On the history of the early meteoritic bombardment of the Moon: was there a terminal lunar cataclysm?

Greg Michael, Alexander Basilevsky, Gerhard Neukum

In Press, Accepted Manuscript,
Available online 4 November 2017


• Statistics of 269 K-Ar datings of lunar highland returned samples are reviewed.
• These are compared against 94 datings of meteoritic highland rocks from 21 localities.
• Cumulative effect of impact gardening on presence of melt at surface is modelled.
• Combined modelling and analysis contradicts idea of a short terminal cataclysm.”

“This work revisits the hypothesis of the so-called ‘lunar terminal cataclysm’ suggested by Tera et al. (1973, 1974) as a strong peak in the meteorite bombardment of the Moon around 3.9 Ga ago. According to the hypothesis, most of the impact craters observed on the lunar highlands formed during this short time period and thus formed the majority of the lunar highland impact breccias and melts. The hypothesis arose from the observation that the ages of highland samples from all the lunar missions are mostly grouped around 3.9–4.0 Ga. Since those missions, however, radiometric dating techniques have progressed and many samples, both old and new, have been re-analyzed. Nevertheless, the debate over whether there was a terminal cataclysm persists. To progress in this problem we summarized results of 269 K-Ar datings (mostly made using the 40Ar-39Ar technique) of highland rocks represented by the Apollo 14, 15, 16, 17 and Luna 20 samples and 94 datings of clasts of the highland rocks from 23 lunar meteorites representing 21 localities on the lunar surface, and considered them jointly with the results of our modelling of the cumulative effect of the impact gardening process on the presence of impact melt of different ages at the near-surface of the Moon.

The considered results of K-Ar dating of the Apollo-Luna samples of lunar highland rocks confirmed a presence of strong peak centered at 3.87 Ga. But since the time when the hypothesis of terminal cataclysm was suggested, it has become clear that this peak could be a result of sampling bias: it is the only prominent feature at the sites with an apparent domination of Imbrium basin ejecta (Apollo 14 and 15) and the age pattern is more complicated for the sites influenced not only by Imbrium ejecta but also that of other basins (Nectaris at the Apollo 16 site and Serenitatis at the Apollo 17 site). Our modeling shows that the cataclysm, if it occurred, should produce a strong peak in the measured age values but we see in the considered histograms and relative probability plots not only the 3.87 Ga peak (due to Imbrium basin), but also several secondary peaks caused by the formation of other basins distributed between 3.87 and 4.25 Ga.

The lunar terminal cataclysm hypothesis is in disagreement with the distribution of K-Ar ages for the highland rocks of the lunar meteorites. The population of lunar meteorites representing localities randomly distributed over the lunar surface, and thus free from the mentioned sampling bias, shows no ∼3.9 Ga peak as it should, if the cataclysm did occur.

We conclude that the statistics of sample ages contradict the terminal cataclysm scenario in the bombardment of the Moon. We also see evidence for the formation of several impact basins between 3.87 and 4.25 Ga which is likewise incompatible with the hypothesis of a short interval cataclysm. There remain other basins, including the largest South Pole – Aitken, the ages of which should be determined in future studies to further clarify the impact history. Sample-return missions targeted to date several key basins need to be planned, and the continued study of lunar meteorites may also bring new details to the general view of the impact bombardment of the Moon.”