Source
Bottke, Vokrouhlický, Dykhuis & Zellner (2026), LIT-009 (The Planetary Science Journal 7, art. 171), read in full in NOTE-002 from the arXiv preprint 2606.05036v1. Figures were read through their captions.
What was actually shown
The crater cluster (§2.1). Seven lunar craters of D ≥ 20 km have mean model ages of 760–820 Ma, from Copernicus (93.1 km) down to Al-Khwarizmi (22.5 km). Their ages are N(1) superposed-crater densities scaled to Copernicus = 800 Ma, the one crater of the seven dated by Apollo samples. The scale is checked on two other sample-dated craters, Aristillus and Autolycus.
The source. The Eulalia family is re-identified among overlapping families. Yarkovsky/YORP chronology fits give five solutions with ages T = 785, 855, 965, 920 and 865 Myr. The text says 800–900 Ma; only two of the five fall in that range (NOTE-002, corrections). The parent body broke up just inside the J3:1 resonance, so part of its debris went straight in and more drifted in over 10⁸ years.
The shower, Solution 5 only (§§5–6). Table 1 gives 6,995 D ≥ 2 km and 833 D ≥ 5 km fragments into the J3:1. That is after a WISE incompleteness factor F1 ≈ 2.3, which the authors call “a knob” and set by eye, and before a ≈1.5× collisional correction. Impact probabilities per resonance entrant are 5.4×10⁻³ (Venus), 4.2×10⁻³ (Earth) and 0.6×10⁻³ (Mars), from 61,065 test particles. The Moon is taken as 1/20 of Earth by cross-section, not simulated. The largest expected impactor is about 8 km on Earth and Venus, 4.5 km on Mars and “>3 km” on the Moon. Most impacts fall in about 700–850 Ma.
The comparison (§6.4). In 10,000 trials of seven lunar projectiles, the observed crater sizes “generally fall within the 1σ envelope”. Copernicus “would require a larger projectile” than the average.
What could have come out otherwise. The best solution could have delivered too few multi-km bodies to make the cluster, or delivered them at the wrong time. It did not. Whether the other four solutions would have is not reported: “If Solution 5 cannot produce an adequate impact shower, the results from Solutions 1–4 are unlikely to be relevant” (§6.2).
What this does not say
- That the shower happened. The paper shows that a source exists which could have produced it. No crater of the cluster but Copernicus has a sample age. The only compositional link is Morgan et al.’s 1973 trace elements in Apollo 12 material, which the paper calls “compelling albeit circumstantial”.
- That the cluster is at 800 Ma. It is a cluster in N(1). If Copernicus’s ray-sample age were not its formation age, every date in it would move together.
- A central estimate. Only the most favourable of five solutions is propagated, F1 is tuned by eye, and the size distribution is extrapolated at both ends. NOTE-002 reads the result as an existence argument.
- That three-quarters of the family went into the J3:1. That fraction is a geometric first approximation (§3.1). The simulations report absolute yields, not the fraction.
- Anything settled about Earth, Mars or Venus. Links to the Bitter Springs Anomaly, Snowball Earth, eukaryote diversification, Martian caldera volcanism and Venus resurfacing are flagged by the authors as “necessarily speculative” (§7).
- That such showers leave the dating clock alone. The argument that a family shower raises the large-impactor flux for 10⁸ years while changing small-crater density by only about 6% (App. A.2) is a back-of-envelope estimate.
Connections
- The model leaves out a possible second, lower-(e, I) half of the family, which could raise the flux “potentially by a factor of two” (§3.2.2). The shower could be larger than modelled as well as smaller.
- The template event is the 465.8 Ma L-chondrite breakup, whose source family the paper leaves open. The record holds no reading of it.
- No other THEORY in the record bears on impact chronology.
Note of 2026-10-30: narrowed and promoted
Promoted as a reading of what the model shows (ADR-065). Whether the shower happened, which sample ages from the other six craters would settle, is THEORY-1195, Proposed, which extends this entry.