Origin of the moon – challenges and prospects

Organiser and discussion leader

Professor David Stevenson FRS, Caltech, USA

• Audio recording (mp3)

Giant impact origin of the moon- forty years on

Dr William Hartmann, Planetary Science Institute, USA

Abstract

Inspirations for the first presentations of the modern giant impact theory of the origin of the moon included discovery of the Orientale impact basin (Hartmann & Kuiper, 1962) and study of the work of Safronov in the later 1960s, as well as failure of the three primary lunar origin theories in vogue during the Apollo era.  This raised questions about the largest impactors to hit Earth during its accretion.  The work was presented by Hartmann and Davis in 1974 (Cornell satellite conference) and 1975 (Icarus paper). Discovery of the equality of lunar and terrestrial O isotope, a few years later, was originally seen as support for the impact hypothesis (although, ironically, it is now cited as a contradiction or “crisis” for the impact hypothesis).  A 1984 conference on lunar origin, held in Hawaii, marked the first wide acceptance of the model, after which numerical models of giant impact processes evolved rapidly. 

The initial 1974/75 work focused on the second-largest body from the terrestrial feeding zone to strike Earth during Earth’s accretion.  We thus suggest that the modern isotope data as constraining the nature of the putative Earth-zone impactor, rather than by assuming a priori that the impactor must have had different isotope ratios, and suggesting a conflict with the hypothesis.  Kortenkamp and Hartmann are investigating the idea of Belbruno and Gott (2005) that such an object (with terrestrial isotope ratios) may have been temporarily trapped at terrestrial Lagrangian point.  (Further work has proposed to NASA Planet. Geol. & Geophys. program in 2013). 

As for recent objections to the giant impact model in terms of lunar water content, it appears that production of OH and H2O on silicate dust grains (during accretion of the moon) should be considered, not to mention likely early scattering of water-rich outer-solar-system material into the terrestrial zone, possibly during lunar accretion. 

REFERENCES 

Belbruno, Edward and J. R. Gott III  2005.  Where did the moon come from?  Astron. J. 129: 1724-1745.

Hartmann, W. K. and G. P. Kuiper 1962.  Concentric Structures Surrounding Lunar Basins.  Comm. Lunar and Planetary Lab., 1, 51-66.

Hartmann, W. K. and D. R. Davis 1974.   Satellite-sized planetesimals, I.A.U. Colloquium 28, “Planetary Satellites,” August 18-21, Cornell University, Abstract.

Hartmann, W. K. and Donald R. Davis 1975.  Satellite-sized planetesimals and lunar origin.  Icarus, 24, 504-515.

• Audio recording (mp3)

How many impacts to form the Moon?

Dr Martin Jutzi, Physics Institute, Center for Space and Habitability, University of Bern, Switzerland

Abstract

The nearly identical isotopic composition of the Moon and the Earth’s mantle suggest a common origin. In recent numerical studies, the formation of a prelunar accretion disc of appropriate chemical composition was demonstrated in the case of a small impact in a fast rotating Earth (Cuk and Stewart, Science 338, 2012) or by involving very large impactors (Canup, Science 338, 2012). An alternative scenario was presented by Reufer et al. (Icarus 221, 2012), who considered hit-and-run collisions. We will present a follow-up study of this work, which includes also impactors with different initial compositions.

As suggested recently, a fraction of the material in the Moon forming disk could have accreted in one of the Trojan points, forming a smaller second moon (Jutzi and Asphaug, Nature 476, 2011). Such a two-Moon configuration could be stable for tens of millions of years after the giant impact (Cuk and Gladman, Icarus 199, 2009). The likely fate of the companion moon would be to collide with the Moon at low (subsonic) speed; a scenario which was suggested to have caused the lunar dichotomy (Jutzi and Asphaug, Nature 476, 2011). We will discuss various aspects and consequences of such Moon-companion moon collisions.

• Audio recording (mp3)

Coupled Thermal-Orbital Evolution of the Early Earth-Moon System

Professor Jack Wisdom, Department of Earth, Atmospheric, and Planetary Sciences, MIT Cambridge, USA

Abstract

The isotopic similarity of the Earth and Moon has motivated a recent investigation of the formation of the Moon with a fast-spinning Earth.

Angular momentum was found to be drained from the system through the evection resonance, a resonance between the Moon and Sun.  However, tidal heating within the Moon was neglected.  Here we explore the coupled thermal-orbital evolution of the early Earth-Moon system, taking account of tidal heating within the Moon.  We find that as the eccentricity rises once the evection resonance is reached, tidal heating within the Moon becomes especially strong.  The large tidal heating in the Moon significantly lowers the tidal Q/k_2 in the Moon, with consequent early escape from the evection resonance and decay of the orbital eccentricity.  Insufficient angular momentum is withdrawn from the system to be consistent with the current configuration of the Earth-Moon system.

• Audio recording (mp3)

Water and volatile elements in the Moon

Professor Francis Albarede, Ecole Normale Supériere de Lyon, France

Abstract

For most of the post-Apollo decades, the lunar interior has been deemed extremely dry. This lack of water can be explained both by the low volatile of the impactor and by the gravitational escape from the lunar gravity field. However, lunar apatite contains enough to support the claim that the lunar mantle holds more water than initially expected. In situ analyses of pyroclastic glass beads also show that melt inclusions in olivine crystals contain hundreds of ppm of water. Evidence of fractionated Zn isotopes, another volatile element, adds to the conundrum. We examine K, Rb, Zn and Ge concentration data on lunar basalts and extrapolate the depletion trend of the to ~300 K, the freezing point of water and show that the water content of the lunar mantle must be in the sub-ppm range. The high water contents in lunar apatites and pyroclastic glasses can be reconciled by the concentrating effects of extreme magmatic differentiation and are also consistent with a dry lunar mantle.

• Audio recording (mp3)

Organiser and discussion leader

Professor Alex Halliday FRS, Vice President (Physical Secretary), the Royal Society

Geophysical constraints on lunar composition and origin

Dr Mark Wieczorek, Institut de Physique du Globe de Paris, France

Abstract

The present day interior structure of the Moon has been elucidated by a number of geophysical measurement techniques. These include data collected from the surface (such as seismic, heat flow, and magnetic field measurements), data collected from orbit (such as topographic, gravimetric and magnetic field measurements), and data collected from Earth-based observatories (such as lunar laser ranging data). Though our current knowledge of the Moon's interior is imperfect, these measurements provide critical constraints on the Moon's bulk composition, initial differentiation, and subsequent thermal evolution. In this presentation, the geophysical approaches that have been used to decipher the Moon's interior structure will be reviewed, the shortcomings of these techniques discussed, and the consequences that these measurements have for lunar origin highlighted. It will be shown that despite problems in interpreting these data, tight constraints exist on the size, composition, and thermal state of the lunar core, the thickness of the Moon's crust, and the bulk silicate abundance of refractory elements. Investigations of the Moon's thermal evolution and magnetic field generation will be shown to offer the possibility of constraining the initial thermal state of the Moon. A future geophysical mission to the Moon's surface would resolve many resolving problems, and a potential mission scenario that is in the realm of a NASA discovery-class or ESA M-class mission will be outlined.

• Audio recording (mp3)

Understanding the origin and evolution of water in the Moon through lunar sample studies

Dr Mahesh Anand, The Open University, UK

Abstract

Several recent lunar sample studies have demonstrated that the lunar interior contains heterogeneously distributed but appreciable quantities of water although estimates for the bulk-water content of the Moon vary over an order of magnitude. In addition, the origin and evolution of this lunar water remain unresolved with a range of possibilities including retention of primordial water during lunar accretion phase to later addition of water after the crystallisation of the putative Lunar Magma Ocean (LMO). Thus far, water (measured as OH) contents (and in some cases hydrogen isotopic composition) of volcanic glasses, apatite, agglutinates, and nominally anhydrous minerals (e.g., plagioclase) have been measured in lunar samples utilising the latest advancements in analytical instrumentation and techniques. In this contribution we review the existing database on water contents and its hydrogen isotopic composition in a variety of lunar samples and evaluate different hypotheses for understanding the history of water in the Moon.  

Mahesh Anand1,2*, Romain Tartèse1, Jessica J Barnes1,2

1Planetary and Space Sciences, The Open University, Milton Keynes, MK7 6AA, UK

2Department of Earth Sciences, The Natural History Museum, London, SW7 5BD, UK

• Audio recording (mp3)

Zinc isotope constraints on lunar formation

Dr James Day, Scripps Institution of Oceanography, USA

Abstract

Zinc isotopes are a powerful tracer of planetary volatiles because fractionation can occur during volatilisation. Recent work has shown that mare basalts have mean d66Zn values of ~1.3‰, and are distinct from carbonaceous chondrites (0-0.5‰), as well as terrestrial (0.28±0.05‰) and martian basalts (0.25±0.03‰) [1,2]. These results suggest volatile depletion of the Moon through evaporation, and are consistent with an impact origin for the Earth-Moon system.

Here we examine why these differences have become manifested using correlations with volatile element abundances and enrichments of the highly siderophile elements (HSE). One explanation requires an episode of melting and evaporation that was subsequently overprinted on Earth by more extensive late accretion than for the Moon. Mass-balance requires ~12% late accretion to account for terrestrial Zn, which is much greater than ≥0.5% estimated from the HSE. This discrepancy can potentially be explained through stochastic accretion of massive impactors combined with partial vaporization and metal-silicate segregation leading to apparent late accretion of moderately volatiles to the HSE of >10:1. Alternatively, oxidation conditions on planetary bodies determine Zn volatility. The mantles of Earth and Mars are more oxidised than the Moon. Combining this evidence with recent work suggesting delivery of chondritic hydrogen into the lunar mantle from late accretion, as well as inheritance from the proto-Earth [3], it is possible that Earth’s mantle was already oxidized prior to the Moon-forming event.

[1] Paniello et al. 2012. Nature, 490, 376-379. [2] Chen et al. 2012. EPSL, 369-370, 34-42 [3] Saal et al. 2013. Science, 340, 1317-1320. 

James M.D. Day1 and Frederic Moynier2

1Geosciences Research Division, Scripps Institution of Oceanography, La Jolla, CA 92093-0244, USA. jmdday@ucsd.edu

2Department of Earth and Planetary Sciences and McDonnell Center for Space Sciences, Washington University, St. Louis, MO 63130, USA. moynier@wustl.edu

• Audio recording (mp3)

Organiser and discussion leader

Professor David Stevenson FRS, Caltech, USA

• Audio recording (mp3)

N-body and SPH Simulations of the Formation of the Terrestrial Planets and Moon

Dr Hidenori Genda, Tokyo Institute of Technology, Japan

Abstract

We have developed the hybrid code that can handle both the long-term orbital evolutions of objects (N-body code) and the short-term collision processes of objects (SPH code). Here, we apply this hybrid code to the last stage of terrestrial planet formation. During this stage, several tens of Mars-sized protoplanets collide each other, that is, giant impacts. Giant impacts are the energetic events, and related to the origin of the large satellite like the Moon. N-body simulation of giant impact stage and SPH simulation of moon-forming impacts have been well investigated independently. Our hybrid code can handle histories of giant impacts such as change in core-mantle ratio, spin axis, spin velocity of protoplanets and so on. From the results of several tens of numerical simulations for giant impact stage, we will statistically discuss formation probability of Moon-like satellite, typical size and composition (also Oxygen isotopic composition) of satellite, and so on.

• Audio recording (mp3)

Modeling the accretion of the Moon from the protolunar disk

Dr Julien Salmon, Southwest Research Institute, USA

Abstract

We have developed a new numerical model to investigate the accretion of the Moon from the protolunar disk. Material inside the Roche limit is represented by a uniform slab of material that spreads viscously, based on the Thompson & Stevenson (1988) disk model. Beyond the Roche limit, the disk is modeled by a collection of individual particles tracked with the N-body code Symba. The latter interact with the Roche-interior disk at the strongest Lindblad resonances. We find that the Moon accretes in 3 consecutive steps: 1) material in the outer disk rapidly accretes and forms a lunar “core” on a timescale of ~1 yr, similar to what found by pure N-body simulations (e.g. Ida et al. 1997, Kokubo et al. 2000); 2) the inner disk, which has been confined by outer objects in phase 1), slowly spreads outward; 3) after ~20 years, material from the inner disk is brought through the Roche limit, forms new moonlets that recoil from the disk and collide with the Moon, prolonging its accretion over ~200 years. The forming Moon scatters and traps into mean motion resonance some of the objects formed at the Roche limit, resulting in an expansion its semi-major axis. For canonical disks, the Moon has a semi-major axis of ~6Re by the end of its formation, compared to an average of 3.8Re found in pure N-body simulations (e.g. Ida et al. 1997, Kokubo et al. 2000). We will present the range of outcomes produced by initial disk profiles inspired from both canonical (e.g. Canup 2008) and non-canonical impacts (Canup 2012, Cuk & Stewart 2012), and discuss what the dynamics of accretion imply regarding the Earth and Moon’s compositional similarities. For non-canonical disks, we will discuss how the Moon’s initial orbital parameters may affect the likelihood of its capture into the evection resonance with the Sun.

• Audio recording (mp3)

Isotopic constraints on physical models

Dr Kaveh Pahlevan, Observatoire de la Côte d'Azur, France

Abstract

Ever since Apollo,  isotopic abundances have been used as tracers to study lunar formation,  in particular,  to study the sources of the lunar material.  In the last decade,  a number of increasingly precise observations have been made that give strong indications that the Moon and the Earth's mantle have a common heritage. To reconcile this observation with the origin of the Moon via the collision of two isotopically distinct bodies,  two ideas have been put forward: (1) it has been proposed that the Earth-Moon system underwent isotopic homogenization into a single reservoir via turbulent mixing during the ~10^2 year molten disk epoch after the giant impact but before lunar accretion and (2) non-canonical impacts have been proposed in which the silicate disk injected into orbit is sourced directly from the mantle of the proto-Earth and the impacting planet in the right proportions to match the isotopic observations.  Simultaneously,  it has been recognized that liquid-vapor separation in the aftermath of the giant impact is capable of evolving small but measurable (mass-dependent) isotopic differences in the Earth-Moon system,  making isotopic measurements sensitive not only to the sources but to the processes involved.  I will discuss how ongoing interaction of theoretical models with isotopic observations can continue to inform our ideas about lunar origin.

• Audio recording (mp3)

Volatile elements in Earth and Moon

Professor Bernard Marty, CRPG-CNRS, Université de Lorraine, France

Abstract

The discovery of indigenous lunar water renewed tremendous interest on the origin of atmophile/hydrophile elements, namely water, carbon, nitrogen, noble gases, in the terrestrial planets.  The Moon is more depleted in moderately volatile elements such as K, Zn... than the Earth, so that retention of water in the lunar interior required specific processes and conditions. The terrestrial H, C, N, noble gas inventory is consistent with contribution of about 2(±1)% carbonaceous chondrite (CC) -type material. Halogens of the bulk Earth requires a comparable contribution of a few percents CC, despite their lower volatility. This implies that the  Earth was dried up for atmophile/hydrophile elements at the end of the giant impact phase , as a result of impact erosion, and was contributed later on by wet material. The timing of this transition is constrained by the ¹²⁹I-²⁴⁴Pu-^129,131-136Xe closure age  of the terrestrial atmosphere of 40-50 Ma (after correction for geological loss of atmospheric xenon). A similar closure age is obtained for the terrestrial mantle, also from I-Pu-Xe systematics. This was probably the time of the Moon-forming event.

• Audio recording (mp3)

Organiser and discussion leader

Professor Alex Halliday FRS, Vice President (Physical Secretary), the Royal Society

Oxygen isotope evolution during accretion of the terrestrial planets

Professor David Rubie, University of Bayreuth, Germany

Abstract

The terrestrial planets and meteorite parent bodies formed with intrinsic oxygen fugacities approximately 5 orders of magnitude higher than that of a solar gas.  A major cause of this was the oxidation of Fe to form a FeO component in silicates.  The most likely oxidant was water.  There is abundant evidence that the oxygen isotopic composition of water was high in 18O/16O and 17O/16O compared with other oxygen reservoirs in the early solar system, implying a link between the oxygen isotope ratios of planetary bodies and their oxidation state.  We have investigated this link in the context of N-body accretion models for planet formation.  We find that the composition of the cores and mantles of the Earth and other terrestrial planets are best reproduced with models in which Fe is significantly oxidized at heliocentric distances greater than 1 to 1.5 AU prior to accretion.  The oxygen isotopic corollary is that the high 18O/16O and 17O/16O ratios of the planets relative to solar is also determined by the FeO component rather than by accretion of H2O ices.  Exchange between Fe-bearing dust and water vapor enriched in the heavy oxygen isotopes prior to planet formation explains both the chemical and oxygen isotopic compositions of the planets. This exchange can be understood as mixing between water produced by photochemical dissociation of CO and oxygen of solar composition.  In addition, in the Grand Tack accretion models, D17O values of all bodies above a certain mass tend to rise together. Consequently the oxygen isotopic composition of the Moon-forming impactor (Theia) could have been similar to that of the proto-Earth.

Co-authors:

E.D. Young (UCLA, Los Angeles, USA)

S.A. Jacobson (OCA, Nice, France)

A. Morbidelli (OCA, Nice, France)

D.P. O’Brien (PSI, Tucson, USA)

• Audio recording (mp3)

Siderophile Element Constraints on the Origin of the Moon

Professor Richard J Walker, University of Maryland, USA

Abstract

The lunar mantle sources of volcanic glasses and basalts were depleted in highly siderophile elements (HSE) relative to the terrestrial mantle by at least a factor of 20. Limited Os isotopic data, however, suggest the HSE are not fractionated from chondritic relative abundances in the lunar mantle. Although there are other possibilities, the most likely explanations for the disparity between the Earth and Moon are either: 1) the Moon received a disproportionately lower share of late accreted materials than Earth, such as may have resulted from stochastic late accretion, or 2) the major phase of late accretion occurred prior to the Moon-forming event, and the putative giant impact led to little drawdown of HSE to the Earth’s core. High precision determination of the 182W isotopic composition of the Moon can help to resolve this issue.

• Audio recording (mp3)

Panel discussion

Professor David Stevenson FRS, Caltech, USA

• Audio recording (mp3)