The bellringer

Between December 16, 1811 and February 7, 1812, a series of four earthquakes -- each estimated to be above magnitude 7, with the first and last perhaps at magnitude 8 -- hit what you might think is one of the most unlikely places on Earth; southeastern Missouri.

The centers of continents are ordinarily thought to be tectonically stable, as they are generally far from any of the three typical sorts of faults -- divergences, or rifts, where plates are moving apart (e.g. the East African Rift Zone); convergences, or thrust faults, where plates are moving together (e.g. the Cascadia Subduction Zone); and strike-slip faults, where plates are moving in opposite directions parallel to the fault (e.g. the San Andreas Fault).  The Midwest is located in the middle of the North American Craton, an enormous block of what should, according to the conventional wisdom, be old, stable, geologically inactive rock. 

But the 1811-1812 earthquake series happened anyhow.  If they'd occurred today, it would likely have flattened the nearby city of Memphis, Tennessee.

So much for conventional wisdom.

The fault responsible was named the New Madrid Seismic Zone for the county right in the center of it, and its capacity for huge temblors is staggering.  The biggest (and final) earthquake of the four was powerful enough that it was felt thousands of kilometers away, and rang church bells in Charleston, South Carolina.  The shift in terrain changed the course of the Mississippi River, cutting off a meander and creating horseshoe-shaped Reelfoot Lake.

So what created a seismic zone where one shouldn't be?

[Image is in the Public Domain courtesy of the USGS]

The topic comes up because I just finished reading seismologist Susan Elizabeth Hough's excellent book Earthshaking Science: What We Know (and Don't Know) About Earthquakes, which is one of the best laypersons' introductions to plate tectonics and seismicity I've come across.  She devotes a good bit of space to the New Madrid earthquakes, and -- ultimately -- admits that the answer to this particular question is, "We're still not sure."  The problem is, the fault is deeply buried under layers of sediments; current estimates are that the hypocenter (the point directly underneath the epicenter where the fault rupture occurred) is between fifteen and thirty kilometers beneath the surface.  And since the quakes in question happened before seismometers were invented, we're going off inferences from written records, and such traces that were left on the surface (such as "sand blows," where compression forces subsurface sand upward through cracks in the stratum, and it explodes through the surface).

As far as the cause, Hough has a plausible explanation; the New Madrid Seismic Zone is an example of a failed rift, where a mantle plume (or hotspot) tried to crack the continent in half, but didn't succeed.  This stretched the plate and created a weak point -- called the Reelfoot Rift -- where any subsequent stresses were likely to trigger a rupture.  Since that time, the North American Plate has been continuously pushed by convection at the Mid Atlantic Rift, which is compressing the entire plate from east to west; those stresses cause buckling at vulnerable points, and may well have been the origin of the New Madrid earthquakes.

One puzzle, though, is what happened to the hotspot since then.  This is still a matter of speculation.  Some geologists think that friction with the rigid and (relatively) cold underside of the plate damped down the mantle plume and ultimately shut down convection.  Others think that as the North American Plate moved, it simply slid off the hotspot, making the plume appear to move eastward (when in actuality, the plate itself was moving westward).  This may be why another anomalous mid-plate earthquake zone is in coastal South Carolina, and it might also be the cause of the Bermuda Rise.

That point is still being debated.

Another open question is the current risk of the fault failing again.  There's paleoseismic data suggesting major earthquake sequences from the Reelfoot Rift/New Madrid Seismic Zone in around 900 and 1400 C.E., suggesting a timing between events of about four to five hundred years.  But these are estimates themselves, and I probably don't need to tell you that earthquake prediction is still far from precise.  Faults don't fail on a schedule -- which is why it annoys me every time I see someone say that an area is "overdue for an earthquake," as if they were on some kind of calendar.

Still, I can say with at least moderate confidence that it's unlikely to generate another big earthquake soon, which is kind of a relief.

So that's our geological curiosity of the day.  I have a curious family connection to the area; my wandering ancestor Sarah (Handsberry) Overby-Biles-Rulong (she married three times, had nine children, and outlived all three husbands) lived in the town of New Madrid in 1800, after traveling there from her home near Philadelphia as a single woman in the last decade of the eighteenth century.  I've never been able to discover what impelled her to leave her home and, with a group of relative strangers, cross what was then trackless wilderness to a remote outpost, and I've often wondered if she might have been either running away from something, or perhaps might have been a prostitute.  I'm not trying to malign her memory; it bears mention that a good eighty percent of my forebears were rogues, ne'er-do-wells, miscreants, and petty criminals, so it would hardly be a surprise to add prostitution to the mix.  And whatever else you can say about my family members, they were interesting.  I've often wished I could magically get a hold of Sarah's diary.

In any case, Sarah was in Lafayette, Louisiana by 1801, so she missed the New Madrid earthquakes by ten years.  But kind of interesting that she lived for a time in the little village that was about to be the epicenter of one of the biggest earthquakes ever to hit the continental United States, one that rang bells thousands of kilometers away, and which created a geological mystery the scientists are still trying to work out.

****************************************

The hidden fault

Between December 16, 1811 and February 7, 1812, a series of four earthquakes -- each estimated to be above magnitude 7 -- hit what you might think is one of the most unlikely places on Earth; southeastern Missouri.

The fault (named the New Madrid Seismic Zone for the county right in the center of it) is located in the middle of the North American craton, an enormous block of what should be old, stable, geologically inactive rock.  But even so, the biggest (and final) earthquake of the four was powerful enough that it was felt thousands of kilometers away, and allegedly rang church bells in Charleston, South Carolina.  The shift in terrain changed the course of the Mississippi River, cutting off a meander and creating horseshoe-shaped Reelfoot Lake.

It's well known that most of the world's earthquakes take place along the "Ring of Fire" and other junctions between tectonic plates, but it's not always so.  The New Madrid Fault is thought to be either a failed rift zone -- when a convection current in the mantle tried, but failed, to split the continent, but created a weakness in the middle of the plate -- or else the rebound of the crust from the passage of the Bermuda Hotspot, which is also one possible explanation for the process that created the Ozark Mountains.

The point is, earthquakes don't always occur where you might expect, and sometimes fault lines can stay hidden until suddenly they slip and catch everyone off-guard.  This is the situation much closer to where I live; the Saint Lawrence Rift System, aligned (as you'd expect) with the Saint Lawrence River, is an active seismic zone in northern New York and southern Canada, and like New Madrid, is very far away from any plate margins.  Here, the weakness is very old -- geologists believe the fault actually dates to the early Paleozoic, and may be related to the Charlevoix Asteroid Impact 450 million years ago -- and has been reactivated by something that is causing super slow convergence on opposite sides of the fault (on the order of 0.5 millimeters a year).

What that something might be, no one is certain.

The reason the topic comes up is a paper in the journal Tectonics this week that I found out about because of my friend, the wonderful author Andrew Butters, who is an avid science buff and a frequent contributor of topics for Skeptophilia.  It describes a newly-discovered 72-kilometer-long fault that runs right down the middle of Vancouver Island -- passing just northeast of the city of Victoria.

To be fair, British Columbia isn't exactly seismically inactive; as I described last month, it's in the bullseye (along with the rest of the coastal Pacific Northwest) of the horrifyingly huge Cascadia Subduction Zone.  But even so, the discovery of a hitherto-unknown fault right near a major city is a little alarming, especially since the southeast corner of Vancouver Island is actually pretty far away from Cascadia.  The authors write:

Subduction forearcs are subject to seismic hazard from upper plate faults that are often invisible to instrumental monitoring networks.  Identifying active faults in forearcs therefore requires integration of geomorphic, geologic, and paleoseismic data.  We demonstrate the utility of a combined approach in a densely populated region of Vancouver Island, Canada, by combining remote sensing, historical imagery, field investigations, and shallow geophysical surveys to identify a previously unrecognized active fault, the XEOLXELEK-Elk Lake fault, in the northern Cascadia forearc, ∼10 km north of the city of Victoria...  Fault scaling relations suggest a M 6.1–7.6 earthquake with a 13 to 73-km-long surface rupture and 2.3–3.2 m of dip slip may be responsible for the deformation observed in the paleoseismic trench.  An earthquake near this magnitude in Greater Victoria could result in major damage, and our results highlight the importance of augmenting instrumental monitoring networks with remote sensing and field studies to identify and characterize active faults in similarly challenging environments.

So that's a little alarming.  Another thing to file under "You Think You're Safe, But..."  I've frequently given thanks for the fact that I live in a relatively calm part of the world.  Upstate New York gets snowstorms sometimes, but nothing like the howling blizzards of the upper Midwest; and we're very far away from the target areas for hurricanes, mudslides, wildfires, and volcanoes.

But the scary truth is that nowhere is natural-disaster-proof.  As New Madrid, the Saint Lawrence Rift System, and -- now -- Victoria, British Columbia show, we live on an active, turbulent planet that is constantly in motion.  And sometimes that motion makes it a little dangerous for us fragile humans.

The Earth is awe-inspiring and beautiful, but also has little regard for our day-to-day affairs.  You can do what is possible to minimize your risk; forewarned is forearmed, as the old saying goes.  But the reality is that the natural world is full of surprises -- and some of those surprises can be downright dangerous.

****************************************