Read on Twitter
Our new paper is now officially published in its final, beautifully-typeset form! I'm really proud of our work on this, and very excited to share our results. In this thread, I'll walk through what we did, why we did it, what we found, and why it matters. https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2020JE006589
The Moon, like the Earth, is stratified, or divided into several layers: core, mantle, and crust. The core is mostly iron and other heavy elements, while the crust and mantle are made up of rocks. But how did this layering originate?
Early in lunar history, the Moon got super hot. Most, if not all, of the Moon melted. This is known as the Lunar Magma Ocean. As the Moon cooled, minerals began to crystallize. Different minerals formed at different times due to changing melt temperatures, pressures+compositions.
What happens to these minerals? It all depends on density. The first mineral to crystallize from the melt is olivine, a magnesium-rich, green-colored mineral with a high melting temperature. Olivine is denser than the melt and therefore sinks, forming the base of the mantle.
Later in the crystallization sequence, something very different happens. A calcium-rich, grey-colored mineral called plagioclase crystallizes. This mineral is LESS DENSE than the coexisting melt, and therefore FLOATS. This leads to a global plagioclase flotation crust.
Sandwiched between sinking olivine+orthopyroxene and floating plagioclase is the final remaining melt. These dregs are enriched in certain elements that don't easily fit into mineral structures. These elements (called KREEP) are dense, rare, and many give off radioactive heat.
Where does the name KREEP come from? Major components of KREEP include potassium (atomic symbol: K), rare earth elements (REE), and phosphorus (P). The late-stage magma ocean melt dregs are also associated with enhanced iron, thorium, and titanium contents.
The distribution of KREEP is a bit of a mystery. One issue is that this KREEP layer expected to form the uppermost mantle at the end of Lunar Magma Ocean crystallization is MUCH DENSER than the underlying mantle minerals. KREEP probably sinks to the base of the mantle - but when?
The other mystery is that KREEP is almost exclusively sequestered on the lunar nearside, usually associated with magmatic activity. Was KREEP asymmetrically distributed in the magma ocean? Or was it initially distributed globally then brought to the nearside by a later process?
To address these questions, we looked at the farside's lone thorium anomaly, associated with the Moon's largest and oldest impact structure (the >2000 km South Pole-Aitken Basin, or SPA). Thorium abundance maps show a crescent-shaped distribution with two hotspots in NW SPA.
Several prior papers looked at these hotspots, proposing a variety of origin scenarios including antipodal ejecta from nearside basins or exposure of volcanic emplacements within the crust. We wanted to test a different hypothesis: the KREEP dreg layer from the lunar magma ocean.
3D impact models by coauthor Jordan Kendall show that SPA formation would excavate entirely through the crust, ejecting huge amounts of material from the upper mantle. These materials are deposited in a ~30-km-thick blanket concentrated in NW SPA, very similar to the thorium map.
The mantle ejecta blanket is subject to >4 billion years of geologic evolution, including impacts and volcanic resurfacing. Small impacts churn, mix, redistribute, and refresh the ejecta, while very large impacts (such as the 537-km Apollo Basin in NE SPA) can remove it entirely.
In our new, detailed thorium maps, that's exactly what we see. The thorium distribution mirrors the distribution of SPA mantle ejecta reshaped by >4 billion years of geological processes. Volcanic materials and soils mask thorium-bearing ejecta, while recent impacts re-expose it.
So far, the distribution of thorium-bearing materials is consistent with excavation from the lunar magma ocean upper mantle dregs. Is this supported by compositional analyses? Models show that the dregs should contain clinopyroxene minerals and be enriched in Fe, Ti, and K (
).
).
BINGO!
Integrating data from several spacecraft-flown instruments, we see that these critical compositional relationships are indeed observed. The SPA thorium-bearing materials are correlated with enhancements in K, Ti, and Fe, and exhibit a clinopyroxene-bearing mineralogy. ✓
Integrating data from several spacecraft-flown instruments, we see that these critical compositional relationships are indeed observed. The SPA thorium-bearing materials are correlated with enhancements in K, Ti, and Fe, and exhibit a clinopyroxene-bearing mineralogy. ✓
The correlation between clinopyroxene and thorium content is also observed at the local scale, within the two hotspots in NW SPA associated with craters Birkeland and Oresme V. Here, the areas with the highest thorium abundance exhibit the strongest clinopyroxene signatures.
Here are a bunch of spectra and spectral parameters, if you care about that kind of thing. If you don't care about that sort of thing, the main takeaway here is that the spectral properties also support our magma ocean dreg hypothesis. Nice.
So we've solved one mystery. The SPA thorium anomaly is the result of excavated KREEPy lunar magma ocean dregs. Nice! But wait, there's more:
Ejecta isn't the only mantle-related SPA material. SPA formation also MELTED mantle material from greater depths than ejecta. The melt is dominated by low-Ca pyroxenes and is low in thorium. Therefore, SPA provides a cross section of two compositionally-distinct mantle layers!
Quick implications for mantle evolution. #1: Because SPA excavated KREEP-bearing mantle material on the lunar farside, KREEP must have been globally distributed in the lunar magma ocean. If KREEP did end up sequestered on the nearside, it must have happened after SPA formed.
Implication #2: The dense thorium-bearing, KREEPy, Ti- and Fe-enriched magma ocean dregs were present in the uppermost mantle at the time of SPA formation. Either (a) SPA formed before these materials sank to the lower mantle, or (b) at least some of these materials never sank.
Thanks to my awesome coauthors - @Ryan_N_Watkins, @Snorthosite, Jordan Kendall, @alexjayevans, @NickDygert, and @nepetro. It was an incredibly rewarding experience working with this group, and I've learned so much from them about the Moon's many mysteries.
Thanks also to @beckyferreira, who gave this paper an awesome writeup in @VICE when it was first accepted! https://www.vice.com/en/article/v7mzpm/how-the-moons-far-side-got-its-radioactive-spots
OK, that's a wrap. Feel free to reach out if you have any questions or want a personal copy of the paper. If you do have access to @jgrplanets, the paper is available here: https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2020JE006589
