What Are the Real Constraints on Commencement of the Late Heavy Bombardment?

Introduction: Early Apollo sample ages demonstrated an early intense bombardment by small bodies lasted until ~3.8 Ga, including a possible spike (Late Heavy Bombardment, or LHB) during a short interval just after 4.0 Gyr [1,2], which also affected other terrestrial planets [3,4]. Some data suggest the LHB affected asteroids [5] and Mars [6]. Recent dynamical models seek to explain the LHB [cf. 7,8]. Tera et al. [1,2] assumed that a histogram of radiometric ages directly reflected the flux of big impacts that reset rock ages. The assumption was later questioned [9,10]; it was proposed that a "stonewall" saturation of big impacts would have destroyed the pre-LHB record. Clearly, from rock ages ascribed (by geological inference) to several basins: many basin impacts happened from just ~3.90 Ga to 3.85 Ga but ended sharply with Orientale (3.82 Ga). But it is hotly debated Whether a similarly sharp rise in impact rate initiated the LHB. Before Nectaris (probably 3.90-3.92 Ga), geological evidence is fragmentary. Impacts were so numerous during the LHB, little earlier geology remains for stratigraphy, crater counts, or to link with rock ages. The flux from crustal solidification until the LHB was much lower than during the LHB [argued by Ryder, 11], or perhaps similar or even higher. If lower, then the LHB was truly a "cataclysm"; otherwise, perhaps only a bump or inflection on a generally monotonically declining post-accretionary impact rate. Some arguments favor a minimal pre-LHB flux. (1) A heavy flux would have pulverized and punctured the lunar crust, more than it is. However, this constrains only top-heavy size distributions, not if there were a cut-off near the sizes of the largest observed basins. (2) A heavy early flux would have contaminated the crust with more impactor signature than is found in samples. This suggestive argument is not robust, depending on the very uncertain projectile retention efficiency. (3) Most persuasively, Ryder [11] argues that the absence of old lunar impact melts, which would have been abundantly produced by early basin impacts, means that such impacts were rare; we evaluate this below.