RSS Feed
iTunes Link

Episode 53: Lunar Formation and Origins

Download the Episode

Recap: A run-down of the various models of how the moon formed. This episode doesn't have much pseudoscience to go with it, but it's a good example of the scientific process at work.

There was no puzzler in episode 52.

Puzzler: There was no puzzler for this episode.

Q&A: There was no Q&A for this episode.

New News:

New News from Other Episodes:

  • From Episode 54
    • Article 1: Cuk, M. and S.T. Stewart, "Making the Moon from a Fast-Spinning Earth: A Giant Impact Followed by Resonant Despinning."
    • Press release for Cuk & Stewart, including Movie showing angular momentum transfer simulation.
    • Article 2: Canup, R.M., "Forming a Moon with an Earth-Like Composition via a Giant Impact."
    • The first article is by Cuk and Stewart who reported on simulations of a giant impact that showed that the conditions for the impact to form a moon are not as restrictive as previously thought. In other words, the initial modeling of the impact event indicated that the Mars-sized object would have had to strike Earth at an angle, like a glancing blow, in order to get an outcome in which a moon formed and didn't destroy Earth.
    • These new simulations show that if the early Earth were to spin faster, the impactor could have struck closer to dead-on. They also showed that the material that would have formed the moon would primarily come from Earth's mantle, which helps to solve some of the isotopes of heavy elements that I briefly addressed in the last episode.
    • Meanwhile, another article came out by Robin Canup, who I may be able to interview on this podcast early next year. She's one of the originators of modeling the giant impact scenario and has been modeling it for over a decade. Her paper suggests that Earth may not have been hit by a Mars-sized object, but by an Earth-sized object. The combined planet would throw off a disk of debris that would also be primarily made of Earth's mantle, which also helps solve the isotope issue.
    • Both papers also rely on the Cuk and Stewart finding that angular momentum can be more easily removed from the Earth-Moon system than previously thought. One of the issues with having such a large impact, or Earth spinning around twice as fast as previously thought, is that the angular momentum of the combined system would be larger than we observe today, which is still abnormally large as I talked about last time.
    • But, Cuk and Stewart showed that soon after the Moon's formation, the Moon would have been in an elliptical orbit, and from gravitational gravitational interactions with the Sun over a very long time, the Moon can effectively bleed angular momentum off from the Earth-Moon system to the Sun, solving that higher angular momentum issue. If that doesn't quite make sense, I'll be posting a movie on the show notes for this episode that shows the simulations on the angular momentum transfer.

Additional Materials:

CORRECTION: KREEP stands for "Potassium, Rare-Earth Elements, Phosphorus."


Claim: Earth's moon exists. How did it get here?

In order to explain how it got here, how it formed, we need to develop a model. That model has to be physically possible, and it also needs to explain properties that we observe today. It also needs to be able to make predictions about things that we DON'T yet observe but could in the future. If the model predicts anything that isn't true, then the model needs to be modified or rejected for a new one.

Hopefully it goes without explanation that we've been observing the moon for a very, very long time. Rather than go through a list of what we see, I'm going to go through the major hypotheses that are out there and, in each, explain what we observe that support it and what we observe that don't support it.

Hypothesis 1: Big Sister

One of the original hypotheses for the moon's formation was that it formed with Earth, the same way that the other planets did. This means that it coalesced out of the solar nebula near Earth, with Earth, and everything was hunky dory.

Pretty much the only good thing that this model has going for it is that it can easily explain the Moon's size. Earth's moon is very large relative to Earth, being about 1/4 the diameter and 1/100 the mass. The co-formation idea was something we already were pretty sure could happen, and it can explain the size.

But there are a lot of things that it doesn't explain and a lot of things that contradict that formation mechanism.

A prediction, if Earth and Moon formed in nearly the same location, is that they should have the same composition. They don't. Earth is very rich in iron relative to the Moon, and there are a large number of elements - what we call KREEP - that are rare on Earth but relatively abundant on the Moon. KREEP stands for Potassium, where K is the symbol for potassium; rare-earth elements, which are things like scandium, lanthanum, neodynium, samarium, europium, and thulium; and then potassium which is what the P stands for.

If the two formed in the same place, they should have the same composition, which they don't.

Another bonk against the hypothesis is that the angular momentum of the Earth-Moon system is anomalously high. Angular momentum is effectively the spin energy stored, and we have a lot of it. If we formed together like any other planet, just two, then there's no reason why we should have a large angular momentum.

Hypothesis 2: Big Net

Perhaps directly because of the high angular momentum, the next formation model proposed was capture. This means that Earth formed alone, the Moon formed alone, but early in the solar system's history, the moon came close by Earth and gravitational interactions resulted in its capture.

This was good. First, because it wasn't impossible.

Second, it could sort of explain the composition differences. If we formed in different places in the solar system, then there wasn't any particular reason we should have the same elemental abundances. Third, it could explain the high angular momentum.

The bad is that it's very improbable. In pretty much every simulation that the Earth and Moon survive, you have to have a third body to remove the energy from the Moon gravitationally to get it into a stable orbit. And getting three bodies to just happen to be in the same place and be going in the right directions at the right speeds to result in one being captured about another is very unlikely to happen. An alternative is to have Earth possess a huge and thick atmosphere to literally aerobrake the passing moon, but that is also very difficult to do based on what we know about the past environment.

Another problem is that it doesn't really explain why the Moon lacks iron. Mercury, Venus, Earth, Mars, some asteroids, and even some outer solar system moons have reasonably large iron cores. Mercury's is around 80% the size of the planet. But if the Moon has an iron core, it's very small relative to its size. Capture hypothesis can't explain that.

Hypothesis 3: Big Spin

Another proposed hypothesis was developed before the theory of plate tectonics, is that Earth spun off the Moon. The idea is that Earth was spinning so gosh darn fast that the Moon just sorta fissed off, where I use the technical definition of "fis" being where it separates. As opposed to bubbling fizzy drinks.

Anyway, the only really good thing about this model is that it can explain the lack of iron in the Moon because the Moon formed from Earth's crust and lower mantle. And we had a big basin - the Pacific Ocean - from which the Moon could have come. Problem with that is as I implied before, this was before we knew about plate tectonics and before we had dated the Pacific Ocean floor to only a few hundred million years old.

I already started to explain the bad things about this model, but to list even more ... one prediction from it is that the Moon should orbit Earth's equator because that's where it would have blobbed off. But it has a 5.2° inclination relative to Earth's equator.

Another basic physics problem is how to get the Moon to fission off. This is a problem that intro astronomy students often do, and the math works out to being that if you ignore the actual strength of the rock, you still have to get Earth spinning faster than once every 85 minutes on its axis for this to happen. It would actually need to go much faster than that because the rocks have their own strength. Getting Earth to spin that fast from any formation method we know isn't really possible.

And, it still doesn't explain the angular momentum anomaly of the Earth-Moon system.

Hypothesis 4: Big Splash

The fourth hypothesis is one that most people have probably heard of because it's the one that's generally accepted today, the Big Splash.

The basic model is that a Mars-sized planet hit Earth. Most of the core from the planet sunk into Earth, debris from Earth's crust and mantle and most of the impactor were launched into orbit, and the coalesced into the moon we know and love today. Unless you're David Nabhan and think it causes earthquakes, but I'll get to that later.

There are a lot of good things about this model that explain a lot of properties that we observe. One is that it explains the Moon's lack of iron because the majority of its core was deposited in Earth. This has the benefit of also explaining Earth's relatively large core.

It also explains the composition differences with KREEP materials, but at the same time, it can explain why the oxygen isotopes are the same. Oxygen isotopes are a way of determining if two objects formed from the same material in the solar system. Oxygen is defined as having 8 protons, but it can have a few different numbers of neutrons, and these would be isotopes. 8 neutrons and you have oxygen-16, 9 neutrons and you get oxygen-17, and so on. When you take the ratio of these isotopes, usually oxygen-18 and -17 relative to -16, and you plot rocks from different bodies in the solar system, they form different groups based on where they came from.

So, if Earth and the Moon formed in different locations and never mixed, they should have different oxygen isotopes, like Earth versus Mars. But they have the same. So, the Big Splash can explain this by some material mixing together, such as the easily vaporized oxygen versus the not easily vaporized neodynium.

This also explains the high angular momentum ... I mean, a friggin' planet hit Earth. That's going to impart a lot of energy.

It can also explain why the Moon doesn't quite orbit around the equator, why Earth is tilted on its axis, and why the moon is tidally locked -- it formed that way. We also know that impacts were very common early in the solar system and sometimes very large objects did hit each other.

What this hasn't yet been able to explain is that there are still some issues with different elements. In fact, this was a listener-requested episode after the Skeptics Guide to the Universe episode 350, where I think Bob way jumped the gun and said that this model needed to be thrown out because of some new research on titanium isotopes on the Moon and Earth. In preparation for this episode, I re-listened to Bob's 10-minute segment and I think that he misinterpreted what the authors were arguing.

He got the basic part right -- the authors were reporting on comparing isotopes of titanium from Earth rocks and lunar rocks. They found that the ratios were very close to the same, the implication being that they very likely needed to have formed from the same material. Problem is that these don't mix well like oxygen would, so how do you get them to be the same if the original pre-lunar body and Earth formed in different locations?

The authors suggested three different methods in their abstract, two of which are, I think, easily compatible with the Big Splash. Basically, both of those methods are that material from Earth blanketed the forming Moon and so that's what we're sampling, as opposed to material from the original pre-moon object.

I think it remains to be seen how the numerical models play out with this to see if it works, but the current consensus is that the Big Splash is likely, at least as a basic concept, how the Moon formed.


There are of course some other models that are out there. One that was done by press release rather than science publication a year or two ago is what I term the "big burp," where they proposed that a freak natural nuclear explosion happened, ejecting a large chunk of Earth, and that chunk formed the Moon.

Another idea is that a giant ice ball crashed into Earth and ejected material.

Another one has to do with more of a glancing blow object knocking off material.

But, to my knowledge - and I'm reasonably up-to-date on this particular field - the Big Splash is the state of the art that best explains what's observed, though it's still being actively worked on.


This episode didn't really have much pseudoscience in it, but believe me, there are people today who will flagrantly misrepresent what the current state of the science is for lunar formation.

In fact, someone who I've been talking about for a few months now, Mike Bara, includes in the first chapter of his latest book several misrepresentations of lunar formation theory. His preferred model, as stated in the beginning of the book, is that God with a "big G" did it, though he looks into evidence throughout the book that it may have been gods with a "little g" that did it, AKA, ancient aliens.

But, this is not something that's often taught, and when minor mistakes are made on very popular productions, it's not an unreasonable endeavor to correct them. And so, that's the state of the science. The Big Splash idea is the one that best explains all the observations today. It's likely to be tweaked further to match new data, such as this titanium isotope data, but I haven't seen anything to suggest that the model needs to be thrown out entirely.

Provide Your Comments:

No comments have been provided.

Your Name:

Your Location:


Your Comment:

Security check *

security image