s growth is probably minuscule, Collins says, but he and his team haven’t yet calculated how much one day’s crack growth might be. Nevertheless, over time the fissure grows to the point at which the bulging on a hot summer day can fracture a slab free of its cliff, triggering a rockfall.

The study “is a very important piece of work that brings a new kind of life to rocky landscapes,” says Jeffrey Moore, a geologist at the University of Utah in Salt Lake City. “A 1-centimeter deflection over the course of a day is not trivial.” Daily swings in temperature can aggravate the same sorts of stresses in layered sandstones, he says, driving cracking—and thus possibly triggering rockfalls—within those types of rocks as well.

Until now, researchers have largely ignored thermal stresses in rocks, says Stephen Martel, a geologist at the University of Hawaii, Manoa. But the new research—“a very nice job of both field measurement and theory,” he notes—shows that thermal stresses can be important. When added to stresses already present in the rocks, those caused by changes in temperature “can be the straw that breaks the camel’s back,” Martel says, and triggers a disaster.

The statistics are rather clear, Collins and Stock report. About 15% of Yosemite’s rockfalls that either don’t have a clear cause or can be linked to thermal stresses occur during the hottest months of the year (July through September) or at the warmest time of day (from noon through 6 p.m. local time). If the distribution of events were totally random, the number occurring within those time windows would be only 6%, the researchers say.

Although the new study doesn’t help researchers predict rockfalls in any meaningful way, it does help geologists understand how such events can be triggered—and possibly makes rock climbers a little more nervous the next time they’re hanging off a cliff on a warm sunny day.