THE EARLIEST LUNAR BOMBARDMENT PRODUCED BY MOON-FORMING IMPACT EJECTA.

Introduction. The Moon likely formed in a collision between a large protoplanet and the proto-Earth [e.g., 1,2]. This giant impact (GI) occurred during the late stages of Earth’s accretion; the abundance of highly siderophile elements in Earth’s mantle indicate the Earth only accreted ~0.5% of its mass from broadly chondritic projectiles after this time [e.g., 3]. This makes the GI one of the youngest largest collisions to take place in the terrestrial planet region. Recently, we used this inference to argue that GI ejecta dominated the population of km-sized bodies in the terrestrial planet region during the late stages of planet formation [4]. As evidence, consider that GI simulations, capable of reproducing Earth-Moon system constraints, often eject several percent of an Earth mass out of cis-lunar space [1-2]. If a considerable fraction of this mass were solid debris, as described by many GI simulations, and the GI ejecta size frequency distribution (SFD) had a steep slope, which we infer from modeling work and data [4], km-sized bodies could plausibly have struck main belt asteroids at velocities V > 10 km/s. This is enough to heat and degas target rock; models show such impacts produce ~1,000 times more highly heated material by volume than typical main belt collisions at ~5 km/s [5]. By tracking the temporal evolution of GI ejecta, we predicted a “signature” of the GI was left behind in the Ar-Ar shock degassing ages of asteroid meteorites, and that they show the Moon formed ~4.48 Ga [4]. If GI ejecta blasted the asteroid belt, a large fraction should have also returned to hit the Moon. Here we examined whether the most ancient lunar craters and basins could plausibly come from these projectiles. Dynamical Model of GI Ejecta. To explore the evolution of GI ejecta, we tracked 30,000 test bodies for 600 My using the numerical integrator SWIFTRMVS3. The planets Venus-Neptune were included in the integrations with starting orbits described in [6]. For their initial orbits, the bodies were assigned a random isotropic trajectory away from Earth’s center, were placed along Earth’s Hill sphere, and were given an initial ejection velocity “at infinity” of 1, 3, 5, 7, or 9 km/s, respectively. The results were combined by weighing the outcomes using an initial velocity distribution corresponding to GI hydrocode simulations; 14%, 27%, 26%, 18%, and 15% of the objects were ejected at 1, 3, 5, 7, 9 km/s [7]. Using [8], we estimated that ~1% of our GI test bodies should have struck the Moon ~0.01-400 My after the GI. The timing and impact velocities V of the test bodies are shown in Fig. 1. We find that 30% and 65% hit within 1 and 10 My of the GI, respectively. Their median V was 10 km/s. Velocities increase as the test bodies are perturbed by the terrestrial planets. Collisional Evolution of GI Ejecta. A key uncertainty here is the nature of the GI ejecta SFD. We infer its properties in part from the ancient lunar impact record. The Moon has ~25 Pre-Nectarian (pN) lunar basins made by the impact of D > 20 km diameter projectiles [9]. Assuming 1% lunar accretion, the GI ejecta SFD only had a few thousand D > 20 km bodies. Mass balance therefore requires the majority of GI ejecta to be in a steep SFD dominated by D < 20 km bodies. Tests suggest that ~10 km-sized projectiles were thrown out of cis-lunar space (Fig. 2) [4]. Support for such steep SFDs can be found in nature (e.g., Rheasilvia basin on (4) Vesta produced fragFig. 1. GI ejecta hits the Moon