Geotimes Banner

Geotimes is now


Customer Service
Geotimes Search

GeoMarketplace Link

EARTH magazine cover

  Geotimes - March 2007 - New Madrid fault dying?
New Madrid fault dying?

A series of devastating earthquakes that altered the course of the Mississippi River in the early 19th century may have been among the last gasps of an old, dying fault system, a controversial new study suggests. The fault system, the source of the earthquakes in the New Madrid seismic zone, is no hotter or weaker than its surroundings, the researchers found — suggesting that it won’t always be a hotbed for seismic activity. Instead, activity could soon migrate elsewhere in the region — a fact that hazard planners should take into account, the team says.

During the winter of 1811 to 1812, three strong earthquakes between magnitude 7 and 8 rocked the New Madrid seismic zone, which runs through parts of eastern Missouri and northeastern Arkansas. The earthquakes changed the Mississippi River’s course, opened deep fissures, destroyed forests and lakes, and produced intense ground-shaking that liquefied the soil, turning the land to the consistency of jelly across an area of 10,000 square kilometers.

Centering around the 240-kilometer-long Reelfoot Fault system, which runs from southern Illinois to northeastern Arkansas, the New Madrid zone is in the center of a tectonic plate, rather than along its boundaries. Such intraplate earthquakes are still poorly understood, and what caused the 19th century temblors, and how much longer they might remain a hazard in the region, remains hotly debated (see Geotimes online, Web Extra, May 28, 2004).

At plate boundaries, “we know that earthquakes tend to repeat,” says Eric Calais, a geophysicist at Purdue University. “There is little doubt that there will be a major earthquake on the San Andreas Fault sometime in the next 1,000 years.” Knowing that, geophysicists and seismologists have developed statistical methods to estimate earthquake hazards in those regions, but the same rules may not apply to intraplate earthquakes, he says.

Some studies have suggested that the reason the New Madrid earthquakes occurred is that the seismic zone is significantly hotter, and therefore is structurally weaker, than its surroundings — creating a zone of weakness where stresses building up in the plate become concentrated in the upper crust by the hot lower crust and mantle. Additionally, a team of scientists examining the movement of the Reelfoot Fault from GPS data reported in Nature in 2005 that the fault appeared to be accumulating strain at rates similar to the San Andreas Fault in California — suggesting that the New Madrid is still very much alive.

If the New Madrid is not hotter and weaker, however, earthquake bursts such as the 19th century temblors may be only transient, says Seth Stein, a seismologist at Northwestern University. That means that the seismic activity could eventually migrate to another “weak zone” in the North American plate, such as the Midcontinent Rift system stretching from Lake Superior to Kansas, he says.

To determine how hot the seismic zone is, Stein and his colleagues analyzed data from a new “geothermal map” of North America, which measures heat flow in boreholes across the continent to determine how much heat is rising to the surface. Reporting at the American Geophysical Union’s December meeting in San Francisco, the team found that based on this new data, the seismic zone does not appear to be significantly hotter than elsewhere in the United States.

Furthermore, they reported, seismic activity in the region may actually be tapering off. Although the seismic zone has been active for at least 2,000 years, rumbling into life about every 500 years (see Geotimes, October 2006), Stein says that GPS data analyzed by he, Calais and others have contradicted this, and suggested the fault has not moved much in recent years. That could mean that the Reelfoot Fault is now “cold and dying,” he says. If so, small aftershocks of the 1811 to 1812 earthquakes will continue, but large earthquakes could not reoccur for thousands of years. Thus, the potential hazard to the region has been significantly “overestimated” by hazard planners, meaning a lot of money could be wasted on constructing seismically safe buildings to California standards, Stein says.

Residents shouldn’t stop preparing for earthquakes, however. “I’m not sure that scientists have the answer yet,” says Calais, who was not an author on the study. The idea that intraplate earthquakes may migrate is “stimulating,” but hard to test, because the evidence may no longer exist, Calais says. If small bursts of earthquakes did occur elsewhere in the Midwest 10,000 years ago before moving to the New Madrid seismic zone, any topographic evidence would have been long since eroded by glaciers or other forces, he says. Still, he says, “I like the idea because it shakes us up a little bit. It’s going to trigger interesting research.”

Carolyn Gramling

"New New Madrid findings," Geotimes online, Web Extra, May 28, 2004
"River bends reveal past quakes," Geotimes, October 2006

Back to top


Advertise in Geotimes

Geotimes Home | AGI Home | Information Services | Geoscience Education | Public Policy | Programs | Publications | Careers

© 2019 American Geological Institute. All rights reserved. Any copying, redistribution or retransmission of any of the contents of this service without the express written consent of the American Geological Institute is expressly prohibited. For all electronic copyright requests, visit: