Although the imagers and observers among us may disparage the Moon when it's full, the Moon has fascinated humanity since we lived in caves, and perhaps before that. Once we realised that the Moon was an astronomical ... as opposed to a theological ... object, perhaps ~400 years ago (think Galileo), mankind has been asking the same fundamental question : "How did the Moon form ?". After more than half a century of active space exploration, you would be forgiven for thinking that maybe scientists have come to an agreement on the origin of our nearest astronomical neighbour, but in 2026 it doesn't seem as though the question is quite settled to everyone's satisfaction !
Over the past 10 years, from this podium I have discussed with you the most recent theories of Moon formation ... specifically in my 2016 and 2019 presentations, that you can find on my website ... if interested. However, the question "How did the Moon form ?" is still being asked within the planetary science community ... and attempts are still being made to test the proposed models. In fact, a paper in a peer-reviewed scientific journal published earlier this year really made me sit up and pay attention ! But more on that later ...
So, with this in mind, plus the media interest in NASA's current and future Artemis program, I thought that this would be a good time to take another look at where planetary science is today (2026) regarding the Moon's origin. It doesn't get much more basic than that !
So, with this in mind, plus the media interest in NASA's current and future Artemis program, I thought that this would be a good time to take another look at where planetary science is today (2026) regarding the Moon's origin. It doesn't get much more basic than that !
Now, it may surprise you, but in order to set the latest ideas on the origin of the Moon in context we have to go all the way back to George Darwin (the son of Charles Darwin of "Evolution" fame ). In 1878, George Darwin proposed his "fission theory" for how the Moon formed. Simply put, he proposed that the early Earth spun faster than it does today (for various reasons) ... and that the faster spin created a bulge on the planet that eventually broke away and began to orbit the Earth as the Moon.
He also suggested that the Pacific Ocean between Asia and the Americas represents the scar from which the Moon was torn out. It's roughly circular and about the right diameter ... !
However, there are at least two fundamental flaws with George Darwin's hypothesis. First the Earth had to spin so fast that its day would have only lasted ~2-3 hours, and at that speed the entire planet would probably have flown apart ! Even if the Earth could hold together, the resulting Earth-Moon system would have had way more angular momentum than it has today, and scientists cannot say where that extra momentum could have gone. Second, in the 19th century, no-one knew that the Pacific Ocean floor is the product of Plate Tectonics ... or that the ocean floor is younger than ~200 My. Plus, they did not know that the Earth is ~4.5 Ga old, but we do ... and so we know that the Pacific Ocean is way too young for George Darwin's fission theory to work.
Now, fast forward to the late 1960s when NASA astronauts landed on the Moon, grabbed some rock samples ... and brought them back home. Chemical analysis of these rocks showed something hitherto unexpected : compared to the Earth, the Moon's chemistry is strongly depleted in volatile elements (Na, K, Rb), and especially H20. If the Moon really was formed out of the Earth, this depletion calls for a massive vaporisation event. But what and how ?? At the same time, planetary scientists were beginning to make quantitative models (calculations) about what might happen if the early Earth was involved in collisions with other bodies left over from the accretionary phase of Solar System construction. A really big collision could provide just such a vaporisation event.
Building on this, early computer simulations suggested that the Moon could have formed by a massive collision between the early Earth and a Mars-sized impactor (Theia). According to this (the classical) collision model, the impact resulted in a massive ejection of debris that formed a ring around the Earth that rapidly coalesced (accreted) into the Moon (in about 1000 to 10,000 years). In addition, Theia would have contributed ~70% to the mass of the Moon.
In addition to coalescing to form the Moon, the debris ring was envisaged as being zoned (sandwich) with an inner ring of droplets of molten rock (a "magma disk") bordered by an "atmosphere" of vaporised rock. Notice how the circulation within the debris disk is highly localised within its constituent parts. Accordingly, the debris disk is zoned and it remains so : in other words the classical debris disk is NOT well mixed.
According to the classical collisional model, over a period of weeks, the part of the debris disk furthest removed from Earth began to cool and form particles of various sizes that then progressively coalesced (accreted) over several hundred years to form the Moon until all of the debris disk was consumed.
Now, I need to add just a wee bit of physics here. The Roche limit around the Earth is the radius within which coalescence (accretion) of material into a solid body cannot happen because the Earth's gravity would tear apart any large accreted body that lies between the Roche limit and the Earth. Hence the Moon would have formed from debris located beyond the Roche limit, while debris disk material located within the Roche limit would have simply fallen back to Earth.
Now, I need to add just a wee bit of physics here. The Roche limit around the Earth is the radius within which coalescence (accretion) of material into a solid body cannot happen because the Earth's gravity would tear apart any large accreted body that lies between the Roche limit and the Earth. Hence the Moon would have formed from debris located beyond the Roche limit, while debris disk material located within the Roche limit would have simply fallen back to Earth.
Post-Apollo, this remained the consensus model for Moon formation for about 40 years until it became apparent that there were a number of problems with it. First, the classical collision model predicted that the debris disk should form over the Earth's equator. However, the Moon's orbit is tilted with respect to the Earth's equator : and that's a physical problem. Second, and more importantly, planetary scientists were able to analyse the samples brought back from the Moon by the Apollo program to such high degrees of accuracy and precision that it became very obvious that there was a chemical problem with the classical collisional model. The mantles (the stuff between the crust on the outside and the Fe-rich core on the inside) of both the Earth and the Moon are chemically indistinguishable. That's what this specific diagram illustrates : when you plot the chemistry of mantle-derived rocks from Earth and from the Moon, they all lie on the same straight line ... and specifically it illustrates Oxygen isotopes normalised to the composition of a standard (Earth's ocean water). We don't need to get into the details here, but all attempts to explain this chemical identity failed : they either relied on too much coincidence (scientists do not like coincidence as an explanation for anything !), or they were simply over-constrained (too many assumptions had to be made instead of observations and measurements). Because of the difficulty in explaining the isotope data, some scientists referred to it as an "Isotope Crisis". But why is this chemical identity so important ? What's the problem ?
Here's another plot of the oxygen isotope signatures. It's pretty easy to read. The dashed horizontal line is the signature of the Earth (TFL). Samples from the Moon sit on that same line, and therefore are indistinguishable from the terrestrial signature. However, Mars and various families of meteorites (that represent asteroids in our Solar System) lie well away from the Earth/Moon line. Remember that the classical collisional model for the origin of the Moon predicts that ~70% of the mass of the Moon comes from the impactor (Theia), but this slide tells us that planetary bodies throughout the inner Solar System are chemically quite different from one another, so it is extremely unlikely that Theia would have had the same isotopic composition as the Earth (it would be a special case) ... and therein lies the problem : if Theia was chemically different to the Earth, why do we not see a hybrid chemical signature in the Moon rocks ?
So, does this mean the end of the collisional hypothesis ? No !! ... far from it. Gifted astrophysicists later came up with further computer simulations and found a way to form the Moon directly from a collision (or a near miss) with the Earth, but without a debris disk forming at all, and without involving too much impactor material in the resulting Moon. This slide shows how a model impact might extract vaporised rock just from the Earth. The green material (in the diagram !!) is Earth material that separates from the yellow impactor material. Most of the green material is placed into orbit where it forms the Moon (the Dot). If I understand correctly, the impactor material (yellow) and part of the Earth material (green) fall back to Earth because they lie within the Earth's Roche limit, thereby explaining that the chemistries of the Earth and the Moon are indistinguishable because they are both made just from Earth materials. It's a very neat model ... but it's not the only one that might account for the indistinguishable chemistries of the lunar and terrestrial mantles !
Perhaps the most popular model at the present time is the one involving a Synestia. What is a synestia ??? It's a doughnut shaped cloud (a torus) of vaporised rock. Where does the idea come from ... and why ? It's all down to the mantle mixing issue. The best way to mix the chemistries of the Earth and impactor mantles is to vaporise both ... totally !! 100 % !! ... and the simplest way to do that is with a very high velocity impact. The result of such a high velocity impact would look like a doughnut ; in fact a vast homogeneous cloud of vaporised rock with no zoning that can mix without internal barriers. As it cools, condensation in the inner part forms the Earth in the middle and the Moon in the outer part, beyond the Roche limit.
So let's take a closer look at the mantle mixing issue. The top row of the cartoon in this slide represents what happens after a classical (low velocity) impact scenario with a debris disk. The terrestrial and lunar mantles are not throughly mixed with each other (or individually), so their chemistries will be different. The bottom row of the cartoon shows the high velocity synestia scenario with a debris cloud, where the terrestrial and lunar mantles are thoroughly mixed and chemically indistinguishable on all scales ... even if their cores are not proportionally similar.
Until the beginning of this year, I was under the impression that the synestia was the favoured model for formation of the Earth-Moon system involving a collision between proto-Earth and Theia. However (and unbeknownst to me !!), for the past decade there have been "rumblings" in the scientific literature about another model : one that sounds very much like George Darwin's 19th century "fission" model ... but this time on steroids. I first heard of this idea last February when I came across a paper that attributed the origin of the Moon to "explosive fission". At the risk of oversimplifying : the author's rationale was that in order to reduce the spin of the early Earth to prevent it from breaking apart and still derive the Moon by terrestrial fission ... ejection of material to form the Moon was going to need an extra energy boost. So the question became : "where do we get that extra energy?"
But before I can address this question, I need to brief you about evolving current ideas regarding the interface between the Earth's mantle and core. Deep below Africa and the Pacific ocean are two rock masses (here "Tuzo" and "Jason"), informally named after two very famous geophysicists (Tuzo Wilson and Jason Morgan). These rock masses are Large Low Velocity Provinces - or LLVPs - and they are big !!! : 1000s km wide by up to 1000 km thick. Now, a fundamental tenet of geophysics is that as you go deeper into the mantle of a planet the density of rocks increases with the pressure, and as the density of rocks increases so the speed at which seismic (sound) waves travel through rocks also increases. So it came as a great surprise to everyone when the Large Low Velocity Provinces were identified as zones wherein the speed of seismic waves is significantly lower than geologists would expect at such depths. What are they ? Quite simply, we do not know !!! However there are theories ... one of which suggests that they might be leftovers from collisions with the proto-Earth. Could they be pieces of Theia ? If they are, then maybe we could solve the "isotope crisis" if Theia is simply "hiding" at the base of the Earth's mantle. But note the emphasis on "maybe" ! ... we really don't know.
Here's a 3D view of the Large Low Velocity Provinces as we see them today with seismic imaging. So back to the question of "where do we get that extra energy?" required to run the fission model of Moon formation ... without tearing the proto-Earth apart. As recently as 2013, a paper suggested that a nuclear explosion at the core-mantle boundary of the Earth might do the trick. Now, there are such things as natural nuclear reactors in the Earth's crust (Gabon is the classical example), but we know very little about the deep mantle in this regard. However, had such an explosion occurred (though it is far from obvious how !!!), it could have allowed for fission of the Moon with a slower rotating Earth (~6 hour day).
Whatever, the 2026 paper tries to combine the Large Low Velocity Provinces and the notion of an "explosive fission" to form the Moon. According to the author's theory : First ... the precursors ("seeds") of the Large Low Velocity Provinces are assumed to pre-date the Theia collision and to simply be part of the mantle, however that might have occurred. Second, the Large Low Velocity Provinces themselves are assumed to form from these "seeds" by a combination of tidal forces driven by the Sun and heat emanating from the Earth's core, but it's not clear how. Third because heat only escapes from the lowermost mantle very slowly at this time (no plate tectonics), the author assumes that it builds up in the Large Low Velocity Provinces.
At this point in the explanation, the author switches from talking about "heat" to talking in terms of "energy" ... and this is where his argument gets difficult to follow. His idea is that the built-up "energy" goes "critical" ... and that leads to an almighty explosion at the core-mantle boundary that expels Earth material that forms the debris disk that forms the Moon with a chemical signature identical to that of the Earth. It's that last step that I have difficulty following. In my opinion it reads rather like "... and then a miracle happens". I'm really not sure how this paper got through peer-review, but it was included in a scientific journal published by Springer, a reputable international science publishing house, and this is just one of a set of papers over the past decade that seem to be going this "explosive fission" route. So let's just say that I am perplexed ... and leave it at that !
So, what's the take-home message here ? Astronomically speaking, the origin of our nearest neighbour (cousin, if you like) continues to puzzle us. As my grandmother used to say : "There's more than one way to feed a cat sour cream", and it appears that the scientific community agrees that there's more than one way to explain the fact that the chemical signature of the Earth's interior is indistinguishable from that of the Moon. I'm sure that you've all heard the old adage : "All models are wrong, but some are more useful than others" We'll just have to wait and see how these competing models play out.
So, what's the take-home message here ? Astronomically speaking, the origin of our nearest neighbour (cousin, if you like) continues to puzzle us. As my grandmother used to say : "There's more than one way to feed a cat sour cream", and it appears that the scientific community agrees that there's more than one way to explain the fact that the chemical signature of the Earth's interior is indistinguishable from that of the Moon. I'm sure that you've all heard the old adage : "All models are wrong, but some are more useful than others" We'll just have to wait and see how these competing models play out.