Prof. Dr. Harry Becker

Institut für Geologische Wissenschaften Malteserstrasse 74-100

D-12249 Berlin, Tel.: +49 30-83870668

A1 Summary

Lunar impact basins and their deposits represent the most detailed and best-preserved record of the timing, flux, and composition of late accretion in the inner solar system. However, reconciling interpretations of isotopic ages (suggesting a late heavy bombardment at 3.9 Ga) with the cratering record remains controversial. Thus, applying improved methods and novel approaches to the geochronology of ancient (>3.8 Ga) lunar impact rocks will produce new constraints for models of the lunar accretion history between 4.5 and 3.8 Ga. New Lu-Hf, Sm-Nd, Rb-Sr, and Pb-Pb ages and applying in situ U-Pb dating of zircons in lunar impactites by ion microprobe will provide new insight into the significance of ages of impact events and isochron resetting due to late basin-forming impacts. In particular the U-Pb zircon and Lu-Hf systems are characterized by high closure temperatures, and will likely display a different response to secondary overprints by later impacts compared to the Ar-Ar plagioclase chronometer. The samples we will analyse have siderophile element compositions that may be related to specific impactors, include samples that have yielded ambiguous Ar-Ar dates, or apparently predate the ~3.9 Ga distribution peak that is dominated by Ar-Ar ages. Linking the new data to impact rocks with specific highly siderophile element (HSE) and siderophile volatile element (SVE) compositions (subproject B1) may also place constraints on the flux of volatile rich planetesimals during late accretion. Another objective will be to provide new data to improve the chronology function of the lunar cratering record, in particular for older geologic units that comprise ejecta deposits. The lunar chronology function, established in the early 1980s, provides the basis for ages of other bodies of the inner solar system that were determined by crater counting. Some ages used for the chronology function were obtained on lunar impactites from the Apollo 16 landing site. However, the significance of many of the ages obtained in the 1970s is unclear. Thus, here we will use multiple decay systems to date petrographically well-characterized impact melt rocks and breccias, both simple and complex. By offering updated constraints on the lunar accretion history, this new data will then be used to test different models of post-4.5 Ga mass accretion rates in the inner solar system (A2, A3).

A3 Summary

The lunar cratering record provides valuable information about the late accretion history of the inner solar system. However, our understanding of the origin, rate, and timing of the impacting projectiles is far from complete. To learn more about these projectiles, we can examine the crater size–frequency distributions (CSFDs) on the Moon. For example, by comparing the shape of lunar CSFDs with the size–frequency distributions of potential projectile families like different groups of asteroids, comets, or even ejecta projectiles from the giant impact that formed the Moon we can potentially identify the origin and composition of the impacting projectiles in the inner solar system. In addition to helping identify these projectiles, the shapes of lunar CSFDs are particularly important for determining crater-based relative and absolute ages of surfaces, because CSFDs are used to define the so-called lunar “production function” (PF) from which these ages are calculated. Also, if any changes to the PF are observed over time, this indicates that more than one impactor population may have formed the lunar cratering record.

The potential existence of more than one impactor population has often been used to support a late heavy bombardment (LHB) around 3.9-4.1 Ga that would have significantly influenced the surface development of the terrestrial planets and the distribution of water and other volatiles. However, there is an ongoing debate about whether the shape of the PF has changed at all, and, if it has, at what time it transitioned.

Still, others question the interpretations of different CSFD shapes in general, because geologic processes can subsequently modify existing CSFDs. Thus, to address the questions of whether the PF has changed with time and when the potential transition occurred to produce differently shaped CSFDs, we propose to reinvestigate the key regions previously used for interpreting CSFDs and to improve upon past methods and techniques that were applied for analysing crater data.

Improvements in this area are crucial because currently, crater-based ages are determined using a constant PF; however, if the PF changed with time, current ages might be inaccurate. Thus, to address these issues, we propose to perform detailed geologic mapping and small-scale CSF measurements on high-resolution imagery and topography data from recent lunar missions to validate the suitability of the regions previously used to analyse CSFDs. Furthermore, using new approaches of GIS spatial analyses in combination with high-resolution digital terrain models, we will be able to test the statistical significance of previous CSFD measurements and to investigate more closely the processes modifying the original CSFDs. Overall, this subproject will help identify new constraints on the potential projectile families and their compositions (relevant for subprojects in areas B and C) and produce new constraints on a time-variable PF, which will improve the method for determining crater-based ages (relevant for subprojects A1, 2 and 4).

B1 Summary

Understanding the composition of highly siderophile elements (HSE; platinum group elements, Re, and Au) and siderophile volatile elements (SVE, e.g. S, Se, Te, Ag) in materials from the Earth and Moon will help to constrain the processes that have fractionated elements with different volatility and affinity to metal and sulfide, and thus will improve models of accretion and core formation in the terrestrial planets. In particular, the composition of objects that formed the 4.5-3.8 Ga old lunar impact basins yields key information about the late-accretion history of the inner solar system.

Similar HSE abundance ratios in the silicate Earth and in ancient lunar impact rocks can be explained, if, prior to 3.8 Ga, the Earth’s surface accreted similar materials as the Moon did, and these materials were subsequently mixed into Earth’s mantle (“late veneer”). Some of these element ratios match those in chondritic meteorites, whereas others are not chondritic.

The non-chondritic ratios have been interpreted to indicate that the late veneer comprised a mixture of carbonaceous chondrite–like material and a minor proportion of fragments of planetesimal or embryo cores. If correct, this hypothesis predicts specific ratios and abundances of some SVEs (e. g., S, Se, Te) in the lunar breccias and in the silicate parts of terrestrial planets, which will be tested in the proposed study. Alternatively, the late veneer may be characterized by a unique HSE composition that reflects different processing conditions of dust precursors of planets and asteroids in the early solar nebula, a hypothesis that can be tested by isotopic studies (subprojects B3, B5). Also, while the Moon is thought to be rather depleted in volatile elements, not all SVEs of similar volatility are depleted equally.

Early studies have shown enigmatic fractionations that remain poorly understood. Some of these fractionations may be affected by limits of analytical detection on concentrations. However, others may reflect volatilization and metal–sulphide–silicate partitioning processes. In order to unravel these complexities, we propose to perform comprehensive studies of the HSE and SVE (chalcophile elements such as S, Se, Te, Cu, Ag, but also others such as Cd, Tl, In) composition of ancient lunar breccias and pristine highland rocks from the lunar crust.

However, as a prerequisite for interpreting these data correctly, we must understand the origin of the fractionation of these elements in pristine lunar highland rocks and in the lunar interior. Thus, to evaluate these partitioning processes we will be using new data from subproject C1 and literature data. Results of this project will provide new constraints on the chemical composition of the late veneer, identify the processes that have fractionated HSEs and SVEs in the Moon, and reveal the connection between the timing of late accretion, volatiles, and magma ocean evolution (C2, C4). It will also further our understanding of relations between lunar breccia compositions, specific impact deposits and their ages (subprojects A1-A4).

B2 Summary

In current formation models of the Earth, the budgets of the siderophile volatile elements (SVE) S, Se, and Te and the highly siderophile elements (HSE) in the silicate Earth are assumed to predominantly derive from the late accretion of material that had a broadly chondritic bulk composition. But, some HSEs, most notably Pd, appear to display less siderophile behaviour at the very high temperatures and pressures that prevailed during Earth’s core formation. Thus, a late veneer is not needed to explain the elevated Pd abundances in the Earth’s mantle. However, the abundances of other HSEs (Ir, Os) cannot be accounted for by equilibrium partitioning during core formation, even under very high temperatures and pressures. These contrasting results can be reconciled if the HSE abundances in the Earth’s mantle reflect not only the late veneer but also a residual signature of core formation. This would also be consistent with the elevated Pd/Ir and Pd/Os of the Earth’s mantle, although these non-chondritic ratios may likewise be a distinctive feature of the late-accreted material.

Similar to the HSEs, the budget of the SVEs S, Se, and Te in the silicate Earth are commonly thought to derive almost entirely from the late veneer. This view has been challenged, however, by recent data on the S isotopic composition of ocean ridge basalts derived from the Earth’s mantle. These data suggest that the S isotopic composition in the Earth’s mantle and in chondrites is different, which was interpreted to reflect mass-dependent fractionation of S isotopes during metal–silicate segregation. This difference would imply that about half of the S in the Earth’s mantle might be left over from metal–silicate segregation during core formation, meaning that only the remaining S was delivered by the late veneer. This conclusion, however, contradicts available data on the metal–silicate partitioning of S, Se, and Te and on the chondritic abundance ratios of S, Se, and Te in mantle rocks, which suggest that >90% of the budget of S and essentially the complete budget of Se and Te were delivered by a late veneer.

Therefore, in light of these discrepancies, the objectives of this project are twofold: first, we aim to study the stable isotope composition of S, Te, and Pd in samples derived from the terrestrial mantle, as well as the stable isotope composition of Te and Pd in chondritic bulk rocks. Second, we aim to experimentally calibrate the stable isotope fractionation imparted by metal–silicate equilibration during core formation by performing isotope fractionation experiments at a range of relevant conditions.

Overall, this study will contribute critical data to help resolve the discrepancy between the isotopic and concentration data of S, and it will also help to assess whether the elevated Pd abundances in the Earth’s mantle can or cannot be a residual signature of core formation. As such, by determining how core formation and the late veneer affected the budget of HSEs and SVEs in the Earth’s mantle, this study will help us quantitatively constrain the relative roles of these crucial events in Earth’s history.