Skip to main content

Can Rivers Cause Earthquakes?

If so, it could help explain some quakes that happen far from tectonic-plate boundaries

This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American


Steel beams in the ceiling rattled as the gym shook at the College of William and Mary in Virginia. With basketball hoops wiggling perceptibly overhead, I evacuated the building along with everyone else. A magnitude 5.8 earthquake rocked the East Coast from Florida to Canada on August 23, 2011. It wasn’t huge by global standards but still caused hundreds of millions of dollars in damage.

Most large earthquakes occur at the boundaries of tectonic plates—the slowly moving, continent-sized slabs of rock that make up Earth’s crust—in places like Nepal and California. But many destructive earthquakes have happened in the eastern United States, far from any plate boundary. What triggers these so-called intraplate earthquakes is something of a mystery. New research by geologists at Colorado State University and the University of Kentucky suggests a surprising culprit: rivers. By eroding trillions of tons of rock from Earth’s surface over millions of years, rivers might remove enough weight to change the balance of forces in Earth’s crust and cause intraplate earthquakes.

Intraplate quakes happen frequently and can have serious consequences. Virginia experiences about one earthquake a year that can be felt at Earth’s surface. In 1886, a magnitude 7.0 quake in Charleston, S.C., damaged almost every building in town, leading to a severe housing shortage and prolonged social unrest. Chimneys toppled in St. Louis in 1812 when a magnitude 7.5 leveled the nearby town of New Madrid. A similar event in New Madrid today would be one of the costliest natural disasters in U.S. history. And at the College of William and Mary, where I was a sophomore geology major, the same earthquake that forced me out of the gym interrupted a well-timed geology faculty meeting. None of these places are within a thousand miles of a plate boundary. Better prediction of intraplate earthquakes could save millions of dollars in property in places traditionally thought of as safe from major quakes.


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


Intraplate earthquakes require three ingredients. First, there must be faults, or weaknesses in the Earth’s crust, far from a plate boundary. Second, stresses must build up at those faults. Finally, an increase in stress above what the crust can sustain triggers an earthquake.

Geologists have a good sense for why the first two ingredients exist on the East Coast; the eastern U.S. has a sordid geologic past. The Appalachian Mountains formed in a collision between North America and Africa. Shortly after that, the two continents wrenched apart as the Atlantic Ocean opened up between them. All of that smashing and tearing of rock has left the crust under the East riddled with faults. Stress on these faults then grows over time as plates move and interact with each other.

But what about the third ingredient? What triggers intraplate earthquakes? If they occurred simply from continuous stress buildup along faults, they would occur all up and down the Eastern Seaboard. Yet East Coast earthquakes are concentrated into several “seismic zones,” areas with frequent quakes. The earthquake I felt in Virginia, for example, hit smack in the middle of the Central Virginia Seismic Zone. Something must be triggering earthquakes specifically within observed seismic zones and not elsewhere. The trigger, at least in one seismic zone in eastern Tennessee, might be a river.

According to Sean Gallen, a Colorado State University geologist who has long been interested in Appalachian landscapes, the Tennessee River removed about 500 vertical feet of rock (a good third of the height of the Empire State Building) from a large patch of eastern Tennessee over the past 9 million years. That’s about 8 trillion tons of rock swept off the landscape in the geologic blink of an eye.

A new study by Gallen and geologist Ryan Thigpen at the University of Kentucky suggests that rapidly removing so much weight from the crust causes earthquakes. Like squeezing a tube of toothpaste while unscrewing the cap, removing the confining pressure of overlying rock may allow parts of the crust to move quickly, resulting in large earthquakes far from plate boundaries. The researchers use a computer model to calculate how stress in the crust would respond to the removal of 500 vertical feet of rock. Their results predict a pattern of stress change that corresponds with the earthquakes observed in the area between 1901 and 2016.

In short, Gallen and Thigpen found that earthquakes in East Tennessee happen most frequently where large volumes of rock have been excavated from Earth’s surface by river erosion, and they argue that the unloading of rock from the surface causes increased earthquake frequency.

While this study’s results are specific to the East Tennessee Seismic Zone, geologist Eric Calais of the École Normale Supérieure in Paris and his colleagues have proposed a similar mechanism for the origin of another intraplate seismic zone in Missouri. Erosion by the Mississippi River some 10,000 years ago may have set the stage for the earthquake that flattened New Madrid and shook St. Louis back in 1812.

Geologists who study rivers, myself included, know that periods of rapid erosion like these are fairly common. This means that river erosion could be causing intraplate earthquakes in other regions that haven’t been investigated yet. The body of evidence for rivers as a control on the spatial distribution of intraplate earthquakes is strong and continues to grow.

Rivers cannot cause the initial faulting that gives rise to weaknesses in the crust, nor can they provide the initial stresses that those faults experience. But rivers may set the location of some intraplate earthquakes by removing huge quantities of rock from some locations but not others over a relatively short time.

The Central Virginia Seismic Zone, which gave rise to the 2011 earthquake that shook the William and Mary gym, has so far not been associated with a pulse of rapid rock removal by rivers. While river erosion is an important part of the story, geologists are still working to understand all the causes of intraplate earthquakes to minimize future damage to life and property. In the meantime, keep in mind that whatever your distance from a plate boundary, the nearest earthquakes are probably much closer.

Charlie Shobe is a PhD candidate at the University of Colorado, Boulder. He uses computer models and field work to understand how Earth's surface changes shape over time. In his spare time, he races bikes in the foothills of the Rocky Mountains.

More by Charlie Shobe