The sleeping dragon

When most people think of seismically-active regions, Bangladesh is not ordinarily near the top of the list.

Cyclones, sure.  They come roaring up the Bay of Bengal with a horrifying regularity, and most of the country is low enough in elevation that the storms barely even slow down.  The worst was the 1970 Bhola cyclone, which still holds a record as the deadliest storm in recorded history.  The official death toll was five hundred thousand, but is likely higher than that, mostly people who lived in the lowlands near the city of Chittagong.

Unfortunately for the citizens of Bangladesh, though, they're also at high risk for earthquakes -- something that has only been recognized recently.  A 2021 study led by Muhammad Qumrul Hassan of the University of Dhaka found that the region is right on top of the junction of three different tectonic plates, the Eurasian Plate, the Indian Plate, and the small Burma Plate ("small" here means geographic area, not capacity for damage -- the devastating 2004 earthquake and tsunami was caused by a slippage of the Burma Plate relative to the Indian Plate).  But the compression and twisting of the land near the junction has created enough stresses that the entire country is crisscrossed with faults, most notably the Dauki Fault and the Haflong Thrust (which crosses into the Indian states of Meghalaya and Assam to the north).

The whole thing is exceedingly complex, and still poorly understood.  Imagine laying a sheet of pie crust on a table, and you and two friends each stand around it and push, pull, or twist it from the edge.  The sheet will wrinkle, tear, and hump up in places, but exactly where those deformations will end up isn't easily predictable because it depends on where there was weakness in the dough before you started messing with it.  This is the situation with the chunk of the Earth's crust that underlies Bangladesh.  Add to that the fact that the region is poor, and much of it is jungle- or swamp-covered and pretty inaccessible to study, and you have a picture of the extent to which we don't understand the situation.

However -- alarmingly -- a 2016 study found that the entire region has been building up stress for at least four hundred years, meaning when the some piece of fault slips, it's likely to be catastrophic.

The whole topic comes up because of a rather terrifying discovery that was the subject of a paper this week in Nature Communications.  Geoscientists Elizabeth Chamberlain (of Wageningen University). Michael Steckler (of Columbia Univeristy), and colleagues were studying a puzzling historical shift in the channel of the Ganges River, and quite by accident -- it was in an area some locals were digging in to create a pond -- they saw the unmistakable signs of seismites.  These are features in rock layers created by massive earthquakes, in this case a column of sand that had erupted through pre-existing strata during a colossal temblor.  Upon analysis, they found that the river had changed course because of a massive earthquake about 2,500 years ago.

Imagine an event big enough to shift the path of a river that size.

A change in the course of a river is called an avulsion, and it normally takes decades or centuries.  (It's an avulsion of the Mississippi River that the levee system in southern Louisiana is attempting to prevent -- something I wrote about a couple of weeks ago.)  Seismic avulsions are much less common, but when they happen it's sudden and spectacular.  The only other one I've ever heard of is the shift in the Mississippi caused by the 1812 New Madrid earthquake, which dropped the land so much it cut off a meander and created Reelfoot Lake.

The seismic record in Bangladesh indicates that they're dangerously at risk for another earthquake -- and because of the complexity and our lack of comprehension of the fault system underlying the country, the geologists aren't certain where is likeliest to rupture.  There's a sleeping dragon underneath one of the poorest countries in Asia -- and we're only beginning to understand when and how it might suddenly awaken.

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

Shaky ground

A little less than six years apart -- on 1 November 1755 and 31 March 1761 -- two major earthquakes struck the country of Portugal, each time generating a tsunami that devastated the capital city of Lisbon.

They were both huge, although given that this was before the invention of the seismometer, we can only guess at how big; estimates are that the 1761 quake was around 8.5 on the Richter Scale, while the 1755 one may have been as high as 9.0.  Each time, the tremors were felt far from the epicenter.  The shaking from the 1755 quake was recorded as far away as Finland.

The effects in Portugal and nearby nations were devastating.  In 1755 the combined death toll in Portugal, Spain, and Morocco -- mostly from the tsunami -- is estimated at fifty thousand.  Over eighty percent of the buildings in Lisbon were damaged or completely destroyed -- and five and a half years later, many of the ones that had survived in 1755 collapsed.

Ruins of the Convento do Carmo, which was destroyed in the Great Lisbon Earthquake of 1755 [Image licensed under the Creative Commons Chris Adams, Convento do Carmo ruins in Lisbon, CC BY-SA 3.0]

What's curious is that Portugal isn't ordinarily thought to be high on the list of seismically-active nations.  It's not on the Ring of Fire, where the majority of the world's earthquakes and volcanoes occur.  The fact is, though, there is a poorly-studied (and poorly-understood) fault zone offshore -- the Azores-Gibraltar Transform Fault -- that is thought to have been responsible for both of the huge eighteenth century quakes, as well as a smaller (but still considerable) earthquake in 1816.

The AGTF, and how it's evolving, was the subject of a paper in the journal Geology last week.  The big picture here has to do with the Wilson Cycle -- named after plate tectonics pioneer John Tuzo Wilson -- which has to do with how the Earth's crust is formed, moved, and eventually destroyed.

At its simplest level, the Wilson Cycle has two main pieces -- divergent zones (or rifts) where oceanic crust is created, pushing plates apart, and convergent zones (or trenches) where oceanic crust is subducted back into the mantle and destroyed.  Right now, one of the main divergent zones is the Mid-Atlantic Rift, which is why the Atlantic Ocean is gradually widening; the Pacific, on the other hand, is largely surrounded by convergent zones, so it's getting smaller.

Of course, the real situation is considerably more complex.  In some places the plates are moving parallel to the faults; these are transform (or strike-slip) faults, like the AGTF and the more famous San Andreas Fault.  And what the new paper found was that the movement along the AGTF doesn't just involve side-by-side movement, but there's a component of compression.

So the Azores-Gibraltar Transform Fault, in essence, is trying to turn into a new subduction zone.

"[These are] some of the oldest pieces of crust on Earth, super strong and rigid -- if it were any younger, the subducting plate would just break off and subduction would come to a halt," said João Duarte, of the University of Lisbon, who lead the research, in an interview with Science Daily.  "Still, it is just barely strong enough to make it, and thus moves very slowly."

The upshot is that subduction appears to be invading the eastern Atlantic, a process that (in tens or hundreds of millions of years) will result in the Atlantic Ocean closing up once more.  The authors write:

[T]he Atlantic already has two subduction zones, the Lesser Antilles and the Scotia arcs.  These subduction zones have been forced from the nearby Pacific subduction zones.  The Gibraltar arc is another place where a subduction zone is invading the Atlantic.  This corresponds to a direct migration of a subduction zone that developed in the closing Mediterranean Basin.  Nevertheless, few authors consider the Gibraltar subduction to be still active because it has significantly slowed down in the past millions of years.  Here, we use new gravity-driven geodynamic models that reproduce the evolution of the Western Mediterranean, show how the Gibraltar arc formed, and test if it is still active.  The results suggest that the arc will propagate farther into the Atlantic after a period of quiescence.  The models also show how a subduction zone starting in a closing ocean (Ligurian Ocean) can migrate into a new opening ocean (Atlantic) through a narrow oceanic corridor.

So the massive Portugal quakes of the eighteenth and nineteenth centuries seem to be part of a larger process, where compression along a (mostly) transform fault is going to result in the formation of a trench.  It's amazing to me how much we've learned in only sixty-odd years -- Wilson and his colleagues only published their seminal papers that established the science of plate tectonics between 1963 and 1968 -- and how much we are still continuing to learn.

And along the way elucidating the processes that generated some of the biggest earthquakes ever recorded.

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

Hot times

In today's contribution from the Completely Useless Advice department: if you own property in southern Africa, you might want to consider selling it some time in the next ten million years or so.

The reason I say this is because of a paper published a couple of months ago in Nature Geoscience that was once again thrown my way by my pal Gil Miller, who seems to have an inordinate talent at ferreting out truly fascinating stuff I hadn't heard about.  The paper is entitled "A Tree of Indo-African Mantle Plumes Imaged by Seismic Tomography," by Maria Tsekhmistrenko, Karin Sigloch, and Kasra Hosseini (of Oxford University), and Guilhem Barruol (of the Université de Paris), and describes the structure of the mysterious "hotspots" -- upwelling of extremely hot magma from deep in the mantle -- that are responsible for such volcanically-active regions as Hawaii, Yellowstone, and Réunion Island.

These hotspots have long puzzled geologists, because they are quite distant from tectonic plate boundaries, where most of the world's seismic and volcanic activity occurs.  Hawaii is the best-studied hotspot; it was one of the most powerful pieces of evidence of plate movement, back in the 1960s when the theory of plate tectonics was first being studied.  The Big Island of Hawaii is just the easternmost point in a chain that extends way beyond what we usually think of as the Hawaiian Islands; even the westernmost island that pokes up above sea level, Kure Atoll, isn't the end of it.  It continues into the Emperor Seamount Chain, which extends underwater all the way to the Kamchatka Peninsula of Siberia. 

My long-ago geology professor described it as being like pulling a piece of fabric (the Pacific Plate) through an upside-down sewing machine (the Hawaiian Hotspot); the needle of the sewing machine punches regular holes upward through the fabric as it moves through, but the sewing machine itself stays in the same place.  The plates are moving; the hotspot isn't.  (And the angle in the chain of seamounts indicates that at some point in the past, the Pacific Plate changed direction, probably because of jostling against other plates.)

The Pacific Ocean floor, showing the Hawaiian-Emperor Seamount Chain [Image is in the Public Domain courtesy of NOAA]

What is still mysterious about hotspots is why they happen at all.  We have a pretty decent idea of why the activity along plate margins occurs -- strike-slip faults like the famous San Andreas, where two plates are moving along each other in opposite directions; trenches/subduction zones like Indonesia, where you get both powerful quakes and huge volcanoes; and mid-ocean ridges/divergent zones like the Mid-Atlantic Ridge, where plates are moving apart and new magma upwells to fill the gaps.  But why would there be a persistent chain of volcanoes out in the middle of a stable plate?

The current paper describes blobs of extremely hot magma originating from the lower parts of the mantle, which rise and then diverge into branches.  The authors write:

Mantle plumes were conceived as thin, vertical conduits in which buoyant, hot rock from the lowermost mantle rises to Earth’s surface, manifesting as hotspot-type volcanism far from plate boundaries.  Spatially correlated with hotspots are two vast provinces of slow seismic wave propagation in the lowermost mantle, probably representing the heat reservoirs that feed plumes...  Using seismic waves that sample the deepest mantle extensively, we show that mantle upwellings are arranged in a tree-like structure.  From a central, compact trunk below ~1,500 km depth, three branches tilt outwards and up towards various Indo-Austral hotspots.  We propose that each tilting branch represents an alignment of vertically rising blobs or proto-plumes, which detached in a linear staggered sequence from their underlying low-velocity corridor at the core–mantle boundary.  Once a blob reaches the viscosity discontinuity between lower and upper mantle, it spawns a ‘classical’ plume-head/plume-tail sequence.

So the Réunion Hotspot is apparently connected to the East African Rift Zone, three-thousand-odd kilometers away.  The EARZ is a developing rift that is ultimately going to shear off the "Horn of Africa," opening a new ocean and creating a new "microcontinent" made up Somalia and bits of Ethiopia, Kenya, and Tanzania.  (As an aside, it's also the site of Olduvai Gorge, where some of the earliest hominin fossils were found.)

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

"From looking at the core-mantle boundary, you can maybe predict where the oceans will open,” said study co-author Karin Sigloch.  "If the new models are accurate, a few tens of millions of years from now, you may not want to be in South Africa — or, perhaps, on planet Earth at all."

The reason Sigloch says this is that the team's analysis of the "tree" of magma that underlies both Réunion and the EARZ suggests that it's in the process of forming another branch -- another mantle plume -- that will ultimately end up underneath what is now South Africa.  "In tens of millions of years, a blob of nightmarishly gargantuan proportions will pinch off from the central cusp," Sigloch said, in an interview with Quanta magazine.  "This would produce cataclysmic eruptions.  The Deccan Traps [one of the largest volcanic eruptions ever, and which probably contributed to the extinction of the non-avian dinosaurs 66 million years ago] were caused by what we would think of as a solitary mantle plume.  This future mega-blob, though, would be capable of producing volcanism so prolific and extensive that the Deccan Traps would be a firecracker in comparison."

Pretty scary.  But like I said, if you want to visit South Africa, or if you live there, you still have a ten-million-year window to take care of business.  What's interesting from a geological perspective is that up till now, South Africa has been very stable tectonically.  The majority of the country is made of extremely old rock, what geologists call a "craton" -- a chunk of some of the oldest continents on Earth.  A massive flood basalt eruption, like the Deccan Traps, the Columbia River Flood Basalts, and the largest of them all -- the Siberian Traps, implicated in the cataclysmic Permian-Triassic Extinction -- would (literally) overturn three billion years of stable geology, with catastrophic results for the entire planet.

So yeah.  That's cheerful.  But since we have ten million years before we have anything serious to worry about, it'd be better if to turn your attention to more pressing concerns, even if you live in Johannesburg.  Like what we're doing to destroy the global ecosystem our own selves by our seeming commitment to burn every last gallon of fossil fuels out there, damn the climate, full speed ahead, and which could make the Earth pretty close to uninhabitable a great deal sooner. 

Which now that I think of it, isn't all that reassuring.

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

Like graphic novels?  Like bizarre and mind-blowing ideas from subatomic physics?

Have I got a book for you.

Described as "Tintin meets Brian Cox," Mysteries of the Quantum Universe is a graphic novel about the explorations of a researcher, Bob, and his dog Rick, as they investigate some of the weirdest corners of quantum physics -- and present it at a level that is accessible (and extremely entertaining) to the layperson.  The author Thibault Damour is a theoretical physicist, so his expertise in the cutting edge of physics, coupled with delightful illustrations by artist Mathieu Burniat, make for delightful reading.  This one should be in every science aficionado's to-read stack!

[Note: if you purchase this book using the image/link below, part of the proceeds goes to support Skeptophilia!]