Clark R. Chapman (SwRI), W.J. Merline (SwRI), S.C. Solomon (DTM, Carnegie Institution), J.W. Head III (Brown Univ.), and R.G. Strom (Univ. Ariz.)

1

First MESSENGER Insights Concerning the Early
Cratering History of Mercury

• Clark R. Chapman (SwRI), W.J. Merline (SwRI),
  S.C. Solomon (DTM, Carnegie Institution), J.W.
  Head III (Brown Univ.), and R.G. Strom (Univ.
  Ariz.)

Workshop on Early Solar System Impact
Bombardment Lunar and Planetary Institute,
Houston TX 19-20 November 2008
2
Origins for Mercurys Craters

• Primary impact cratering
• High-velocity comets (5x lunar production rate)
• Sun-grazers, other near-parabolic comets
• Jupiter-family comets
• Crater chains may be solar-disrupted comets
  (Schevchenko Skobeleva 2005, COSPAR)
• Near-Earth, Aten, and Inter-Earth asteroids
• Ancient, possibly depleted, impactor populations
• Late Heavy Bombardment
• Outer solar system planetesimals (outer planet
  migration)
• Main-belt asteroids (planetary migration,
  collisions)
• Trojans and other remnants of terrestrial planet
  accretion
• Left-over remnants of inner solar system
  accretion
• Vulcanoids (bodies that primarily impact Mercury
  only)
• Secondary cratering
• Craters lt8 km diam. from larger impacts
• Basin secondaries up to 30 km diam. (Wilhelms)
• Endogenic craters (volcanism, etc.)

3
Main Observed Crater Populations on Mercury

• Basins dozens of multi-hundred km peak-ring and
  multi-ring basins tentatively identified by
  Mariner 10 (lower bound due to high sun)
• Schaber et al. (1977)
  35gt200 km, 19gt250 km
• Frey Lowry (1979)
  40gt200 km, 21gt250 km
• Pike (1988) 33 72-199 km, 46gt200
  km, 29gt250 km
• Highlands craters like heavily cratered terrains
  on the Moon, but fewer craters lt40 km diameter
  (due to embayment by widespread Intercrater
  Plains which may simply be older smooth
  plains)
• Light cratering of younger smooth plains
• Secondary craters chains and clusters of
  craters associated with large craters and basins
  

4
Mariner 10 Imaged 45 of Surface Vivaldi
Crater/Basin Then and Now
Mariner 10 Image Shaded Relief
MESSENGER image
5
Vivaldi Basin at Sunset Sunrise
M1
M2
6
Basins on Mercury
Caloris Basin, MSGR M1 Mariner 10
Basin X, MSGR M2 M1
7
Raphael (340 km diameter)
8
Schematic Representations of Lunar Cratering
History Alternatives

• Can various steady declining flux models have a
  high enough rate at 3.9 Ga without being too
  massive early on? (Destroy the crust,
  contaminate itor require unrealistically massive
  projectile population.)
• From cratering/age-dating perspective, we cant
  observe the history before 3.9 Ga grey areas

Strom et al. (2006)
Zahnle et al. (2007)
Mercury focus
9
Late LHB Population 1 Main-Belt Asteroids
(Strom et al., 2005)
As LHB declines, cratering by modern NEAs
dominates Population 2

• Shape of main-belt asteroid SFD matches lunar
  highland craters
• Shape of NEA SFD matches lunar maria craters
• Size-selective processes bring NEAs from main
  belt to Earth/Moon
• A solely gravitational process bringing main-belt
  asteroids into Earth-crossing orbits could
  produce highland SFD (e.g. resonance sweeping)
• BUT, main-belt SFD may not be uniquecould
  reflect a collisionally evolved population
  anywhere in the solar system
• The Nice Model could produce a comet shower
  followed by an asteroid shower

Pop. 1
Pop. 2
10
Caloris Interior and Exterior Plains
MESSENGER M1

• Counts of craters gt8 km diameter within plains
  units, both inside and exterior to Caloris
• New counts from best images from Mariner 10 and
  first MESSENGER flyby

Exterior Plains
Interior Plains
Exterior Plains
Mariner 10
11
Caloris Interior Plains 25 Older than Exterior
Plains
12
Caloris Basin Cratering Stratigraphy
Important issue raised by these results If
exterior plains are volcanic, then interpretation
of knobby texture of Odin Formation as
Cayley-Plains-like Caloris ejecta may be wrong

• Caloris mountains on rim (measured by Caleb
  Fassett) show old, Pop. 1 signature
• Crater density much higher than on plains
• SFD shape resembles that of highlands on Moon and
  Mercury
• Hence interior plains must have volcanic origin,
  cannot be contemporaneous impact melt
• Interior and exterior plains have low density,
  and flat Pop. 2 signa-tureso they formed mainly
  after the LHB had ended

13
Modeling Different Percentages of Lunar
Population 1 in Lunar Population 2(Strom)(The
lunar Class 1 craters have been subtracted from
the lunar highlands craters.)
Issue raised by Matija Cuk What is the meaning
of Class 1 lunar craters? They are fresh and
relatively new. But does the boundary between
Classes 1 and 2 for 10 km craters signify the
same age as the boundary for 100 km craters? Is
it proper to just subtract Class 1 craters from
the total highlands craters to determine the SFD
for the LHB? I dont think we know.
14
Comparison of the Caloris Interior and Exterior
Plains

• The Caloris interior plains SFD has a steeper
  slope at a higher density than the exterior
  plains.
• The interior plains are not only older, but they
  have a crater SFD (-2.5 slope) consistent with
  30 proportion of Population 1.
• The exterior plains are younger and have a crater
  SFD (-2.8 slope) consistent with only 5
  proportion of Population 1.
• The interior plains formed as the Late Heavy
  Bombardment was rapidly declining, and the
  exterior plains formed when the LHB was almost
  over (3.8 Ga).

In this model, there are 2 different (and
consistent) indications of age (a) the crater
density and (b) the shape of the crater SFD
(Population 1 or 2, or a mixture).
15
Mercury Cratering Components
but what about the absolute chronology?

• New data consistent with M10 view Pop. 1 (LHB),
  Pop. 2 (recent NEAs)
• Secondary branch upturn near 8 km (vs 2 km on
  Moon)
• Variations in R near 2 km due to proportions of
  Pop. 1, chains, clusters
• Smooth plains are 25 younger than plains on
  floor of Caloris

Population 1
Sample of MSGR cratered terrains more densely
cratered than Mar. 10 avg.
variable 2ndry SFD
?
Caloris plains
25 older than
the smooth exterior plains
Population 2
Secondaries
16
South Pole-Aitken, OrientaleWhat are their
Absolute Ages?

• South-Pole Aitken is relatively old and very
  large. Is its age 4.3 or 4.0 Ga?
• Orientale is the youngest basin. But is its age
  3.72 or 3.84 Ga?

17
Double-Ring Basin Raditladi (260 km)
18
(No Transcript)
19
Raditladi Flooded Floor and Ejecta Blanket

• Segment above excluded region is on ejecta
  deposits
• Segment below is floor of basin
• Craters on rare non-flooded regions ex-cluded
  from analysis of floor
• Note the very fresh, crater-free terrains

20
Smooth Plains West of Caloris Craters, Hills

• 770 craters, green
• 190 positive relief features (PRFs), yellow
• Cluster of PRFs on right side of image (a)
  lunar Marius Hills (b) Odin/Cayley Plains

21
Mercurys Absolute Chronology Raditladi Example

• Sequence Caloris basin ? smooth plains
  volcanism ? Raditladi basin/plains
• If lunar chronology applies, then
• If smooth plains formed early (3.9 Ga), then
  Raditladi is 3.8 Ga (red arrows)
• If smooth plains formed 3.75 Ga then Raditladis
  age is lt1 Ga! (green arrows)

Preferred!
22
Possible Role of Vulcanoids

• Zone interior to Mercurys orbit is dynamically
  stable (like asteroid belt, Trojans, Kuiper Belt)
• If planetesimals originally accreted there, they
  may or may not have survived mutual collisional
  comminution
• If they did, Yarkovsky drift of gt1 km bodies in
  to Mercury could have taken several Gyr
  (Vokroulichy et al., 2000) and impacted Mercury
  alone long after LHB
• Telescopic searches during last 20 years have so
  far failed to set stringent limits on current
  population of vulcanoids and MESSENGER is looking
  (but absence today wouldnt negate earlier
  presence)
• Vulcanoids could have cratered Mercury after the
  Late Heavy Bombardment, with little leakage to
  Earth/Moon zone that would compress Mercurys
  geological chronology toward the present (e.g.
  thrust-faulting might be still ongoing)

?
23
Two Chronologies for Mercury
4.5 4 3.5 3
2.5 2 1.5
1 0.5 NOW
Formation to magma ocean/crustal solidification
Bombardment, LHB, intercrater plains formation
Smooth plains formation, volcanism
Cratering, rays
Lobate scarps, crustal shortening
Classical Chronology
Vulcanoid Chronology
Formation to magma ocean solidification
Bombardment, LHB
Vulcanoid bombardment, intercrater plains
Smooth plains volcanism
Cratering, ray formation
Lobate scarps, crustal shortening
24
Conclusions

• Mercury has many basins, but we must await
  MESSENGERs orbital mission for a complete census
• Extensive geological activity on Mercury
  (intercrater plains, smooth plains volcanism,
  lobate scarp formation) differs from that on Moon
  Mars
• Raditladi basin may be very young
• Absolute chronology during and after LHB may
  mimic lunar chronology, but hypothetical
  vulcanoids are a wild card

25
(No Transcript)
26
Regional Variations (Strom) What Causes the
Differences?
27
Pantheon Fossae (the spider)
28
SFDs on Floor and Ejecta Deposits of Double-ring
Basin

• Total craters and fresh craters (near 1.5-2 km)
  are slightly more numerous on Raditladi floor
  than on ejecta blanket
• Ejecta deposits must be equal to or older than
  floor
• Butdifference is not statis-tically significant,
  anyway
• Large craters (gt2 km) peek through ejecta
  deposits ? these are older, Pop. 1 craters not
  thoroughly covered or destroyed by ejecta
• Steep SFD for Class 1 Dlt2 km these could be
  far-field secondaries Class 2 and clustered
  craters are comparatively rare
• Very low crater densities ? this basin is very
  young

29
It is not just the interpretations that differ
the data disagree!

• Neukum says lunar/Martian production function
  shape was the same during LHB and present Strom
  says it changed dramatically.

Strom
We need to understand why these differences
persist!
Neukum
30
Role of Late Heavy Bombardment

• LHB (whatever its cause) probably cratered
  Mercury similarly to the Moon and Mars
• What happened beforeand afteris not clear

The basin-forming epoch on the Moon (LHB) was of
brief duration compared with the period when
lunar rock ages were re-set, or the still longer
period of bombardment apparently recorded in the
HED meteorites (Bogard 1995). Chapman, Cohen
Grinspoon (2004) argue that the different
histograms may reflect sampling biases. But if
taken literally, the differences might instead
mean that different populations of bodies and/or
dynamical processes affected different planets.
Was the lunar LHB responsible for Mercurys
cratered terrains?
31
Crater Production Function

• Shoemaker first proposed steep branch as
  secondaries
• Neukum (and most others eventually) considered it
  an attribute of primaries
• Evidence from Europa and Mars now suggests
  Shoemaker was right after all
• Another question Big, secondaries from basins?
  (Wilhelms)

Secondary Branch
T.P. Highlands
32
The LHB Lunar Impact Basins
See Chapman, Cohen, Grinspoon (2007), Icarus
189, 233-245.

• The Moon is covered with multi-ringed impact
  basins.
• Paul Spudis map (lower left) shows only the most
  prominent ones.
• Wilhelms, Spudis, and Wood believe that there are
  at least 45 basins, many highly degraded.
• Nectaris and younger frontside basins have been
  dated, from argued geological associations with
  dated lunar rocks.
• At least 2/3rds of all basins are pre-Nectarian.

Basins are also common on Mars, Mercury,
Ganymede, Callisto, Iapetus, Rhea, Tethys, Vesta,
etc. But (except possibly Vesta), we have no
absolute radiometric ages for these basins. The
only reasons for believing that they were created
during the same epoch, come from indirect
dynamical and geophysical arguments.
33
Late Heavy Bombardment or Terminal Cataclysm

• Proposed in 1973 by Tera et al. to explain peak
  in radiometric ages of lunar samples 4.0 -- 3.8
  Ga
• Wilhelms (1987) shows sharp decline in
  basin-formation rate between Nectaris (3.92 Ga)
  and final basin, Orientale (3.82 Ga)
• There are few rock ages, and virtually no impact
  melt ages, prior to 3.92 Ga (probable Nectaris
  age) (Ryder, 1990)
• Basins produce copious melts (10 of involved
  materials)
• Small craters produce few melts because
  efficiency of melt-production increases with
  crater size and basin-forming projectiles
  volumetrically dominate shallow SFD
• So impact melts should be a robust marker of the
  history of basin formation

After Wilhelms (1987)
?
(Cumulative) Crater Density
Implies short, 50-100 Myr bombardment, with
minimal earlier basin formation between crustal
formation and this LHB
LHB
34
Basin Ages
(Stöffler Ryder, 2001, Space Science Reviews
critical re-evaluation of isotopic ages of lunar
geologic units.)

• Numerous un-certainties remain in the association
  of dated samples to specific basins
• Bottke et al. (2007) explore extremes Nectaris
  as old as 4.12 Ga, Imbrium as young as 3.72 Ga

35
Dynamical/Collisional Models Try to Produce Late
Basins
Modern view is that outer planets must have
migrated 100s of Myr after solar system
formation, with sweeping resonances stirring up
main-belt asteroids cataclysmically, and perhaps
with Uranus/Neptune stirring up KBOs. It is
difficult to avoid a late terminal cataclysm!

• Planetesimals left over after terrestrial planet
  formation have a main-belt-like size
  distribution.
• They are dynamically depleted (just like
  modern-day Near-Earth Asteroids) and
  collisionally evolve.
• The lunar impact flux declines by 4 orders of
  magnitude by the time visible lunar basins
  formed.
• In order to form the 4 largest, most reliably
  dated basins during broadest allowed time
  interval for LHB, initial planetesimal population
  must be 1 to 10 Earth masses!
• To avoid a ridiculously massive solar nebula in
  the terrestrial planet region, there must have
  been a late cataclysm to produce even a few
  basins around 3.9 Ga.

(Bottke et al., 2007, Icarus Can planetesimals
left over from planetary formation form basins as
late as 4.1 to 3.7 or 3.8 Ga?)
36
Basin Degradation due to Viscous Relaxation
(Baldwin, 2006)

• Lunar viscosity 1025 poises at 4.3 Ga, increased
  factor of 4 by time Orientale formed
• Viscosity would have had to increase an
  (unphysical) factor of 40 if all basins were
  formed during a short LHB
• Assumption degradation is by viscosity only not
  by erosion, filling, crater overlap, etc.

Baldwins Crater Degradation Classes
Basin Class Calc. Age Orientale
2 3.8 Ga Imbrium 3 3.84
Ga Crisium 4 3.91 Ga Nectaris
7 4.1 Ga Humorum 9 4.23
Ga WernerAiry 10 4.3 Ga
37
Terrestrial Planet Cratering (Robert Strom)

• Old Mercury, Mars, Moon similarbut
• Mars lt40 km diam. depleted by erosion, filling
  (climate)
• Mercury lt40 km depleted by intercrater
  plainsbut what are they? (Volcanic plains?)
• Mercury Post-Caloris
• Strom argues that shape is similar to highlands
• Error bars are large may be shallower
• Recent cratering (Moon, Mars) horizontal
• Strom interpretation
• LHB produced highlands
• NEAs made recent craters
• Neukum interpretation cratering population
  invariant in time and location