Can planetesimals left over from terrestrial planet formation produce the lunar Late Heavy Bombardment?

The lunar Late Heavy Bombardment (LHB) defines a time between similar to 3.8 to possibly 4.1 Gy ago when the Nectarian and early-Imbrium basins on the Moon with reasonably well-constrained ages were formed. Some have argued that these basins were produced by a terminal cataclysm that caused a spike in the inner Solar System impactor flux during this interval. Others have suggested the basins were formed by the tail end of a monotonically decreasing impactor population originally produced by planet formation processes in the inner Solar System. Here we investigate whether this so-called declining bombardment scenario of the LHB is consistent with constraints provided by planet formation models as well as the inferred ages of Nectaris. Serenitatis, Imbrium, and Orientale. We did this by modeling the collisional and dynamical evolution of the post-planet formation population (PPP) for a range of starting PPP masses. Using a Monte Carlo code, we computed the probability that the aforementioned basins were created at various times after the Moon-forming event approximately 4.54 Ga. Our results indicate that the likelihood that the dechning bombardment scenario produced Nectaris, Serenitatis, Imbrium, and Orientale (or even just Imbrium and Orientale) at any of their predicted ages is extremely low and can be ruled out at the 3 sigma confidence level, regardless of the PPP's starting mass. The reason is that collisional and dynamical evolution quickly depletes the PPP, leaving behind a paucity of large projectiles capable of producing the Moon's youngest basins between 3.8-4.1 Gy ago. If collisions are excluded from our model, we find that the PPP produces numerous South Pole-Aitkenlike basins during the pre-Nectarian period. This is inconsistent with our understanding of lunar topography. Accordingly, out results lead us to Conclude that the terminal cataclysm scenario is the only existing LHB paradigm at present that is both viable from a dynamical modeling perspective and consistent with existing constraints. (c) 2007 Elsevier Inc. All rights reserved.