Where did the Moon come from? Scientists are using a supercomputer to find out

Researchers are running simulations of giant impacts on a supercomputer to find out how new planets come about.
Written by Daphne Leprince-Ringuet, Contributor

About 4.5 billion years ago, the early proto-planet Earth was in a huge collision with a giant object about the size of Mars. One of the huge chunks of debris that resulted ended up sticking around. And that, in theory at least, is how the Moon was born.

Many unanswered questions still remain surrounding the creation of the Earth's only satellite – how big the impactor was, at what speed and angle it was travelling before it hit the Earth, and even whether the collision happened with one or several other bodies. 

Now, a team of researchers armed with a supercomputer has come up with new tools for scientists to investigate the consequences of giant impacts, and the new insights could shed light on the creation of the Moon.

"We think that these giant impacts of proto-planets colliding is a common path of planet formation in our solar system," Jacob Kegerreis, who led the research, told ZDNet. "And one piece of the puzzle that is interesting to look at, is how much atmosphere these collisions directly remove."

SEE: Managing AI and ML in the enterprise 2020: Tech leaders increase project development and implementation (TechRepublic Premium)

For example, studies have already shown that, in the collision that is thought to have created the Moon, the Earth lost between 10% and 60% of its atmosphere. How and why is less understood, because so far, simulations of the impact have failed to account for the many parameters that determine atmospheric erosion during a collision. 

Kegerreis and his team, based at the University of Durham in the UK, therefore set out with the objective to find out how atmospheres behave when rocky planets undergo a huge collision such as the one that resulted in the formation of the moon.

"A few different projects have touched on this in the past, but they didn't explore the huge range of possible parameters," said Kegerreis. "Giant impacts can happen at different angles, speeds, masses and so on. We were trying to explore that whole parameter space, and the diversity of impacts, to understand how much atmosphere these collisions directly remove."

To do so, the researchers ran about 300 simulations of different giant impacts, altering parameters like speed, angle, impactor mass and composition to study the consequences of different types of collisions. 

Test scenarios included masses three times the Earth's mass down to a few percent of the planet's mass; head-on and grazing angles; and speeds ranging from roughly ten to 30 km/s.

The simulations were run on an open-source simulation code and carried out on the COSMA supercomputer at the University of Durham, which is part of a digital research infrastructure facility dubbed DiRAC (Distributed Research using Advanced Computing) that provides high-performance computing for science and technology research in the UK.

Kegerreis said that the compute power at hand enabled the team to effectively run 3D high-resolution scenarios of different colliding planets, at unprecedented speed. 

"Once we came up with a list of 300 or so scenarios that we thought were most interesting, it was basically a matter of equation-solving," he said. "Nothing stops us from doing this by hand, of course, but that would have taken an incredibly long time."


Cross-section snapshots of 3D simulations of giant impacts using 30 to 100 million particles, colored by their material or their internal energy, related to their temperature. 

Jacob Kegerreis, Durham University

The results showed the various effects that a giant impact can have on a planet's atmosphere based on the different factors that the researchers input in each of their scenarios. 

That is not to say that the researchers solved the mystery of the Moon's creation. Rather, they came up with a series of correlations that could be used by scientists to validate the hypotheses they may have about the giant impact that formed the satellite planet.  

"The Moon-forming impact has been studied a lot, and there are five or six different plausible theories about it," said Kegerreis. "We haven't found out exactly how the Moon was formed, but thanks to the wide range of scenarios we looked at, we got a solid handle on the effect of the parameters that were at play in forming the Moon."

For example, according to the study, what is known as the "canonical Moon-forming impact" –  whereby the Earth would have collided with a Mars-sized, low-velocity and oblique impactor – would have resulted in a loss of 10% of atmosphere. A more head-on impact at slightly greater speeds, on the other hand, would have removed much more atmosphere from the Earth.

SEE: This AI can discover the hidden links between great works of art

Beyond the Moon-forming impact, other correlations came out of the study that were previously unknown. Giant impacts between young planets and massive objects, for instance, were shown to potentially add significant atmosphere to a planet if the impactor also has a lot of atmosphere.

More worryingly, in some cases changing even only one of the variables led to the complete obliteration of the impacted planet.

Kegeirris and his team used the results from the study to develop a new prediction tool, called a scaling law, which can anticipate the atmospheric loss that can result from different giant impacts. The scaling law can be applied to any type of collision that involves planets with Earth-like thin atmospheres.

The researchers hope that the tool will speed up research in the field of giant impacts and how those huge collisions can have lasting effects on the later stages of planet formation. Kegeirris, for his part, is particularly excited at the prospect of applying the technology to study the evolution of exo-planet atmospheres.

Editorial standards