Lake Superior, North America
A failure helps explain why some continents don’t separate.
Around 1.1 billion years ago, a process capable of redrawing the planet’s map was halfway done. In the center of what is today the United States, a huge tectonic fracture, called Midcontinent Riftbegan to open up, in an attempt to divide North America.
If it had advanced all the way, explains , it would have separated the continent into two distinct masses. But the “tear” failed: the weakened lithosphere gave way under pressure, formed a basin in the crust and ended up contributing to the geological configuration that would eventually host the Lake Superior — the largest of the 5 Great Lakes in North America and the largest freshwater lake in the world.
Despite being one of the most striking geological structures on that continent, especially because it crosses the Great Lakes region and is associated with a band of around 3000 km of deeply buried igneous and sedimentary rocks, the reason why this rift did not evolve into a complete rupture remained for decades as a scientific puzzle.
But now, high-resolution computer simulations are beginning to shed light on the mechanisms that lead to the failure of continental rifts.
A new study, in Scientific Reports in October, describes how scientists used supercomputing provided by the NSF ACCESS program and the Stampede3 system at the Texas Advanced Computer Center (TACC) to run a series of numerical models to systematically test how and when a rift loses the “momentum” needed to progress toward complete separation from a continent.
According to the authors, the behavior of a rift as it approaches collapse can be understood from three key parameters. These parameters are: How much the driving force (the tension that stretches the plate) decreases, how quickly this decrease happens and how mature is the rift at the moment it begins to lose strength.
By varying these factors in dozens of simulations, the researchers were able to delimit the conditions in which continental extension proceeds to a complete rupture and those in which the process stagnates, generating a failed rift.
To reproduce tectonic evolution over millions of years, the team simulated the stretching of a tectonic plate from the margins, but instead of imposing a fixed speed of separation, they applied forcing boundary conditions. The models were built with the forces that promote rift opening (including the gravitational component) and with the factors that resist the process, such as the cooling and strengthening of the lithosphere, the intrinsic resistance of the rift and the drag of the mantle.
Each two-dimensional model used 128 cores and ran for approximately two days to achieve a simulation time equivalent to 20 million years. In total, the manuscript includes 23 models.
The results confirm that a marked reduction in driving force tends to stop rifting and prevents continental rupturewhile a small reduction has little effect, allowing the process to continue until separation. But the study also revealed that, if the driving force decreases very slowly, the rift can gain time to mature, and this maturation makes it progressively weaker, increasing the likelihood that the continent will eventually break apart.
Practically, a rift has a better chance of completing rupture if it only starts to lose strength at a sufficiently advanced stage of its development.
The authors argue that this approach helps to link, in a quantitative and testable way, the Earth’s deep dynamics to surface evolution, explaining why there are both “successful” rifts, which lead to the formation of oceans, and persistent rifts that never materialize.
