The lunar surface as a recorder of astrophysical processes

Abstract

The lunar surface has been exposed to the space environment for billions of years and during this time has accumulated records of a wide range of astrophysical phenomena. These include solar wind particles and the cosmogenic products of solar particle events which preserve a record of the past evolution of the Sun, and cosmogenic nuclides produced by high-energy galactic cosmic rays which potentially record the galactic environment of the Solar System through time. The lunar surface may also have accreted material from the local interstellar medium, including supernova ejecta and material from interstellar clouds encountered by the Solar System in the past. Owing to the Moon's relatively low level of geological activity, absence of an atmosphere, and, for much of its history, lack of a magnetic field, the lunar surface is ideally suited to collect these astronomical records. Moreover, the Moon exhibits geological processes able to bury and thus both preserve and ‘time-stamp' these records, although gaining access to them is likely to require a significant scientific infrastructure on the lunar surface.
This article is part of a discussion meeting issue ‘Astronomy from the Moon: the next decades'.

Full text

The lunar surface as a recorder of astrophysical processes
1
Ian A. Crawforda,b, Katherine H. Joyc, Jan H. Pasckertd, Harald Hiesingerd
aDepartment of Earth and Planetary Sciences, Birkbeck College, University of London, Malet
Street, London WC1E 7HX, UK. Orchid ID: 0000-0001-5661-7403
bCentre for Planetary Sciences at UCL/Birkbeck, Gower Street, London WC1E 6BT
cDepartment of Earth and Environmental Sciences, The University of Manchester, Oxford
Road, M13 9PL, Manchester, UK. Orchid ID: 0000-0003-4992-8750
dInstitut für Planetologie, Westfälische Wilhelms-Universität, Wilhelm-Klemm-Str. 10, 48149
Münster, Germany. Orchid IDs: 0000-0002-1338-5638, 0000-0001-7688-1965
Keywords: Moon, lunar exploration, lunar regolith, solar wind, cosmic rays, supernovae
Summary
The lunar surface has been exposed to the space environment for billions of years and during this time has
accumulated records of a wide range of astrophysical phenomena. These include solar wind particles and the
cosmogenic products of solar particle events which preserve a record of the past evolution of the Sun, and
cosmogenic nuclides produced by high-energy galactic cosmic rays which potentially record the galactic
environment of the Solar System through time. The lunar surface may also have accreted material from the local
interstellar medium, including supernova ejecta and material from interstellar clouds encountered by the Solar
System in the past. Owing to the Moon’s relatively low level of geological activity, absence of an atmosphere,
and, for much of its history, lack of a magnetic field, the lunar surface is ideally suited to collect these
astronomical records. Moreover, the Moon exhibits geological processes able to bury and thus both preserve
and ‘time-stamp’ these records, although gaining access to them is likely to require a significant scientific
infrastructure on the lunar surface.
1. Introduction
There are multiple scientific reasons for wishing to resume the robotic and human exploration of the lunar
surface, ranging from lunar geology to astrobiology (for reviews see [1-4]). Other papers in this volume are
mostly concerned with the potential of the lunar surface as a platform for astronomical observations of various
kinds, whereas in this contribution we argue that the lunar surface itself will have recorded much of
astrophysical interest. In this sense, the Moon itself can be viewed as a giant ‘telescope’ which has been
observing and recording astrophysical processes ever since it first developed a solid surface some 4.5 billion
years ago.
1
Accepted manuscript, as published in Philosophical Transactions of the Royal Society, A379: 20190562 (2020):
https://royalsocietypublishing.org/doi/10.1098/rsta.2019.0562

Several factors combine to make the lunar surface an ideal, and perhaps unique, recorder of a wide range of
astrophysical processes throughout Solar System history. Primarily, this is because the lunar surface,
unprotected by an atmosphere or, for much of its history, a magnetic field, has been directly exposed to the
space environment for most of the last 4.5 Gyr. As a consequence, particles and radiation from space have
impacted the lunar surface unimpeded, leaving evidence of their presence in the rocks and soils of the lunar
regolith. Examples include solar wind particles and cosmogenic products of solar energetic particle events, and
thus a record of the past evolution of the Sun, and cosmogenic nuclides produced by galactic cosmic rays, and
therefore a record of the past galactic environment of the Solar System. The lunar surface may also have accreted
material from the local interstellar medium, including supernova ejecta and material from interstellar clouds
encountered by the Solar System in its journey around the Galaxy. Equally as important as the collection of
these astrophysical records, however, are lunar geological processes which facilitate their long-term
preservation. As a relatively low-mass planetary body, whose own internal geological processes largely ceased
billions of years ago, the Moon has preserved an ancient surface with some sampled crustal rocks dating from
4.3-4.4 Gyr [4]. Crucially, and unlike the possibly equally ancient surfaces of some asteroids, for most of its
history the Moon has been sufficiently geologically active to bury, preserve, and ‘time-stamp’ ancient
astrophysical records in near-surface rocks and soils. Key processes include the covering of old surfaces by lava
flows, pyroclastic deposits and impact crater ejecta blankets, and ancient records preserved by these processes
may be recoverable by future space missions [1,5].
2. Lunar records of solar activity
Our knowledge of the past evolution of the Sun comes mainly from theoretical modelling (e.g. [6,7]) and
observations of other solar-type stars having a range of ages (e.g. [8,9]). These studies indicate that, whereas the
overall solar luminosity was probably only ~70% of its present value when the Sun formed, its faster rotation
would have resulted in greatly enhanced magnetic activity and associated solar wind and UV and X-ray
emission. Thus, we expect the luminosity of the Sun to have increased, and the strength of the solar wind and
high-energy photon and particle emission to have decreased, throughout Solar System history. Both of these
effects will have had implications for the habitability of the terrestrial planets and, in particular, for the
environment within which life originated and evolved on Earth [10]. In addition, the decrease in solar magnetic
activity has likely resulted in a corresponding increase in the galactic cosmic ray (GCR) flux in the inner Solar
System owing to the shrinkage of the heliosphere [11], which may also have had consequences for biological
evolution on Earth.
Although generally accepted, the low total luminosity of the Sun in its early history is difficult to understand
given the evidence for liquid water on the surfaces of early Earth and Mars (i.e. the ‘faint-young-Sun’ paradox
[12]). Most proposed explanations invoke enhanced greenhouse gas concentrations in the atmospheres of these
planets, although difficulties remain with these models [13]. One suggested alternative explanation is that the
Sun may have been more massive, and thus more luminous, in the past, but that it lost several percent of its
initial mass in strong solar winds early in its history [14,15]. Observations of limited mass-loss from young solar-
type stars have cast doubt on this proposal [16], but direct measurements of the strength of the solar wind
through time could in principle settle the issue. In any case, it is clear that obtaining direct observational
evidence of solar activity through time would not only provide important insights into the evolution of the Sun
as a star, but also improve our understanding of the evolution of the atmospheres, surface environments and

habitability of the terrestrial planets. Here we argue that the near-surface environment of the Moon has the
potential to provide this valuable information.
The Sun is an emitter of both low and high energy particles which may potentially yield information about solar
processes and evolution. Solar wind particles (electrons and protons, alpha particles, and trace heavy ions with
energies up to ~10 MeV) are emitted constantly from the Sun's corona, varying in intensity with solar activity.
Higher energy (~10 to 103 MeV) particles, often referred to as solar cosmic rays (SCRs) or solar energetic particles
(SEPs), are episodically emitted during solar flares and coronal mass ejection events and are able to produce a
range of cosmogenic nuclides when they impact planetary surfaces (e.g. [17-19]).
Analyses of samples returned by the Apollo and Luna missions have revealed that the lunar regolith is an
efficient collector of solar wind particles and cosmogenic nuclides produced by SCRs [5,17,20], and that it
therefore potentially contains a record of past solar activity (e.g. [21-24]). In this context, determining the time
dependence of both the flux and composition of the solar wind would be of interest. Whereas the overall solar
wind flux is probably the most direct indicator of solar activity, in practice the surfaces of regolith particles can
become saturated with solar wind [25], and the solar wind concentration retained in the regolith is influenced
by each individual grain's exposure history [e.g. 25-29].These factors may make regolith particles insensitive to
recording bulk temporal flux variations. However, there is evidence that changes in solar activity also affect the
relative abundances of ions in the solar wind owing to differential ionisation in the solar wind source regions
[30], potentially making the composition of the solar wind a proxy for solar activity. In addition, early solar
activity, and especially the frequency of coronal mass ejection events, may have been responsible for the
wholesale depletion of moderately volatile elements such as Na and K in the surficial regolith, which may also
detectable in lunar samples [31].
The implantation depths of solar wind ions into regolith particles depend on the irradiation energy, the mass of
the irradiating particles, and the composition (chemistry and mineral lattice structure) of the target material (e.g.
[32,33]). To access these records, research has focused on determination of the light element (H, C, O, N) and
noble gas isotope (He, Ne, Ar, Kr, Xe) budgets of small rock fragments or individual mineral grains (e.g.
[25,34,35]), and analysis of depth-dependent concentrations of these elements within the grains (e.g. [36])
utilising noble gas acid-step leaching techniques (e.g. [37]) and secondary ion microprobe analyses (e.g. [38]).
However, all these efforts are constrained by the nature of the existing lunar sample collection. Of necessity,
samples collected by the Apollo and Luna missions (see [39] for a review) were obtained from the present-day
surface of the Moon and most have had a very long, but generally indeterminate, exposure to the solar wind.
Moreover, unless they have been deeply buried and recently exhumed, surface samples are unlikely to have
sampled the most ancient (i.e. several Gyr-old) solar wind that is of greatest interest in investigating early solar
evolution, although progress may be made by studying solar wind trapped in ancient regolith breccias dating
from that time [40]. Similar considerations apply to inferring past solar activity from cosmogenic nuclides
produced by SCRs (e.g. [17,41]).
A key requirement for further progress would be to obtain independent information on the absolute ages of
both the start and end times of solar wind and SCR exposure for a range of samples exposed to the space
environment at widely different times in the past. Fortunately, just such 'time-stamped' samples likely exist on
the Moon in the form of ancient regoliths (hereinafter 'palaeoregoliths') that were exposed to the solar wind and

SCRs at discrete times in the past and then covered up, and thus preserved, by later geological processes.
Obtaining such samples would greatly help in reconstructing a record of solar activity through time (although
the individual grain exposure histories within sampled palaeoregolith deposits will still need to be considered).
Before leaving this discussion of lunar records of past solar activity, we draw attention to the possibility that the
vertical temperature profile in the uppermost few metres of the lunar regolith may record the history of the
solar irradiance over the last several centuries [42]. Such measurements have been applied to reconstructing
terrestrial surface temperature changes over similar timescales [43], but implementation on the Moon would
yield a measure of solar irradiance variations free of the complexities introduced by Earth's atmosphere and
climate. Building on results from the Apollo heat flow experiments, Miyahara et al. [42] calculated that a
temperature measurement precision of ~0.01°C over a depth range of ~10 m would be able to distinguish
between different models of the total solar irradiance back to the Maunder Minimum in the mid- to late-
seventeenth century. In addition to providing information on very recent solar activity, such measurements
may be helpful in understanding the historical evolution of Earth’s climate system. Obtaining them will require
the drilling of multiple boreholes to ~10 m depth (i.e. five times the depths of the Apollo heat-flow
measurements of ~2 m).
3. Lunar records of galactic processes
As the Solar System has been orbiting the Galaxy for the last 4.6 Gyr it will have experienced a wide range of
different galactic environments. The recent review of galactic rotation constants provided by Vallée [44] implies
that the Sun traverses the entire spiral pattern of the Galaxy every ~720 to 1760 Myr (where the uncertainty
arises from continuing uncertainties in the angular velocity of the spiral arms). As the Galaxy appears to have
four major spiral arms [45], this implies spiral arm passages every ~180 to 440 Myr, during which periods the
Solar System may have experienced a range of interesting astrophysical phenomena including nearby
supernova (SN) explosions and transits through dense interstellar clouds. Reconstructing this history would
provide astronomically valuable information on the structure and evolution of the Galaxy, as well as
astrobiologically important information relevant for understanding the past habitability of our own planet [46-
49]. Previous attempts to find correlations between spiral arm crossings and Earth’s climate and extinction
records have been controversial and ambiguous (e.g. [47]), in part be due to a lack of reliable geological records
of the Solar System’s astrophysical environment.
As reviewed in earlier publications [50,51], the lunar surface is likely to be a much better repository of this
information for the same reasons that it will have preserved a record of ancient solar activity – i.e., it has been
constantly exposed to the space environment throughout Solar System history, while also manifesting
geological processes able to preserve records of this exposure. There are at least three forms that such records
might take:
• Variations in the Galactic Cosmic Ray (GCR) flux, as recorded in the abundances of cosmogenic nuclides
and/or radiation damage preserved in lunar surface rocks and soils
• Direct accretion of interstellar matter and/or supernova ejecta onto the lunar surface
• Variations in the lunar cratering rate driven by gravitational perturbations of the orbits of comets
and/or asteroids by changes in the galactic gravitational environment.

(a) Recording variations in GCR Flux
Several galactic processes affect the GCR flux in the inner Solar System, operating on a range of timescales [52-
54]. On the longest timescales (>1 Gyr), the average GCR flux may reflect the galactic star formation rate, which
could provide useful constraints on models of galactic evolution (although it would be necessary to account for
an expected secular increase in GCR flux reaching the inner Solar System due to decreasing solar activity [11]).
On timescales of the order of a few 100 Myr, the GCR flux is likely to be moderated by an enhanced supernova
(SN) rate, and/or collapses of the heliosphere owing to encounters with dense interstellar clouds, associated
with the Sun passing through galactic spiral arms [55-57]. On still shorter timescales (tens of Myr), additional
variations in the GCR flux may be expected owing to the oscillation of the Sun about the plane of the Galaxy
(with a period of ~64 Myr and amplitude ~70 parsecs [57]), and possible short-term variations in the size of the
heliosphere owing to fluctuations in the local interstellar medium density [11]. Stochastic events, such as nearby
(say, ≤50 parsecs) SN explosions, may be superimposed on these secular and quasi-periodic galactic influences.
For example, Melott et al. [58] have considered the GCR flux at the Earth due to a SN at distance of 50 parsecs
and find that the GCR flux could be elevated by between one and three orders of magnitude above its current
value for several thousand years (see figure 1 in ref. [58]).
There are at least two ways in which evidence for GCR variations might be recovered from exposed Solar System
samples. Firstly, the high energy GCR particles leave tracks of radiation damage in exposed materials, the
density of which is proportional to the GCR flux and exposure duration ([5,22,59]). Secondly, when GCRs
interact with atomic nuclei in geological materials a variety of cosmogenic nuclides (e.g. 3He, 10Be, 21Ne, 36Cl,
38Ar) are produced as a result of spallation and neutron capture reactions (e.g. [60]). Typically, these interactions
occur within the uppermost metre or so of the exposed surface (e.g. [5,61]). Results of searches for GCR
variations based on meteorite samples have proved to be controversial and inconclusive (e.g. [62,63]), in part
because they only record an integrated GCR flux since becoming exposed to the space environment. As Wieler
et al. [62] note “because of the limited sensitivity of the time-integrated GCR signals provided by meteorites, it
is wise to consider ... also the differential GCR flux signals provided by terrestrial sediment samples.” Because
terrestrial samples can be dated independently of the cosmic ray flux, this is a potentially powerful approach,
but is limited by the relatively recent ages of terrestrial sedimentary samples, the complexity of Earth’s
geological and erosional history, and by the fact that the primary GCR flux is attenuated by the Earth’s
atmosphere and magnetic field. It is here that the lunar geological record may be able to help.
Several cosmogenic nuclides have been measured in lunar samples (e.g. [22,29,33,64]), and in principle the GCR
flux could be inferred by measuring the density of cosmic ray tracks and/or the concentrations of cosmogenic
nuclides in exposed lunar materials. In practice there are a number of complications, especially the generally
unknown exposure and shielding histories (i.e. burial depths) of existing lunar samples. These uncertainties
would be mitigated if the start and end times of the exposure of a given lunar sample, together with its burial
history, could be determined independently. This will likely be key to reconstructing GCR records from which
the changing galactic environment of the Solar System might be inferred. As for the solar wind history, the
recovery of GCR records from buried palaeoregolith layers would be one possibility, although unlike the solar
wind case the development of a surficial regolith may not be required to preserve GCR records because they
will also occur at ~ metre depths within crystalline rocks.

We note in passing that a nearby SN explosion would also produce an enhancement in the neutrino flux, which
in principle might be detected by damage tracks produced in mineral lattices by nuclei recoiling from a neutrino
interaction. It has been proposed to search for such signals in terrestrial rock samples [65], but lunar samples
would provide a longer temporal baseline and avoid the neutrino background produced in Earth’s atmosphere.
2
(b) Recording the direct accretion of interstellar matter
In its journey around the Galaxy, the Solar System will have been exposed to a range of different interstellar
medium densities. As reviewed elsewhere ([66] and references cited therein), at present the Solar System
appears to be located close to the boundary of a low density (nH ~ 0.1-0.2 cm-3, where nH is the density of
hydrogen nuclei) interstellar cloud (the ‘Local Interstellar Cloud’, LIC), which is itself located in the even lower
density (nH ~ 0.005 cm-3) and ~100 parsec radius Local Bubble within the Local (‘Orion’) Arm of the Galaxy. As
noted by Cohen et al. [11], even small changes in the extent of the heliosphere caused by the Solar System
moving in and out of low-density clouds like the LIC may have produced variations in the inner Solar System
GCR flux on Myr timescales. On longer timescales, and especially during spiral arm passages, much denser
interstellar environments are likely to be encountered, possibly resulting in the direct accretion of interstellar
gas and dust onto the atmospheres and surfaces of the terrestrial planets [67-70].
In their study of the interaction of the Solar System with interstellar clouds, Yeghikyan & Fahr [68] found that
for interstellar densities of nH ≥1000 cm-3 the size of the heliosphere would shrink to less than one astronomical
unit, leaving the Earth (and the Moon) directly exposed to interstellar material. The frequency with which the
Solar System encounters such dense interstellar clouds is uncertain, with estimates varying between ~300 Myr
and 3 Gyr [68,70-72]. For both astronomical and astrobiological reasons it would be desirable to reduce this
uncertainty, and to determine whether or not encounters with dense interstellar clouds have influenced life on
Earth. Pavlov et al. [67] calculated that if it took 200,000 years to cross a cloud with nH ~1000 cm-3 then ~1 kg m-
2 of interstellar dust would be deposited on exposed planetary surfaces, where it might be identified by
distinctive chemical and isotopic signatures. Given the likely ages and relatively short durations of interstellar
cloud encounters, the ancient and relatively undisturbed surface of the Moon appears far more likely to retain
a record of such events than the dynamic surface of the Earth, especially if they have been preserved within
independently dateable palaeoregolith deposits.
In addition to collecting interstellar dust, it is also possible that airless surfaces such as that of the Moon will
provide a record of the gaseous component of interstellar clouds through which the Solar System has passed.
One possibility would be interstellar pick-up ions, ionized and accelerated within the heliosphere and then
implanted into the surfaces of lunar regolith grains [73]. In addition, now that there is abundant evidence for
volatiles trapped in permanently shadowed regions (PSRs) at the lunar poles [74], where temperatures are
typically of the order of 40 K [75], it may be worth considering whether directly accreted interstellar gas
penetrating the inner heliosphere during interstellar cloud traverses could have become cold-trapped onto PSR
surfaces; this might be a fruitful topic for future investigation.
2
We thank Joe Silk for this observation.

As reviewed elsewhere [51], another component of interstellar material that might be identified on the lunar
surface would be ejecta from nearby SN explosions. There has been a long-standing recognition that SN
occurring within a few tens of parsecs, and perhaps as distant as 100 parsecs, may deposit debris enriched in
radioactive elements within the Solar System (e.g. [76-79), and evidence for two such events, in the age ranges
of ~2 and ~8 Myr, has been reported from 60Fe deposition in ocean sediments [80-82]. Cook et al. [83] argued
that the lunar surface has some advantages as a collector of SN ejecta as the much slower rate of surface re-
working would allow it to accumulate in more concentrated layers than on Earth, and in 2016 this group [84]
identified 60Fe enhancements in Apollo 12, 15 and 16 soil samples (collected from depths ≤few cm) consistent
with the ~2 Myr-old SN event recognized in Earth ocean sediments. However, where the lunar record is likely
to come into its own is in identifying debris from much older SN events than are likely to be preserved by Earth's
dynamic surface environment. It is true that this will be complicated by the short half-lives (T1/2 ≤ a few Myr) of
radioisotopes likely to be present in SN ejecta (see table 1 of Fry et al. [79] for a summary), but two such isotopes,
146Sm (T1/2 = 100 Myr) and 244Pu (T1/2=80 My), are sufficiently long-lived to have recorded one or more spiral arm
passages. In addition, careful analysis of the decay products of once-live isotopes in SN ejecta (e.g. 26Al, 53Mn,
60Fe, 41Ca) might also reveal the signatures of ancient SN events. We also note that Siraj and Loeb [85] have
recently suggested that SN-accelerated dust grains might leave detectable tracks in mineral surfaces exposed at
the lunar surface. Any such detections of SN ejecta would be expected to correlate with evidence for enhanced
GCR (and neutrino) fluxes, so these different lines of evidence for ancient SN events would be mutually
supportive.
Finally, we briefly mention an even more exotic possibility: recently it has been proposed that geological
materials may record interactions with some candidates for dark matter particles [86] and, should the Solar
System have encountered variations in the density of such particles in its orbit around the Galaxy, the long-
lived lunar geological record would appear ideally suited to recording them.
(c) Recording variations in impact cratering rate
The third category of lunar geological records with the potential to provide insights into galactic processes
concerns the changing gravitational environment of the Solar System. It has long been speculated that the
changing gravitational potential as the Solar System oscillates above and below the galactic plane, and passes
through galactic spiral arms, may perturb the orbits of comets in the Oort Cloud and increase the impact
cratering rate in the inner Solar System (e.g. [46,49,55,87-89]). Identifying possible periodicities and/or episodic
spikes in the impact cratering rate, which might then be correlated with the Solar System’s galactic environment,
is problematical primarily because the terrestrial impact record is so sparse [90]. In contrast, the lunar surface
holds an essentially complete impact record for most of Solar System history [91]. It follows that, by obtaining
and dating samples of impact melt from a sufficiently large number (possibly hundreds) of lunar craters, it
ought to be possible to determine unambiguously whether or not temporal variations in the impact flux have
occurred and are correlated with galactic structure (although if most of the impactors were from the asteroid
belt rather than from comets [92] any galactic signal might be muted). Note that we would expect any galactic
periodicity in cratering rate to be correlated with the GCR flux, which is also recorded on the Moon. Obtaining
such a complete impact record for the Moon would also have many other benefits for planetary science,
including refining the inner Solar System impact cratering chronology and constraining models of the
dynamical evolution of planetary orbits (e.g. [4] and references cited therein).

4. Preserving the record
The Moon will only record a history of past astrophysical processes if evidence for them has reached the surface
and then been preserved by lunar geological processes. Before turning to the means of preservation, it is
important to consider whether the lunar surface has always been as open to external influences as it is today.
There are at least two aspects to consider: the ancient lunar magnetic field, and a possible early atmosphere.
Palaeomagnetic studies of Apollo samples suggest that between ~4.2 and ~3.6 Gyr ago the Moon had a core-
generated magnetic field comparably strong to the Earth’s present-day magnetic field (i.e. several tens of µT),
which then declined by at least an order of magnitude prior to 3.2 Gyr ago [93]; it may have persisted at this
reduced level (~5 µT) until about 2.5 Gyr ago, finally ceasing (<0.1 µT) sometime before ~1 Gr ago [94]. As a
consequence, at least during the early part of this time period, the Moon’s surface would have been partially
shielded from the solar wind, so this would need to be taken into account in interpreting the most ancient solar
wind records; possibly samples collected close to palaeomagnetic poles, where the magnetic field lines would
tend to channel charged particles to the surface, would be preferred for such studies. On the other hand, high-
energy GCRs, and uncharged particulate material (e.g. SN ejecta and interstellar dust particles) would not be
expected to be significantly affected by an early lunar magnetic field. The presence of an ancient lunar
atmosphere would potentially impede a wider range of exogenous material from reaching the surface, including
some fraction of the primary GCRs. Given the Moon’s low gravity, its only realistic opportunity to accumulate
an atmosphere would be during periods of intense volcanic activity when the rate of magmatic degassing might
transiently exceed the rate of atmospheric loss. Based on these arguments, Needham and Kring [95] estimated
that a transient lunar atmosphere having a surface pressure of up to ~9 mbar (i.e. 1.5 times higher than the
current atmospheric pressure on Mars) might have persisted for ~70 million years at the peak of mare volcanism.
On the other hand, Wilson et al. [96] have argued that the intervals between individual mare eruptions (~20,000
to 60,000 years) would have been too long for such a transient atmosphere to accumulate. In any case, it appears
that any ancient lunar atmosphere would only have affected the accumulation of astronomical records on the
Moon’s surface for geologically brief periods ~3.5 Gy ago.
Assuming that they reach the lunar surface unimpeded, the various astronomical records discussed in this paper
will only have survived in detectable quantities if geological processes have acted to preserved them. This is
especially true of material deposited directly onto the surficial regolith (e.g. solar wind particles, interstellar
pick-up ions, and accreted interstellar material) because otherwise these records will be disturbed and diluted
by the continuous comminution and overturning (“gardening”) of the regolith by the unremitting impact of
small meteoroids [20]. Owing to the deeper (order metre) penetration depths of GCRs, the record of cosmogenic
nuclide formation, required to reconstruct variations in the GCR flux, is probably less sensitive to surficial
regolith processing, but any record dating back hundreds of millions of years will still need to be protected from
the disturbing effects of larger meteorite impacts. To illustrate this point, consider the solar wind and
cosmogenic nuclei extracted from surface regolith samples collected by the Apollo missions. Solar wind and
cosmogenic noble gases have been extracted from regolith samples collected at all six landing sites (see [33] for
a recent review). However, these regoliths have mostly been developed on surfaces with ages exceeding 3 Gyr
(e.g. [97]) so any evidence for temporal variations within them will have been smeared out by impact-induced
gardening during these vast spans of time. Some exceptions are provided by samples collected on the ejecta
blankets of young impact craters, such as South Ray crater (Apollo 16; estimated age 2 Myr), Cone crater (Apollo

14; 25 Myr), and North Ray crater (Apollo 16; 53 Myr), but these estimated ages [91] are mostly derived from
cosmic ray exposure, so are not independent of the GCR fluxes that we seek to determine.
What we really need is to identify materials (e.g. palaeoregoliths) that were once exposed at the lunar surface
for a known duration and which were subsequently covered by overlying material so that they have been
isolated from the space environment ever since. Fortunately, there are at least three geological processes that
will have acted to cover, and therefore preserve, pre-existing surfaces throughout lunar history:
• Eruption of low-viscosity basaltic lava flows
• Deposition of pyroclastic deposits around sites of explosive volcanism
• Emplacement of impact crater ejecta blankets
There are pros and cons associated with all three preservation mechanisms, which we now discuss.
(a) Lava flows
Basaltic lava flows cover ~17% of the lunar surface, mostly on the nearside, and their generally low viscosity
and apparently laminar flow [98,99] suggests that palaeoregolith layers may be preserved beneath or between
them. Once sampled, basalts can be dated to high accuracy using standard radiometric techniques (e.g. [97]), so
the ages of palaeoregoliths trapped between lava flows could in principle be well constrained [5]. However,
most mare basalts appear to have been erupted within the relatively narrow time interval between about 3.8
and 3.3 Gyr [100] (although older lava flows may be buried by younger ones now exposed at the surface), and
there is no evidence for large-scale basaltic volcanism more recently than ~1 Gyr ago. It follows that
palaeoregoliths dating from the last ~1 Gyr, which more or less encompasses the Solar System’s most recent
traverse of the galactic disk, are unlikely to be preserved between lunar lava flows unless younger, smaller scale,
eruptions have occurred. In this context, it is worth noting that small patches of basaltic lavas, apparently
erupted within the last 100 Myr, may have been identified on orbital imagery [101]. However, although these
localities could potentially preserve valuable astronomical records from an important time interval, such young
lavas are unexpected on the basis of our current understanding of lunar geology and the original interpretation
of some of these features has been questioned [102]. Clearly further work to determine whether or not basaltic
lava flows with ages < 1 Gyr exist on the Moon would be desirable.
Probably the main disadvantage of lunar lava flows as preservers of astronomical records within buried regolith
layers is the heating of the regolith substrate by the overlying lava when it is emplaced. Detailed studies of this
process [103] indicate that the uppermost ~20 cm of regolith covered by a 1 m thick lava flow would likely
experience at least partial degassing of implanted volatiles, with thicker lava flows requiring approximately
proportional thicker regolith to provide adequate insulation. This sets a lower limit to the thickness of
palaeoregoliths able to preserve a good record of solar wind and other implanted volatiles. As regolith
production rates are thought to be in the range 1-5 mm Myr-1 (where the lower value is the estimated present-
day rate and the higher value relates to fresh basaltic surfaces ~3.8 Gyr ago [104,105]), it follows that fresh
surfaces would need to be exposed for tens to hundreds of Myr to accumulate regoliths thick enough to shield
implanted volatiles from an overlying lava flow. These long regolith accumulation times would lead to a loss of
temporal resolution for any astrophysical records they contain. However, it is worth mentioning the recent

suggestion [106] that degassing of some mare lavas may result in the formation of metre-thick fragmented
layers, ‘auto-regoliths’, on their surfaces. If present, these would assist in the preservation of volatiles should
they subsequently be covered by younger lavas. We also reiterate that a thick palaeoregolith is less of a necessity
for the preservation of non-volatile records (e.g. heavy nuclei delivered as SN ejecta and some of the less mobile
cosmogenic nuclei produced by GCRs) as these may be preserved within layered basalts lacking interbedded
regoliths (such as those shown in Figure 1).
3
Figure 1. (a) Mare basalt stratigraphy in Mare Serenitatis exposed in the wall of the 16 km diameter impact crater Bessel (21.8°N,
17.9°E; LROC image M135073175R/NASA/GSFC/ASU). (b) Similar layering exposed in walls of a collapse pit in Mare Ingenii on the
farside (36.0°S, 166.1°E LROC image M184810930L/NASA/GSFC/ASU). (c) Metre-scale basalt layers exposed in the wall of Hadley Rille
(26.1°N, 3.6°E) photographed by Apollo 15 astronaut David Scott using a 500mm focal length lens; the outcrop is about 1300m from
the camera (NASA image AS15-89-12104).
We illustrate the concept of palaeoregolith preservation and dating with the aid of a prominent lava flow on the
surface of Mare Imbrium (Figure 2). Here, a younger lava flow overlies older mare basalts, so we would expect
a palaeoregolith layer to be preserved underneath it. This palaeoregolith layer will contain material reaching
the surface of the Moon in the time period between the emplacement of the under- and over-lying lavas. Ideally,
radiometric dating of returned samples would provide these ages, but as none have yet been obtained from this
region we have used standard Crater Size-Frequency Distribution (CSFD) measurements to model the ages of
these flows (for details see Electronic Supplementary Information, ESM). Based on our CSFD measurements,
the underlying basalts have an absolute model age of
3.30!"."$
%"."&
Gyr, while the over-lying lava flow has an
absolute model age of
3.03!".'(
%".')
Gyr. We therefore hypothesize that sandwiched between these two lava flows
there will be a trapped palaeoregolith layer containing records of the solar wind, SCRs, GCRs, and possibly
other astronomically valuable records, that reached the surface of the Moon within the ~300 Myr period
separating the emplacement of these two lava flows.
3
Although insulation provided by even a thin regolith may be helpful in preserving more volatile GCR
products (e.g. 36Cl, 37,39Ar) and in preventing the annealing of radiation damage tracks (we thank one of our
referees, Gregory Herzog, for this observation).

Figure 2. (a) A prominent lava flow (arrows) on the surface of Mare Imbrium (31.5°N, 338.0°E; Kaguya Terrain Camera/JAXA). Based on
CSFD measurements (see ESM), we obtain an absolute model age of
3.03!".$%
&".$'
Gyr for this lava flow, and an age of
3.30!"."(
&".")
Gyr for
the older lava flows over which it has flowed, so we expect a ~3 Gyr-old palaeoregolith to be trapped between the two (see text). (b)
Schematic illustration of the trapping of a palaeoregolith layer by an overlying lava flow; a paleoregolith such as this would be expected
to contain solar wind, GCR, and other astrophysical records that were implanted during its time on the surface, meanwhile a new
regolith develops on the younger lava flow and captures more recent astrophysical records.
Estimating the extent to which any underlying palaeoregoliths will have been heated by overlying lava flows
such as this, as well as drawing up plans for sampling them, requires an estimate of the lava flow thickness. In
this case, the CSFD of the underlying surface shows that all measured crater sizes have been resurfaced by the
upper lava flow, yielding a minimum flow thickness of ~20m (see ESM). If emplaced as a single lava flow of
this thickness the modelling of Rumpf et al [103] would imply that a palaeoregolith below it would need to be
several metres thick in order to prevent thermal degassing of solar wind and other trapped volatiles. Such a
thick regolith is unlikely to have been generated in the ~300 Myr available given the estimated 1-5 mm/Myr
regolith formation rates. On the other hand, there is evidence, for example layers exposed in the wall of Hadley
Rille (Figure 1(c)) that some mare basaltic lava flows are built up of multiple thinner (~1 m thick) layers, and if
this were the case here it would help reduce the propagation of heat into an underlying regolith [103]; the same
would apply if the underlying lava had developed a fragmental ’auto-regolith’ on eruption [106]. In any case,
as stressed above, non-volatile and some GCR-produced cosmogenic nuclides are much less susceptible to
thermal disturbance and are likely to be preserved even under such a relatively thick lava flow.
We stress that we have here merely used these Imbrium lava flows as an example. The comprehensive mapping
and dating of mare basalts presented by Hiesinger et al. [100] indicates that there are hundreds of large (≥50 km
in size; see figure 17 of [100]) lava flows on the lunar surface spanning the age range ~4.0 – 1.0 Gyr, and doubtless
many thousands of smaller examples such as that discussed here, all with the potential to preserve underlying
palaeoregolith layers with a correspondingly wide range of ages. Moreover, as noted above, stacks of layered
lava flows (Figure 1) have the potential to preserve some astronomically important records (e.g. the GCR flux)
in the absence of interleaved palaeoregolith layers. That said, unless evidence for small-scale basaltic volcanism
within the last ~1 Gyr is confirmed (e.g., [101]), lava flows are unlikely to preserve astronomical records within
this timeframe.

(b) Pyroclastic deposits
In addition to the effusive eruption of low-viscosity mare basalts, lunar magmatic processes have also resulted
in occasional explosive or pyroclastic volcanism [107,108]. The resulting pyroclastic deposits are fine-grained
units of basaltic glass fragments mantling surfaces around volcanic vents (Figure 3), with sizes ranging from
~10 km2 to ~50,000 km2 and thicknesses estimated at several metres [107,109]. McKay [110] has argued that
pyroclastic deposits may be the best preservers of palaeoregoliths owing to their relatively gentle mode of
emplacement. Moreover, because the small (typically tens of microns in diameter) basaltic glass fragments that
make up the deposits would have mostly cooled and solidified before impacting the surface, the buried regolith
would not be subject to the thermal disturbances associated with burial by active lava flows.
Figure 3. (a) Dark pyroclastic materials surrounding a presumed volcanic vent in the Schrӧdinger Basin on the lunar farside (75.3°S,
139.2°E; LROC image/NASA/GSFC/ASU). (b) Schematic illustration of a pyroclastic eruption covering a pre-existing regolith to preserve
a palaeoregolith underneath (adapted with thanks from LPI CLSE Higher Education Resources).
The main disadvantage of pyroclastic deposits in the present context is the ancient, and relatively brief, time
periods in which they formed, dating from the main phase of lunar mare volcanism with an estimated age range
of 3.8 to 3.2 Gyr [100,110]. While doubtless preserving valuable records of the early Sun, palaeoregoliths buried
by currently identified pyroclastic deposits are therefore unlikely to preserve more recent galactic influences on
the Solar System. As in the case for basaltic lava flows, a search for more recent pyroclastic deposits would be
valuable.
(c) Ejecta blankets and impact melt deposits
Impact cratering has been continuous throughout lunar history [91], so palaeoregoliths covered by crater ejecta
blankets and/or impact melt deposits (Figure 4) have the potential to preserve records from more recent times
than those covered by lava flows or pyroclastic deposits. This will be especially important if we seek well-
resolved temporal records for the Solar System’s most recent orbit around the Galaxy. Dating the emplacement
of crater ejecta blankets could be achieved by sampling and radiometrically dating associated impact melt (e.g.
[111-113]. Importantly, such ages would be independent of assumptions regarding the GCR and solar wind
fluxes that we wish to reconstruct from the underlying palaoregoliths. If the pre-existing regolith was developed

on a basaltic lava flow, this too could be sampled and dated, thereby locating the palaeoregolith sandwiched
between lava flow and ejecta blanket in a well-constrained time window.
Figure 4. (a) An unnamed fresh 185 m diameter impact crater in Mare Nubium (20.9°S, 350.3°E; LROC image M183588912R: NASA/GSFC/ASU); note the
prominent ejecta blanket. (b) Schematic illustration of a crater ejecta blanket covering a pre-existing regolith to preserve a an underlying palaeoregolith.
Although impact melt deposits have the potential to preserve palaeoregoliths in the same way as lava flows,
they suffer from the same potential disadvantage of thermally disturbing volatile records. The same may be true
of palaeoregoliths buried by thick ejecta blankets from large craters which may contain significant volumes of
impact-heated materials [114]. On the other hand, the ejecta of small craters is not thought to be at a high
temperature when emplaced, so the main uncertainty in their value as preservers of buried palaeoregoliths is
the degree to which the pre-existing surface is mechanically disturbed in the process. As the ejecta is mostly
emplaced ballistically [115], this is likely to depend on the size of the impact and the distance from it, and may
be quite variable. It is also likely to disturb some records more than others, with very surficial records (e.g. solar
wind, pick-up ions and SN ejecta) being more disturbed than deeper-lying cosmogenic nuclides produced by
GCR interactions. Further work on the mechanical and thermal effects of impact ejecta emplacement would be
desirable.
5. Locating and accessing the record
Gaining access to these astronomical records will present considerable technical challenges. There are two main
aspects: identifying the most promising locations where such records are likely to be preserved, and accessing
and sampling these locations.
(a) Locating the records
We have argued that the astronomical records we seek will be preserved in sub-surface layers, such as buried
palaeoregoliths or lava flows, that were once exposed at the lunar surface. Practical considerations suggest that
an initial search must be based on surface features accessible to remote-sensing techniques that are indicative of
the likely presence of suitable sub-surface deposits. Examples include the geological mapping and dating of

surface lava flows (e.g. [100]), pyroclastic deposits (e.g. [107]), and the ejecta blankets of small Copernican-aged
craters. In addition, high resolution images of the lunar surface have revealed multiple locations where sub-
surface layers out-crop in the walls of rilles, craters, and collapse pits ([99]; Figure 1). Studies of areas where
small impact craters have penetrated overlying materials to reveal sub-surface boundaries may also help
identify suitable locations [116], as would orbital ground penetrating radar measurements [117]. Ultimately, it
will probably be necessary to visit a sub-set of identified localities with robotic or human explorers employing
geophysical techniques, such as ground penetrating radar [118,119] or refraction seismology [120], to confirm
the existence of suitable sub-surface deposits and to assess the practicalities of sampling them.
(b) Accessing the records
An optimal architecture for accessing and sampling sub-surface deposits would provide for the following
capabilities [50,51]:
• Ability to deploy equipment at a wide (preferably global) range of locations on the lunar surface
• Surface mobility, ideally with a range of several tens of km around a given landing site (for example,
this would permit access to the boundaries of lava flows having a wide range of ages; see, e.g., the mare
basalt maps provided by [100])
• Detection of sub-surface palaeoregolith deposits (e.g. using ground penetrating radar or active seismic
profiling [119,120])
• Access and sampling of outcrops on steep slopes such as crater walls or entrances to collapse pits (e.g.
[121])
• Drilling from 10s of metres to perhaps ~100 metre depths (for a review of suitable planetary drilling
technology see [122,123])
• Return of samples to Earth for analysis; the quantity required will depend on the number and types of
sites (e.g. palaeoregolith layers) sampled, but based on the analysis of Shearer et al. [124] we estimate
this to be ~100 kg for each exploration mission or sortie (which would presumably visit multiple
individual localities; compare with the 110 kg returned by the Apollo 17 mission).
Although some of these capabilities might be achievable with suitable designed robotic missions (e.g. [121]), as
argued elsewhere (e.g. [1,2]) large scale exploratory activities such as these would be enhanced by renewed
human operations on Moon (Figure 5). Especially enabling would be the creation of one or more permanently,
or semi-permanently, occupied scientific research stations on the lunar surface, as exemplified by the ‘Moon
village’ concept advocated by the Director General of the European Space Agency [125]. The creation of such an
outpost would offer significant opportunities by providing a scientific infrastructure on the lunar surface, just
as human outposts in Antarctica facilitate research activities across multiple scientific disciplines [126-128].

Figure 5. Artist’s concept of astronauts supervising a lunar drilling system. Such a capability would permit access to the sub-surface, for
example to extract palaeoregolith samples containing ancient solar wind and GCR records, and is an example of how astrophysics,
among other sciences, could benefit from establishing a human-tended research infrastructure on the Moon (image credit: NASA).
6. Conclusion
The Moon is likely to preserve a rich historical record of astrophysical processes relevant to understanding the
evolution of the Sun and its changing galactic environment. In order to access these records, it will be necessary
to collect samples from sub-surface strata that were directly exposed to the space environment at known times
in the past and for known durations. Such geological records undoubtedly exist on the Moon but accessing them
will require a greatly expanded programme of lunar exploration. Ideally, this will include the eventual
establishment of Antarctic-style research stations to support large-scale exploration activities. Such a research
infrastructure would also support a wide range of other scientific activities on the Moon [2,4], including, in the
present context, observational astronomy and astrophysics.
Additional Information
Data accessibility. Additional data can be found in the associated electronic supplementary material.
Authors’ contributions. IAC conceived the paper and drafted most of the manuscript. KHJ provided most of
the expertise relating to solar wind in the lunar regolith and produced the diagrams illustrating palaeoregolith
preservation. JHP and HH performed the CSFD measurements of the Imbrium lava flows. All authors
provided intellectual content and critically reviewed the whole manuscript prior to submission.

Competing interests. We declare that we have no competing interests.
Funding. We acknowledge STFC grant ST/M001253/1 and Royal Society grant UF140190 to KHJ and
Leverhulme Trust grant RPG-2019-222 to KHJ and IAC.
Acknowledgements. We thank our two referees, Gregory Herzog and the other anonymous, for helpful
comments that have improved the manuscript. We thank Joe Silk for drawing our attention to possible
geological recorders of astrophysical sources of neutrinos and dark matter particles. IAC also thanks Ellie
Ganpot for helpful discussions.
References
[1] Spudis PD. 1996 The Once and Future Moon. Washington, DC: Smithsonian Institution Press.
[2] Crawford IA, Anand M, Cockell CS, Falcke H, Green DA, Jaumann R, Wieczorek MA. 2012 Back to the
Moon: the scientific rationale for resuming lunar surface exploration. Planet. Space Sci. 74, 3-14. (DOI:
10.1016/j.pss.2012.06.002)
[3] Crotts, A. 2014 The New Moon: Water, Exploration. And Future Habitation. Cambridge: Cambridge University
Press.
[4] Tartèse R, Anand M, Gattacceca J, Joy KH, Mortimer JI, Pernet-Fisher JF, Russell S, Snape JF, Weiss BP.
2019 Constraining the evolutionary history of the Moon and the inner Solar System: a case for new returned
lunar samples. Space Sci. Rev. 215, 54. (DOI: 10.1007%2Fs11214-019-0622-x)
[5] McKay DS, Heiken GH, Basu A, Blanford G, Simon S, Reedy R, French BM, Papike J. 1991 The lunar
regolith. In The Lunar Sourcebook: A User's Guide to the Moon (eds GH Heiken, D Vaniman, BM French), pp. 285-
356. Cambridge: Cambridge University Press.
[6] Gough DO. 1981 Solar interior structure and luminosity variations. Solar Physics 74, 21-34.
(DOI:10.1007/BF00151270)
[7] Guzik JA, Neuforge-Verheecke C, Young AC Epstein RI, Poulin FM, Schissel JR. 2001 Standard and non-
standard solar models. Solar Physics 200, 305-321. (10.1023/A:1010398627973)
[8] Güdel M. 2007 The sun in time: activity and environment. Living Rev. Solar Physics 4, 3. (DOI: 10.12942/lrsp-
2007-3)
[9] O'Fionnagáin D, Vidotto AA. 2018 The solar wind in time: a change in the behaviour of older winds? Mon.
Not. R. Astr. Soc. 476, 2465-2475. (DOI: 10.1093/mnras/sty394)
[10] Airapetian VS, Glocer A, Gronoff G, Hébrard E, Danchi W. 2016 Prebiotic chemistry and atmospheric
warming of early Earth by an active young Sun. Nat. Geosci. 9, 452-455. (DOI: 10.1038/ngeo2719)
[11] Cohen O, Drake JJ, Kóta J. 2012 The cosmic-ray intensity near the Archean Earth. Astrophys. J. 760, 85. (DOI:
10.1088/0004-637X/760/1/85)

[12] Sagan C, Mullen G. 1972. Earth and Mars: evolution of atmospheres and surface temperatures. Science 177,
52–56. (DOI: 10.1126/science.177.4043.52)
[13] Feulner G. 2012. The faint young Sun problem. Rev. Geophys. 50, RG2006. (DOI: 10.1029/2011RG000375)
[14] Graedel TE, Sackmann I-J, Boothroyd AI. 1991 Early solar mass loss: a potential solution to the weak sun
paradox. Geophys. Res. Lett. 18, 1881-1884. (DOI: 10.1029/91GL02314)
[15] Whitmire DP, Doyle LR, Reynolds RT, Matese JJ. 1995 A slightly more massive young Sun as an explanation
for warm temperatures on early Mars. J. Geophys. Res. 100, 5457-5464. (DOI: 10.1029/94JE03080)
[16] Gaidos EJ, Güdel M, Blake GA. 2000 The Faint Young Sun Paradox: an observational test of an alternative
solar model. Geophys. Res. Lett. 27, 501-503. (DOI: 10.1029/1999GL010740)
[17] Reedy RC. 1980 Lunar radionuclide records of average solar-cosmic-ray fluxes over the last ten million
years. In The Ancient Sun: Fossil Record in the Earth, Moon and Meteorites (eds RO Pepin, JA Eddy, RB Merrill),
pp. 365-386. New York: Pergamon Press.
[18] Reames DV. 2013. The two sources of solar energetic particles. Space Sci. Rev. 175, 53-92. (DOI:
10.1007/s11214-013-9958-9)
[19] Schmelz JT, Reames DV, Von Steiger R, Basu S. 2012. Composition of the solar corona, solar wind, and
solar energetic particles. Astrophy. J. 755, 33. (DOI: 10.1088/0004-637X/755/1/33)
[20] Lucey PG et al. 2006 Understanding the lunar surface and space-Moon interaction. Rev. Min. Geochem. 60,
82-219. (DOI: 10.2138/rmg.2006.60.2)
[21] Kerridge JF. 1975 Solar nitrogen: evidence for a secular change in the ratio of nitrogen-15 to nitrogen-14.
Science 188, 162-164. (DOI: 10.1126/science.188.4184.162)
[22] Crozaz G, Walker R M, Zinner E, Morrison DA, Poupeau G. 1977 The record of solar and galactic radiations
in the ancient lunar regolith and their implications for the early history of the Sun and Moon. Phil. Trans. R. Soc.
A285, 587-592. (DOI: 10.1098/rsta.1977.0103)
[23] Wieler R, Kehm K, Meshik AP, Hohenberg CM. 1996 Secular changes in the xenon and krypton abundances
in the solar wind recorded in single lunar grains. Nature 384, 46-49. (DOI: 10.1038/384046a0)
[24] Wieler R. 2016 Do lunar and meteoritic archives record temporal variations in the composition of solar
wind noble gases and nitrogen? A reassessment in the light of Genesis data. Chemie der Erde 76, 463-480. (DOI:
10.1016/j.chemer.2016.06.001)
[25] Pepin RO, Becker RH, Schlutter DJ. 1999 Irradiation records in regolith materials. I: Isotopic compositions
of solar-wind neon and argon in single lunar mineral grains. Geochim. Cosmochim. Acta 63, 2145-2162. (DOI:
10.1016/S0016-7037(99)00002-2)
[26] Nishiizumi K, Arnold R, Klein J, Middleton R, Goswami JN. 1985 Cosmogenic nuclides in 15011 rocklets.
In 16th Lunar and Planetary Science Conf., Houston, Texas, March 1985, pp. 620-621.
(http://articles.adsabs.harvard.edu//full/1985LPI....16..620N/0000621.000.html)
[27] Jull AJT, Pillinger CT. 1977 Effects of sputtering on solar wind element accumulation. Proc. Lunar Planet.
Sci. Conf. 8, 3817-3833.
[28] Housley RM 1980. Toward a model of grain surface exposure in planetary regoliths. In The Ancient Sun:
Fossil Record in the Earth, Moon and Meteorites (eds RO Pepin, JA Eddy, RB Merrill), pp. 401-410. New York:
Pergamon Press.
[29] Levine J, Renne PR, Muller RA. 2007 Solar and cosmogenic argon in dated lunar impact spherules. Geochim.
Cosmochim. Acta 71, 1624-1635. (DOI: 10.1016/j.gca.2006.11.034)
[30] Baker D, Brooks DH, van Driel-Gesztelyi L, James AW, Démoulin P, Long DM, Warren HP, Williams DR.
2018 Coronal elemental abundances in solar emerging flux regions. Astrophys. J. 856, 71. (DOI: 10.3847/1538-
4357/aaadb0)

[31] Saxena P, Killen RM, Airapetian V, Petro NE, Curran NM, Mandell AM. 2019 Was the Sun a Slow Rotator?
Sodium and Potassium Constraints from the Lunar Regolith. Astrophys. J. 876, L16. (DOI: 10.3847/2041-
8213/ab18fb)
[32] Wieler R. 1998 The solar noble gas record in lunar samples and meteorites. Space Sci. Rev. 85, 303-314. (DOI:
10.1023/A:1005166904225)
[33] Curran NM, Nottingham M, Alexander L, Crawford IA, Füri E, Joy, KH. 2020. A database of noble gases
in lunar samples in preparation for mass spectrometry on the Moon. Planet. Space Sci. 182, 104823. (DOI:
10.1016/j.pss.2019.104823)
[34] Pillinger, C. T. Solar-wind exposure effects in lunar soil. Rep. Prog. Phys. 42, 897–968 (1979)
[35] Palma RL, Becker RH, Pepin FO, Schlutter DJ. 2002 Irradiation records in regolith materials II: solar wind
and solar energetic particle components in helium, neon, and argon extracted from single lunar mineral grains
and from the Kapoeta howardite by stepwise pulse heating. Geochim. Cosmochim. Acta 66, 2929-2958. (DOI:
10.1016/S0016-7037(02)00853-0)
[36] Grimberg A, Baur H, Bochsler P, Bühler F, Burnett DS, Hays CC, Heber VS, Jurewicz AJG, Wieler R. 2006
Solar wind neon from Genesis: implications for the lunar noble gas record. Science 314, 1133-1135. (DOI:
10.1126/science.1133568)
[37] Wieler R, Baur H, Signer P. 1986. Noble gases from solar energetic particles revealed by closed system
stepwise etching of lunar soil minerals. Geochim. Cosmochim Acta 50, 1997-2017. (DOI: 10.1016/0016-
7037(86)90255-3)
[38] Hashizume K, Chaussidon M, Marty B, Robert F. 2000 Solar wind record on the Moon: deciphering
presolar from planetary nitrogen. Science 290 1142-1145. (DOI: 10.1126/science.290.5494.1142)
[39] Heiken GH, Vaniman D, French BM eds. The Lunar Sourcebook: A User's Guide to the Moon. Cambridge:
Cambridge University Press.
[40] Fagan AL, Joy KH, Bogard DD, Kring DA. 2014 Ages of globally distributed lunar paleoregoliths and soils
from 3.9 Ga to the present. Earth Moon Planets 112, 59-71. (DOI: 10.1007/s11038-014-9437-7)
[41] Kovaltsov GA, Usoskin IG. 2014 Occurrence probability of large solar energetic particle events:
assessment from data on cosmogenic radionuclides in lunar rocks. Solar Phys. 289, 211–220. (DOI:
10.1007/s11207-013-0333-5)
[42] Miyahara H, Wen G, Cahalan RF, Ohmura A. 2008 Deriving historical total solar irradiance from lunar
borehole temperatures. Geophys. Res. Lett. 35, L02716. (DOI: 10.1029/2007GL032171)
[43] Pollack HN, Huang S. 2000 Climate reconstruction from subsurface temperatures. Ann. Rev. Earth Planet
Sci. 28, 339-365. (DOI: 10.1146/annurev.earth.28.1.339)
[44] Vallée JP. 2017 Recent advances in the determination of some Galactic constants in the Milky Way.
Astrophys. Space Sci. 262, 79. (DOI: 10.1007/s10509-017-3058-3)
[45] Vallée JP. 2017 A guided map to the spiral arms in the galactic disk of the Milky Way, Astr. Rev. 13, 113-
146. (DOI: 10.1080/21672857.2017.1379459)
[46] Raup DM, Sepkoski JJ. 1984 Periodicity of extinctions in the geologic past. Proc. Natl. Acad. Sci. USA 81,
801-805. (DOI: 10.1073/pnas.81.3.801)
[47] Overholt AC, Melott AL, Pohl M. 2009 Testing the link between terrestrial climate change and
galactic spiral arm transit. Astrophys. J . 705, L101–L103. (DOI: 10.1088/0004-637X/705/2/L101)
[48] Filipovic MD, Horner J, Crawford EJ, Tothill NFH, White GL. 2013 Mass Extinction and the Structure of
the Milky Way. Serb. Astr. J. 187, 43-52. (DOI: 10.2298/SAJ130819005F)
[49] Gillman MP, Erenler HE, Sutton PJ. 2019 Mapping the location of terrestrial impacts and extinctions onto
the spiral arm structure of the Milky Way. Internat. J. Astrobiol. 18, 323-328 (DOI: 10.1017/S1473550418000125)

[50] Crawford IA, Fagents SA, Joy KH, Rumpf ME. 2010 Lunar palaeoregolith deposits as recorders of the
galactic environment of the Solar System and implications for astrobiology. Earth Moon Planets 107, 75-85.
(DOI: 10.1007/s11038-010-9358-z)
[51] Crawford IA. 2016 The Moon as a recorder of nearby supernovae. In Handbook of Supernovae (eds AW
Alsabti, P Murdin), pp. 2507-2522. Springer. (DOI: 10.1007/978-3-319-21846-5_115)
[52] Forman MA, Schaeffer OA. 1979 Cosmic ray intensity over long time scales. Rev. Geophys. Space Phys. 17,
552-560. (DOI: 10.1029/RG017i004p00552)
[53] Shaviv NJ. 2006 Long-term variations in the galactic environment of the Sun. In Solar Journey: The
Significance of our Galactic Environment for the Heliosphere (ed PC Frisch), pp. 99-131. Dordrecht, The
Netherlands: Springer.
[54] Scherer K., et al. 2006 Interstellar-terrestrial relations: variable cosmic environments, the dynamic
heliosphere, and their imprints on terrestrial archives and climate. Space Sci. Rev. 127, 327–465. (DOI:
10.1007/s11214-006-9126-6)
[55] Leitch EM, Vasisht G. 1998 Mass extinctions and the sun’s encounters with spiral arms. New Astron. 3, 51-
56 (DOI: 10.1016/S1384-1076(97)00044-4)
[56] Shaviv NJ. 2003 The spiral structure of the Milky Way, cosmic rays, and ice age epochs on Earth. New
Astron. 8, 39-77. (DOI: 10.1016/S1384-1076(02)00193-8)
[57] Gies DR, Helsel JW. 2005 Ice age epochs and the Sun’s path through the Galaxy. Astrophys. J. 626, 844-848.
(DOI: 10.1086/430250)
[58] Melott AL, Thomas BC, Kachelrieß M, Semikoz DV, Overholt AC. 2017 A supernova at 50 pc: effects on
the Earth's atmosphere and biota. Astrophys. J. 840, 105. (DOI: 10.3847/1538-4357/aa6c57)
[59] Crozaz G. Solar flare and galactic cosmic ray tracks in lunar samples and meteorite: what they tell us
about the ancient Sun. In The Ancient Sun: Fossil Record in the Earth, Moon and Meteorites (eds RO Pepin, JA
Eddy, RB Merrill), pp. 331-346. New York: Pergamon Press.
[60] Wieler R. 2002 Cosmic-ray-produced noble gases in meteorites. Rev. Mineral. Geochem. 47, 125-170. (DOI:
10.2138/rmg.2002.47.5)
[61] Leya I, Neumann S, Wieler R, Michel R. 2001 The production of cosmogenic nuclides by GCR-particles for
2 pi exposure geometries. Meteorit. Plan. Sci. 36, 1547-1561. (DOI: 10.1111/j.1945-5100.2001.tb01845.x)
[62] Wieler R, Beer J, Leya I. 2013 The galactic cosmic ray intensity over the past 106 – 109 years as recorded by
cosmogenic nuclides in meteorites and terrestrial samples. Space Sci. Rev. 176, 351-363. (DOI: 10.1007/s11214-
011-9769-9)
[63] Smith T, Cook D.L, Merchel S, Pavetich S, Rugel G, Scharf A, Leya I. 2019. The constancy of galactic
cosmic rays as recorded by cosmogenic nuclides in iron meteorites. Meteoritic. Planet. Sci., 54, 2951-2976. (DOI:
10.1111/maps.13417)
[64] Lorenzetti S, Busemann H, Eugster O. 2005 Regolith history of lunar meteorites. Meteoritic. Planet. Sci., 40,
315-327. (DOI: 10.1111/j.1945-5100.2005.tb00383.x)
[65] Baum S, Edwards TDP, Kavanagh BJ, Stengel P, Drukier AK, Freese K, Górski M, Weniger C. 2020.
Paleodetectors for galactic supernova neutrinos. Phys. Rev. D. 101, 103017. (DOI:
10.1103/PhysRevD.101.103017)
[66] Crawford IA. 2011 Project Icarus: a review of local interstellar medium properties of relevance for space
missions to the nearest stars. Acta Astronaut. 68, 691-699. (DOI: 10.1016/j.actaastro.2010.10.016)
[67] Pavlov AA, Toon OB, Pavlov AK, Bally J, Pollard D. 2005 Passing through a giant molecular cloud:
‘‘snowball’’ glaciations produced by interstellar dust. Geophys. Res. Lett. 32, L03705. (DOI:
10.1029/2004GL021890).

[68] Yeghikyan A, Fahr H. 2006 Accretion of interstellar material into the heliosphere and onto Earth. In Solar
Journey: the significance of our galactic environment for the heliosphere and Earth (ed PC Frisch), pp. 317-348.
Dordrecht: Springer.
[69] Yabushita S, Allen AJ. 1989 On the effect of accreted interstellar matter on the terrestrial environment.
Mon. Not. R. Astr. Soc. 238, 1465-1478. (DOI: 10.1093/mnras/238.4.1465)
[70] Kokaia G, Davies MB. 2019 Stellar encounters with giant molecular clouds. Mon. Not. R. Astr. Soc. 489,
5165-5180. (DOI: 10.1093/mnras/stz813)
[71] Talbot RJ, Nreman MJ. 1977 Encounters between stars and dense interstellar clouds. Astrophys. J. Suppl. 34,
295-308. (DOI: 10.1086/190452)
[72] Smith DS, Scalo JM. 2009 Habitable zones exposed: astrosphere collapse frequency as a function of stellar
mass. Astrobiology 9, 673-681. (DOI: 10.1089/ast.2009.0337)
[73] Wimmer-Schweingruber RF, Bochsler P. 2001 Lunar soils: a long-term archive for the galactic
environment of the heliosphere? AIP Conf. Proc. 598, 399-404. (DOI: 10.1063/1.1434029)
[74] Colaprete A, et al. 2010 Detection of water in the LCROSS ejecta plume. Science 330, 463-468. (DOI:
10.1126/science.1186986)
[75] Paige DA, et al. 2010 Diviner Lunar Radiometer observations of cold traps in the Moon’s south polar
region. Science 330, 479-482. (DOI: 10.1126/science.1187726)
[76] Knie K, Korschinek G, Faestermann T, Wallner C, Scholten J, Hillebrandt W. 1999 Indication for
supernova produced 60Fe activity on Earth. Phys. Rev. Lett. 83,18-21. (DOI: 10.1103/PhysRevLett.83.18)
[77] Fields BD, Hochmuth KA, Ellis J. 2005 Deep-ocean crusts as telescopes: using live radioisotopes to probe
supernova nucleosynthesis. Astrophys. J. 621, 902-907. (DOI: 10.1086/427797)
[78] Athanassiadou T, Fields BD. 2011 Penetration of nearby supernova dust in the inner Solar System. New
Astron. 16, 229-241. (DOI: 10.1016/j.newast.2010.09.007)
[79] Fry BJ, Fields BD, Ellis JR. 2015 Astrophysical shrapnel: Discriminating among near-Earth stellar
explosion sources of live radioactive isotopes. Astrophys. J. 800, 71. (DOI: 10.1088/0004-637X/800/1/71)
[80] Knie K, Korschinek G, Faestermann T, Dorfi EA, Rugel G, Wallner A. 2004 60Fe anomaly in a deep-sea
manganese crust and implications for a nearby supernova source. Phys. Rev. Lett. 93,171103. (DOI:
[81] Wallner A et al. 2016 Recent near-Earth supernovae probed by global deposition of interstellar radioactive
60Fe. Nature 532, 69-72. (DOI: 10.1038/nature17196)
[82] Fields, B et al. 2019 Near-Earth supernova explosions: evidence, implications, and opportunities. Decadal
Survey on Astronomy and Astrophysics, science white papers, no. 410, Bull. Am. Astr. Soc. 51(3), 410
(https://arxiv.org/abs/1903.04589)
[83] Cook DL et al. 2009 60Fe, 10Be, and 26Al in Lunar Cores 12025/8 and 60006/7: search for a nearby
supernova. In 40th Lunar and Planetary Science Conf., Woodlands, Texas, 23–27 March 2009,
abstract no. 1129. (https://www.lpi.usra.edu/meetings/lpsc2009/pdf/1129.pdf)
[84] Fimiani L, Cook DL, Faestermann T, Gómez-Guzmán JM, Hain K, Herzog G, Knie K, Korschinek G,
Ludwig P, Park J, Reedy RC, Rugel G. 2016 Interstellar 60Fe on the surface of the Moon. Phys. Rev. Lett. 116,
151104. (DOI: 10.1103/PhysRevLett.116.151104)
[85] Siraj A, Loeb A. 2020 Impacts of dust grains accelerated by supernovae on the Moon. Astrophys. J. 895,
L42. (DOI: 10.3847/2041-8213/ab93d5)
[86] Drukier AK, Baum S, Freese K, Górski M, Stengel P. 2019 Paleo-detectors: searching for dark matter with
ancient minerals. Phys. Rev. D. 99, 043014. (DOI: 10.1103/PhysRevD.99.043014).
[87] Torbett MV. 1986 Dynamical influence of galactic tides and molecular clouds on the Oort cloud of comets.
In: The Galaxy and the Solar System (eds. R. Smoluchowski, J.N. Bahcall, M.S. Matthews), pp 147-172. Tucson:
Univ. Arizona Press.

[88] Matese JJ, Whitman PG, Innanen KA, Valtonen MJ. 1995 Periodic modulation of the Oort cloud comet flux
by the adiabatically changing galactic tide. Icarus 116, 255-268. (DOI: 10.1006/icar.1995.1124)
[89] Rampino, MR. 2015 Disc dark matter in the Galaxy and potential cycles of extraterrestrial impacts, mass
extinctions and geological events. Mon. Not. R. Astr. Soc. 448, 1816-1820. (DOI: 10.1093/mnras/stu2708)
[90] Schmieder M, Kring DA. 2020 Earth's impact events through geologic time: a list of recommended ages
for terrestrial impact structures and deposits. Astrobiology 20, 91-141. (DOI: 10.1089/ast.2019.2085)
[91] Stöffler D, Ryder G, Ivanov BA, Artemieva NA, Cintala MA, Grieve RAF. 2006 Cratering history and
lunar chronology. Rev. Min. Geochem. 60, 519-596. (DOI: 10.2138/rmg.2006.60.05).
[92] Neukum, G., Ivanov, B.A., Hartmann, W.K., 2001. Cratering records in the inner solar system in relation
to the lunar reference system. Space Sci. Rev. 96, 55–86. (DOI: 10.1023/A:1011989004263)
[93] Weiss BP, Tikoo SM. 2014. The lunar dynamo. Science 346, 1246753 (DOI: 10.1126/science.1246753)
[94] Mighani S, Wang H, Shuster DL, Borlina CS, Nichols CIO, Weiss BP. 2020 The end of the lunar dynamo.
Sci. Adv. 6, eaax0883. (DOI: 10.1126/sciadv.aax0883)
[95] Needham DH, Kring DA. 2017 Lunar volcanism produced a transient atmosphere around the ancient
Moon. Earth Planet. Sci. Lett. 478, 175-178. (DOI: 10.1016/j.epsl.2017.09.002)
[96] Wilson L, Head JW, Deutsch AN. 2019 Volcanically-induced transient atmospheres on the Moon:
assessment of duration and significance. In 50th Lunar and Planetary Science Conf., Woodlands, Texas, 18–22
March 2019, abstract no. 1343. (https://www.hou.usra.edu/meetings/lpsc2019/pdf/1343.pdf)
[97] Snape JF.; Nemchin AA, Whitehouse MJ, Merle RE, Hopkinson T, Anand M. 2019 The timing of basaltic
volcanism at the Apollo landing sites. Geochim. Cosmochim. Acta 266, 29-53. (DOI: 10.1016/j.gca.2019.07.042)
[98] Taylor GJ, Warren P, Ryder G, Delano J, Pieters C, Lofgren G. 1991 Lunar Rocks. In The Lunar Sourcebook:
A User's Guide to the Moon (eds GH Heiken, D Vaniman, BM French), pp. 183-284. Cambridge: Cambridge
University Press.
[99] Robinson MS, Ashley JW, Boyd AK, Wagner RV, Speyerer E.J., Ray Hawke B, Hiesinger H, van der Bogert
CH. 2012 Confirmation of sublunarean voids and thin layering in mare deposits. Planet. Space Sci. 69,18-27.
(DOI: 10.1016/j.pss.2012.05.008)
[100] Hiesinger H, Head JW, Wolf U, Jaumann R., Neukum G. 2011 Ages and stratigraphy of lunar mare
basalts: a synthesis. Geol. Soc. Am. Special Paper No. 477. (DOI: doi.org/10.1130/2011.2477(01))
[101] Braden SE, Stopar JD, Robinson MS, Lawrence SJ, van der Bogert CH, Hiesinger H. 2014 Evidence for
basaltic volcanism on the Moon within the past 100 million years. Nat. Geosci. 7, 787–791. (DOI:
10.1038/ngeo2252)
[102] Qiao L, Head J, Wilson L, Long X, Kreslavsky M, Dufek J. Ina pit crater on the Moon: Extrusion of
waning-stage lava lake magmatic foam results in extremely young crater retention ages.
Geology 45, 455-458. (DOI: 10.1130/G38594.1)
[103] Rumpf ME, Fagents SA, Crawford IA, Joy KH. 2013 Numerical modeling of lava-regolith heat transfer
on the Moon and implications for the preservation of implanted volatiles. J. Geophy. Res. (Planets) 118, 382-397.
(DOI: 10.1029/2012JE004131)
[104] Hörz F, Grieve RAF, Heiken GH, Spudis P, Binder A. 1991 Lunar surface processes. In The Lunar
Sourcebook: A User's Guide to the Moon (eds GH Heiken, D Vaniman, BM French), pp. 61-120. Cambridge:
Cambridge University Press.
[105] Xie M, Xiao Z, Xu A. 2018 Modeling the growth of regolith on the Moon: implication for the evolution of
crater and impactor populations. In 49th Lunar and Planetary Science Conf., Woodlands, Texas, 19–23 March
2018, abstract no. 1992. (https://www.hou.usra.edu/meetings/lpsc2018/pdf/1992.pdf)
[106] Head JW, Wilson L. 2019 Rethinking lunar mare basalt regolith formation: new concepts of lava flow
protolith and evolution of regolith thickness and internal structure. In 50th Lunar and Planetary Science Conf.,
Woodlands, Texas, 18–22 March 2019, abstract no. 2532.
(https://www.hou.usra.edu/meetings/lpsc2019/pdf/2532.pdf)

[107] Gaddis LR, Staid MI, Tyburczy JA, Hawke, BR, Petro N. 2003 Compositional analyses of lunar pyroclastic
deposits. Icarus 161, 262-280. (DOI: 10.1016/S0019-1035(02)00036-2)
[108] Head JW, Wilson L. 2017 Generation, ascent and eruption of magma on the Moon: new insights into
source depths, magma supply, intrusions and effusive/explosive eruptions (part 2: observations). Icarus 283,
176-223. (DOI: 10.1016/j.icarus.2015.12.039)
[109] Huff AE, Hagerty JJ, Skinner JA, Fortezzo CM, Gaither TA. 2015 Excavation depths of small diameter
craters on a pyroclastic deposit near northern Rimae Bode, the Moon. In 46th Lunar and Planetary Science Conf.,
Woodlands, Texas, 16–20 March 2015, abstract no. 2386.
(https://www.hou.usra.edu/meetings/lpsc2015/pdf/2386.pdf)
[110] McKay DS. 2009 Do lunar pyroclastic deposits contain the secrets of the Solar System? In Lunar
Reconnaissance Orbiter Science Targeting Meeting, Tempe, Arizona, 9-11 June 2009 abstract no. 6014.
(https://www.lpi.usra.edu/meetings/lro2009/pdf/6014.pdf)
[111] Cohen BA, Swindle TD, Kring DA 2005. Geochemistry and 40Ar-39Ar geochronology of impact-melt
clasts in feldspathic lunar meteorites: implications for lunar bombardment history. Meteoritic. Planet. Sci. 40,
755-777. (DOI: 10.1111/j.1945-5100.2005.tb00978.x)
[112] Zellner NEB, Delano JW. 2015. 40Ar/39Ar ages of lunar impact glasses: relationships among Ar
diffusivity, chemical composition, shape, and size. Geochim. Cosmochim. Acta 161, 203-218. (DOI:
10.1016/j.gca.2015.04.013)
[113] Gombosi DJ, Baldwin SL, Watson EB, Swindle TD, Delano JW, Roberge WG. 2015 Argon diffusion in
Apollo 16 impact glass spherules: implications for 40Ar/39Ar dating of lunar impact events. Geochim.
Cosmochim. Acta 148, 251-268. (DOI: 10.1016/j.gca.2014.09.031)
[114] Fernandes V, Artemieva N. 2012 Impact ejecta temperature profile on the Moon: what are the effects on
the Ar-Ar dating method? In 43rd Lunar and Planetary Science Conf., Woodlands, Texas, 19–23 March 2012,
abstract no. 1367. (https://www.lpi.usra.edu/meetings/lpsc2012/pdf/1367.pdf)
[115] Melosh HJ. 1989 Impact Cratering: a geologic process. Oxford: Oxford University Press.
116] Weider SZ, Crawford IA, Joy KH. 2010 Individual lava flow thicknesses in Oceanus Procellarum and
Mare Serenitatis determined from Clementine multi-spectral data. Icarus 209,323-336. (DOI:
10.1016/j.icarus.2010.05.010)
[117] Ono T et al. (2009) Lunar Radar Sounder observations of subsurface layers under the nearside maria of
the Moon. Science 323, 90-92. (DOI: 10.1126/science.1165988)
[118] Fa W, Zhu M-H, Liu T, Plescia JB. 2015 Regolith stratigraphy at the Chang'E-3 landing site as seen by
lunar penetrating radar. Geophys. Res. Lett. 42, 10,179-10,187. (DOI: 10.1002/2015GL066537)
[119] Li C et al. (2020) The Moon’s farside shallow subsurface structure unveiled by Chang’E-4 lunar
penetrating radar. Sci. Adv. 6, eaay6898. (DOI: 10.1126/sciadv.aay6898)
[120] Cooper MR, Kovach RL, Watkins JS. 1974 Lunar near-surface structure. Rev. Geophys. Space Phys. 12,
291-308. (DOI: 10.1029/RG012i003p00291)
[121] Kerber L et al. 2020 Moon Diver: journey into the ancient lavas of the Moon. In 51st Lunar and Planetary
Science Conf., Woodlands, Texas, 16–21 March 2020, abstract no. 1857.
(https://www.hou.usra.edu/meetings/lpsc2020/pdf/1857.pdf)
[122] Zacny K et al. 2008 Drilling systems for extraterrestrial subsurface exploration. Astrobiology 8, 665-706.
(10.1089/ast.2007.0179)
[123] Zacny K et al. 2013 LunarVader: development and testing of lunar drill in vacuum chamber and in lunar
analog site of Antarctica. J. Aerosp. Eng. 26, 74-86. (DOI: 10.1061/(ASCE)AS.1943-5525.0000212)
[124] Shearer C et al. 2007 Analysis of lunar sample mass capability for the lunar exploration architecture.
White paper by the NASA Curation and Analysis Planning Team for Extraterrestrial Materials (CAPTEM)
(http://www.lpi.usra.edu/lunar/strategies/captem_sample_return.pdf)

[125] Woerner J, Foing B. 2016 The “Moon Village” concept and initiative. Annual Meeting of the Lunar
Exploration Analysis Group, 1-3 November 2016, Columbia, Maryland, abstract no. 5084.
(https://www.hou.usra.edu/meetings/leag2016/pdf/5084.pdf)
[126] Taylor GJ. 1985 The need for a lunar base: answering basic questions about planetary science’. In: Lunar
Bases and Space Activities of the 21st Century (ed. WW Mendell), pp. 189–197. Houston Texas: Lunar and
Planetary Institute.
[127] Ehrenfreund P et al. 2012 Toward a global space exploration program: a stepping stone approach. Adv.
Space Res. 49, 2-48. (DOI: 10.1016/j.asr.2011.09.014)
[128] McKay C. 2013 The case for a NASA research base on the Moon. New Space, 1, 162-166. (DOI:
10.1089/space.2013.0018)

Electronic Supplementary Information
Crater Size-Frequency Distribution (CSFD) Measurements
1. Method
1.1 Absolute Model Ages
Crater Size-Frequency Distribution (CSFD) measurements are a widely used method
to derive relative and absolute model ages of planetary surfaces using remote
sensing data [e.g., 1-8]. We used CraterTools in ArcGIS developed by Kneissl et al.
[9] to measure the diameters and frequencies of primary craters and the sizes of the
respective counting area. For the analyses of the measured CSFDs, CraterStats II,
developed by Michael and Neukum [10], has been utilized. To derive relative and
absolute ages from the CSFD measurements, the lunar production (PF) and
chronology (CF) functions have to be known. For our study, we used the PF of
Neukum et al. [5] which is given by an eleventh-order polynomial:
!"#
$
%!"#
&
' ($)
*
(%
&&
%'&
$
!"#
$
+
&&
%
(1)
where Ncum is the cumulative number of craters per km², a0 represents the time over
which a certain unit has been exposed to the meteorite bombardment, and D is the
crater diameter in km [2,5,11].
We also applied the empirically derived lunar chronology function (CF) of Neukum
et al. (2001), which is given by the following equation:
,()*
$
- . /012
&
' 3455 6/7-&,
8
9:;
$
<4= 6 >
&
?/
@
) A4BA 6/7-,>
(2)
In this equation, t is the crater accumulation time (crater retention age) in Ga, and
Ncum is again the cumulative number of craters per km².
In order to achieve reliable relative and absolute model ages, we used spectrally,
topographically, and morphologically homogenous counting areas and excluded
areas with secondary crater clusters [e.g. 12-17].

To determine relative ages of the mapped geologic units, we calculated the sizes of
the count areas, counted all craters apart from secondary and endogenic craters,
measured their diameters measured, and fitted the data with the PF of Neukum et al.
[5]. The derived crater frequency of craters larger or equal 1 km were linked to the CF
of Neukum et al. [5] to obtain absolute model ages (AMAs). We used Poisson statistics,
introduced by Michael at al. [18], to fit the CSFDs and to derive AMAs. This approach
uses a probability density function, which directly links the chronology function with
the measured crater density of the surface. To evaluate our CSFD measurements we
used the randomness analyses described by Michael et al. [19]. This method
determines the level of clustering by measuring the mean 2nd-closest neighbor
distances (M2CND) of every used crater size bin.
1.2 Flow Thickness
The flow thickness can be estimated based on the rim height of flooded or partly
flooded impact craters (e.g., Hiesinger et al. [20]). The relation between the rim
height and the crater diameter for simple craters < 15 km is given by Pike [13]:
Re = 0.036 Dr
1.014 (3)
where Re is the rim height (km) and Dr is the crater diameter (km).
2 Results
We derived absolute model ages for two lava flows within the Imbrium basin (see
Figure 2 of the main paper). The two flows are separated by a distinct morphologic
boundary. Based on stratigraphic observations on Kaguya images the topographically
higher flow seems to have overflown the lower flow. However, as only the upper flow
shows a clear flow front, and spectral differences are low, the overflown surface
cannot be separated into individual lava flows. Thus, the relative stratigraphic
relationship between the upper and the lower flow could be more complex than
visible with the available data sets. To determine the relative and absolute age
difference of the two flows we performed CSFD measurements of the two flows,
respectively. The counting area locations of these measurements are shown in
Supplementary Figure 1.

Based on our CSFD measurements, the lower flow shows an AMA of 3.3 +0.04/-0.05
Ga (Supplementary Figure 2). The upper flow shows an AMA of 3.03 +0.12/-0.17 Ga
(Supplementary Figure 3) being ~300 Ma younger than the lower flow. Since the error
bars of both AMAs are relatively small, this age difference seems to be reliable.
The CSFD of the lower flow shows that all measured crater sizes have been
resurfaced by the upper flow. The largest measured crater shows a diameter of 473
m and allows us to estimate the minimum flow thickness of the upper flow. Based
on the rim height / crater diameter relation for lunar simple craters determined by
Pike [13] this translates into a minimum flow thickness of 18.6 m.
Supplementary Figure 1: Counting areas of the low (left) and the upper (right) flow
in blue. The measured craters are marked in red. Base map is the 10 m/pixel Kaguya
Terrain Camera (TC) mosaic (Image centre at 32.3°N and 22.1° W).

Supplementary Figure 2. CSFD analysis of the lower lava flow. (a) Summary plot of
the M2CND randomness analyses plotted in standard deviations (!), above or below
the Monte Carlo-derived mean; as most of the crater size bins are between +1 and -1
standard deviations, the influence of clustering due to secondary cratering is
negligible. (b) CSFDs of the lower flow, shown in a differential plot; the calculated
AMA (µ) and error bars come along with a plot of the uncertainty as a probability
density function, marked at the 50 and 50 ±34 percentiles.

Supplementary Figure 3. CSFD analysis of the upper lava flow. (a) Summary plot of
the M2CND randomness analyses plotted in standard deviations (!), above or below
the Monte Carlo-derived mean; as most of the crater size bins are between +1 and -1
standard deviations, the influence of clustering due to secondary cratering is
negligible. (b) CSFDs of the upper flow, shown in a differential plot; the calculated
AMA (µ) and error bars come along with a plot of the uncertainty as a probability
density function, marked at the 50 and 50 ±34 percentiles.
References
1. Crater Analysis Techniques Working Group (Arvidson, R.E., Boyce, J., Chapman,
C.R., Cintala, M.J., Fulchignoni, M., Moore, H., Neukum, G., Schultz, P.,

Soderblom, L., Strom, R., Woronow, A., Young, R.), 1979. Standard Techniques for
Presentation and Analysis of Crater Size-Frequency Data. Icarus 37, 467–474.
https://doi.org/doi:10.1016/0019-1035(79)90009-5
2. Neukum, G., Ivanov, B.A., 1994. Crater size distributions and impact probabilities
on Earth from lunar, terrestrial-planet, and asteroid cratering data., in: Gehrels, T.
(Ed.), Hazards Due to Comets and Asteroids. University of Arizona Press,
Tucson. pp. 359–416.
3. Hiesinger, H., Jaumann, R., Neukum, G., Head, J.W., 2000. Ages of mare basalts
on the lunar nearside. J. Geophys. Res. 105, 29239–29275. https://doi.org/0048-
0227/00/2000JE001244
4. Stöffler, D., Ryder, G., 2001. Stratigraphy and isotope ages of lunar geologic units:
Chronological standard for the inner Solar System. Space Sci. Rev. 96, 9–54.
5. Neukum, G., Ivanov, B.A., Hartmann, W.K., 2001. Cratering records in the inner
solar system in relation to the lunar reference system. Space Sci. Rev. 96, 55–86.
https://doi.org/10.1023/A:1011989004263
6. Ivanov, B.A., 2001. Mars/Moon cratering rate ratio estimates. Space Sci. Rev. 96,
87–104. https://doi.org/10.1023/A:1011941121102
7. Hartmann, W.K., Neukum, G., 2001. Cratering Chronology and the Evolution of
Mars. Space Sci. Rev. 96, 165–194. https://doi.org/10.1023/A:1011945222010
8. Stöffler, D., Ryder, G., Ivanov, B.A., Artemieva, N.A., Cintala, M.J., Grieve, R.A.F.,
2006. Cratering History and Lunar Chronology. Rev. Mineral. Geochemistry 60,
519 LP-
9. Kneissl, T., van Gasselt, S., Neukum, G., 2011. Map-projection-independent crater
size-frequency determination in GIS environments - New software tool for
ArcGIS. Planet. Space Sci. 59, 1243–1254.
https://doi.org/10.1016/j.pss.2010.03.015
10. Michael, G.G., Neukum, G., 2010. Planetary surface dating from crater size–
frequency distribution measurements: Partial resurfacing events and statistical
age uncertainty. Earth Planet. Sci. Lett. 294, 223–229.
https://doi.org/10.1016/j.epsl.2009.12.041
11. Neukum, G., 1983. Meteoritenbombardement und Datierung planetarer
Oberflächen. Ludwig-Maximilians-Universität München.
12. Oberbeck, V.R., Morrison, R.H., 1973. On the formation of the lunar herringbone
pattern, in: Planetary Science Conference IV. pp. 107–123.
13. Pike, R.J., 1980. Geometric Interpretation of Lunar Craters. U.S. Geol. Surv. Prof.
Pap. 77.

14. Melosh, H.J., 1989. Impact cratering: A geologic process. Oxford: Oxford
University Press
15. Bierhaus, E.B., Chapman, C.R., Merline, W.J., 2005. Secondary craters on Europa
and implications for cratered surfaces. Nature 437, 1125–1127.
16. McEwen, A.S., Preblich, B., Turtle, E.P., Artemieva, N.A., Golombek, M.P.,
Hurst, M., Kirk, R.L., Burr, D.M., Christensen, P.R., 2005. The rayed crater Zunil
and interpretations of small impact craters on Mars. Icarus 176, 351–381.
https://doi.org/10.1016/j.icarus.2005.02.009
17. Hiesinger, H., Head, J.W., Wolf, U., Jaumann, R., Neukum, G., 2011. Ages and
stratigraphy of lunar mare basalts: A synthesis. GSA Spec. Pap. 477, 1–51.
https://doi.org/10.1130/2011.2477(01).
18. Michael, G.G., Kneissl, T., Neesemann, A., 2016. Planetary surface dating from
crater size-frequency distribution measurements: Poisson timing analysis. Icarus
277, 279–285. https://doi.org/10.1016/j.icarus.2016.05.019
19. Michael, G.G., Platz, T., Kneissl, T., Schmedemann, N., 2012. Planetary surface
dating from crater size–frequency distribution measurements: Spatial
randomness and clustering. Icarus 218, 169–177.
https://doi.org/10.1016/j.icarus.2011.11.033
20. Hiesinger, H., Head, J.W., Wolf, U., Jaumann, R., Neukum, G., 2002. Lunar mare
basalt flow units: Thicknesses determined from crater size-frequency
distributions. Geophys. Res. Lett. 29, 89-1-89–4.
https://doi.org/10.1029/2002GL014847