Monday Minute: Triple Junctions and the Age of the Earth

In the wake of World War II, American warships were increasingly being cast into new, scientific roles for the sake of physical oceanography. By the late 1950’s, the hastened efforts to develop effective sonar systems produced in remarkably detailed maps of the ocean floor, revealing that it was anything but a flat and passive abyss. On the contrary, it seemed the Earth had been tearing itself apart for eons, consistent with speculations about “continental drift”. The greatest testament to this fact was the discovery of Earth’s longest mountain chain—the Mid-Atlantic Ridge—which demarcates the ongoing division of the Atlantic Ocean.

We can imagine the rifting process of plate tectonics rather simply, for example by pulling on two sides of a sheet of paper. Once the tension exceeds the strength of the paper, a weak point will form and the paper should tear along a roughly straight edge:

linear fracture of a 2D plane under extensional stress

Makes sense, right? If we try to apply this concept to the Mid-Atlantic Ridge, however, things get a little messy. Consider how the ridge trends in the southern Atlantic Ocean, between South America and Africa (which were famously joined in the days of Pangaea). Obviously, the plates have separated along the ridge, but unlike our sheet of paper, the tear is anything but straight:

Mid Atlantic Ridge in the South Atlantic Sea

As you might have guessed (there is a hint in the title of my cartoon above), Earth’s tectonic plates do not fracture along straight lines over a great distance because the Earth is not flat. With a flat surface, one force is pulling the paper 180° relative to the other force. Over a spherical surface, however, this tension extends radially in all directions. To envision how the process leads to tectonic rifting, try the following experiment. First, drape tissue paper over a globe and tie the loose ends together. Now, slowly tighten the loose ends to place tension on the center of the tissue paper from all directions at once. If the force is evenly distributed, the result will look something like this:

Formation of a triple junction over a spherical surface

Under radial stress, the weak point extends along three branches, each pair forming an angle of ~120°. These branches converge at a central point known as a triple junction (represented by the red star). So long as there is tension pulling away from the triple junction, each branch will rift in a more or less linear fashion. Therefore, the layout of Earth’s tectonic plates is characterized not so much by linear fractures at right angles, but rather by pseudo-hexagonal polygons, like one finds on a Settlers of Catan game board. The process can also be displayed brilliantly in 2D using hot wax models (rifting begins at 1:30):

Oceans, rifts, and rivers in the making

As tectonic plates pull away from one another, they typically create depressions in the Earth’s surface, due to a thinning of the crust. If this process occurs within a continent, the depression often forms a linear rift valley, where the anomalously thin crust results in low elevations and active volcanism. The East African Rift, which has been forming for at least 11 million years, provides one of the best examples of the continental branch of a triple junction. Given enough time, the depression may eventually subside below sea level, giving birth to a new ocean basin.

Alternatively, it is possible for one branch of the triple junction to “fail”, meaning that rifting proceeds only for a few millions of years until the extensional stress is no longer sufficient. This phenomenon (called an aulacogen) is rather common in large continents like North America and Africa, where the Mississippi and Niger river basins now fill the remnants of two failed rifts. The other two branches continued to diverge, however, forming part of the Gulf of Mexico and the South Atlantic Ocean.

Triangles, triangles, everywhere!

Once you understand how triple junctions form and how to recognize them, they appear everywhere on physiographic maps of the world. To this end, GoogleEarth is a great tool for exploring how Earth’s modern surface came to be. Look closely at mid-ocean ridges and continental edges: how many of them are characterized by ~120° angles that separate one plate from another or define valleys and ridges? Triple junctions thus explain the shape of familiar continents, like the west coast of Africa and its Brazilian counterpart, the west coast of South America, the Arabian peninsula and Red Sea, and even the Kara Sea and Yenisei river basin in western Siberia.

A triple injunction against Ken Ham’s view of Earth history

What do these hexagonal rift zones tell us about geological history? In short, that it extends deeply into the past, far more than the ~4,500 years posited by Young-Earth Creationists who ascribe to ‘Catastrophic Plate Tectonics’.

First of all, rift valleys have the potential to accumulate massive amounts of sediment, but they do so in a predictable manner. In the early stage, sediments are characterized by weathered volcanic rocks, much like you’d find in the East African valleys today. As the rift proceeds, sediments should transition to those characteristic of a shallow oceanic basin, similar to the modern Red Sea. Finally, these will be overlain by deep-ocean silts, clays, and/or calcareous oozes.

In the young-Earth scenario, however, rifting must proceed too rapidly for the accumulation of such characteristic sediments. There is no time, for example, for volcanic flows to cool, crystallize, weather, and erode into fine sediments. Rather, we’d expect a mishmash of whatever the ocean currents of Noah’s flood dumped into the depression as it opened up.

Secondly, if triple junctions had formed rapidly in the past, then we’d expect the crust to be anomalously hot in these regions. So hot, in fact, that the crust would be ultra buoyant, due to thermal expansion and associated low density (much like a blob of hot wax in a lava lamp). Rather than definitive depressions, therefore, triple junctions would form broad, hot plateaus, that would still be boiling off the oceans today.

Thirdly, the detailed and complex history of past ecosystems recorded in aulacogens (e.g. the Mississippi River Valley) makes a mockery of the young-Earth timeline. The sedimentary successions in these basins indicate slow rifting and associated sedimentation, followed by gradual subsidence as the crust cooled (cold rock is more dense) to form wide river valleys, which accumulated eons worth of floodplain communities, including glacial-interglacial transitions.

But how do we know that we haven’t misinterpreted the history of sedimentation in rift zones?

Quite simply, because these timelines are amply corroborated by the agreement of GPS measurements with radiometric dates of ancient lava flows that formed during rifting. The correlation is so perfect as to be unequivocal. All evidence considered, therefore, we find that triple junctions are a testament to the long, dynamic processes that shaped our multibillion-year-old Earth.


Featured image: Settlers of Catan game board


7 responses to “Monday Minute: Triple Junctions and the Age of the Earth

  1. But-? What makes the crust stretch in the first place? Is It upwelling from hotspots as I have seen suggested? Does that provide enough power to split apart whole continents?


    • Good question 🙂 And there is no unanimously received answer, as far as I know. Advances in tomography, though, give some incites, and seem to suggest that forces originate from a combination of factors: 1) upwelling as you mentioned, which partially melts the crust and weakens it, 2) mantle convection beneath the plate, and 3) subduction at the other end of the plate.

      I wish I could do better, but it’s not my area of expertise. I only know enough to know that nobody knows for sure.


  2. I’m curious, what’s the fastest possible rate for oceanic crust to cool? Is it even physically possible for the oceanic crust to have reached the temperature it’s at today within 4500 years?


    • Hello Daniel,

      The short answer is no—it’s not at all possible.

      The very surface of the crust (in contact with seawater) can cool rather quickly, including the upper kilometer or so, where water circulates through the rock.

      But oceanic crust is typically >6 km thick and cools mainly by conductive heat transfer, while also being heating from the underlying mantle. We can estimate the time needed to cool using Newton’s laws of cooling.

      I’m not the best person to try and apply these laws to a complex system, however, so I would defer to others (search GoogleScholar for more papers):

      Essentially, it should take on average a few million years for the temperature of newly formed oceanic crust to drop below 500°C.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s