Support Provided ByLearn More
Planet EarthPlanet Earth

Midwest Earthquakes: Expert Q&A

On September 8, 2009, Beatrice Magnani answered selected viewer questions about the New Madrid Seismic Zone, the two new faults she and her team discovered along the Mississippi River, and more.


Receive emails about upcoming NOVA programs and related content, as well as featured reporting about current events through a science lens.

Q: How likely is it that this fault will slip, and what is the potential magnitude of the fault and consequent danger to the surrounding area? Bryon Black, Woodruff, Wisconsin

Support Provided ByLearn More
Beatrice Magnani: We have no reason to believe that the two faults we discovered during our seismic survey along the Mississippi River are presently active. Although the faults show displacement of the Quaternary alluvium, which indicates activity at least during the last two million years, no seismicity is observed along them today.

Q: Is there any evidence when or if there were any earthquakes before the large New Madrid 1811 quake?

Also, what is the best current answer as to why such large quakes are occurring on a supposedly non-tectonic boundary? Ted Pierce, Berkeley, California

Magnani: I am not aware of any foreshock sequence in the case of the 1811–1812 events. Consider that at that time the Mississippi Valley was scarcely populated and, unlike today, there were no seismometers on the ground to record even the slightest ground motion. If there were foreshocks before the destructive events of the 1811–1812, they might have gone undetected.

As for why there are large magnitude earthquakes so far away from a plate boundary is still a puzzle, and we don't have a good answer yet. We are working on it!

Q: Hi. From time to time we hear about some earthquake in the news, and I'm wondering if it is possible to predict the onset of an earthquake say a few days before it happens. Thanks! Kennedy Junior High School, 8th Grade, Naperville, Illinois

Magnani: Kennedy,
I wish we could predict earthquakes a few days before they happen! Alas, we can't. We can't tell when and where exactly an earthquake of a particular magnitude will occur, but we can calculate the probability of an earthquake of a certain magnitude occurring in a determinate place over a certain number of years.

Q: Dear Dr. Magnani,
I heard that "sand blows"—evidence of previous major seismic events—have recently been found in far southeast Arkansas and northeast Louisiana. This is well over 100 miles south of the New Madrid Seismic Zone. Is it likely that these features were formed by earthquakes with epicenters in far southeastern Arkansas, an area not traditionally known for seismic activity? Chris King, North Little Rock, Arkansas

Magnani: Indeed, it is likely that a fault or more faults, additional to the faults that are now generating earthquakes in the New Madrid Seismic Zone, might have been active in the area and might have caused the sand blows found in southeast Arkansas and northeast Louisiana. Discovering these faults is the main goal of the Mississippi River seismic survey my colleague and I are conducting.

Q: Are we looking at a California-type potential of an earthquake environment in Memphis? If so, what is the time frame that we are looking at for this to begin, according to the current geological circumstances? Randy G., Memphis, Tennessee

Magnani: I am assuming you are referring to the faults newly discovered along the river. We have no indication that these faults are active today, although they show activity at least during the Quaternary, which, geologically speaking, is considered recent compared to the long tectonic history of this part of the continent.

Q: I have heard the New Madrid Fault described as a "failed rift." Can you please explain the difference between this and a regular garden-variety fault like the San Andreas? Michael LaVoice, Millington, Tennessee

Magnani: In general, a failed rift (or aulacogen) is a tectonic feature that forms when a continent is pulled apart by tensional stresses imposed by plate tectonic forces. If the rifting process continues long enough (i.e., millions of years), the continent continues to thin, eventually breaks apart, and a new ocean forms between the rifted margins. The Atlantic Ocean is a typical example of a successful rifting process that broke the supercontinent Pangea during the Mesozoic.

In some cases, the rifting process ends before the continental break-up, and the failed rift becomes inactive, subsides, and is buried under sediments, remaining like a scar in the middle of the continent.

Rifting processes are associated with normal (extensional) faults, where the blocks slip away from each other. Although a component of strike-slip movement can be present, vertical motion dominates. Along strike-slip faults, such as the San Andreas, the blocks slide past each other, with little vertical movement relative to the horizontal motion.

Faults that were active during the rifting process might become reactivated at a later time, under favorable conditions. This is the case of the New Madrid fault system, located within the Reelfoot rift that failed ~500 million years ago.

Q: Is there a correlation between the path of the Mississippi River and the fault? Which is older, the river or the fault? Mike Spagle, Huntington Beach, California

Magnani: We have reasons to believe that the faults we discovered have been active for a long time, possibly for the past 60 million years. The present Mississippi river valley was mostly shaped during the recent glacial stage of the Great Ice Age, between 75,000 and 12,000 years ago. So where the Mississippi River flows through the New Madrid Seismic Zone, the faults that are active today are deforming the landscape, controlling the path of the river.

Q: Does mining in an earthquake area exacerbate an earthquake? A. Wiseman, Alton, Illinois

Magnani: Mining usually occurs at depths that are much shallower than most seismogenic zones (i.e., the portion of the crust that is capable of generating earthquakes). However, if mining activity is extensive (e.g., in South Africa), then it could affect the state of stress of the upper crust, and it might, in turn, affect the occurrence of earthquakes along favorably oriented faults.

Q: A week ago on NOVA they talked about how fractals can be used to measure shorelines, veins, trees, and heartbeats. Could fractals be used to measure and predict earthquakes? James Fritzler

Magnani: Fractals can be used to model both distribution of faults with characteristic earthquakes and distribution of earthquakes along one fault. Tectonic models for fractal distribution of faults have been proposed since the 1980s.

Q: Could an earthquake in the Midwest cause a shift in the flow of the Mississippi River, as it did in the 1811–12 earthquakes, and could that have effects as far south as New Orleans?

Secondly, can anything be done to prepare the people living in the Midwest for such an earthquake? Anonymous

Magnani: The most active seismic area in the Midwest, the New Madrid Seismic Zone, is indeed capable of generating earthquakes that can rupture the Mississippi River bed and cause it to flow backwards. We know that the 1811–1812 earthquakes were felt over a large region (up to about 1,000 miles away) and that damage was widespread over an area of 450 miles radius.

A great effort is being made at the federal, state, and city level to prepare the Midwest for the occurrence of a large-magnitude earthquake such as those that hit this region in 1811–1812. These efforts include increasing the awareness of the seismic hazard in this region; reducing earthquake risk to people, property, systems, and communities; and supporting research to understand the spatial and time distribution of earthquakes as well as their effects.