Transcript Forming Planetary Crusts II - Lunar and Planetary Laboratory

PTYS 554
Evolution of Planetary Surfaces
Forming Planetary Crusts II
PYTS 554 – Forming Planetary Crusts II

Forming Planetary Crusts I




Forming Planetary Crusts II





Tour of planetary surfaces
Terrestrial planet formation
Differentiation and timing constraints
Giant impacts and the end of accretion
Magma oceans and primary crust formation
KREEP
Late veneers and terrestrial planet water
Forming Planetary Crusts III



One-plate planets vs. plate tectonics
Recycling crust
Plate tectonic changes over the Hadean and Archean
2
PYTS 554 – Forming Planetary Crusts II

The first few 107 years to 108 years


Kleine et al., 2009
T0 = 4568.2 ± 0.6 Myr formation of the CAIs
Rapid formation of planetesimals < 1Myr
 Intense Al26 heating
 Melting and differentiation into iron meteorite parent
bodies

Formation of Chondrules and Chrondrites a few
Myr later
 No differentiation due to lower 26Al levels



Vesta-like bodies formed with volcanic activity in
progress
Gas disk dissipates ~10Myr
Mars in ~10 Myr
 Silicate differentiation ~40 Myr

Earth in ~30-100Myr
 Ends with the moon-forming impact, 50-150Myr
 At 163Myr Earth has a solid surface (zircons)

Next phase (~50 Myr) involves giant impacts –
the leading theory for…



Stripping of Mercury’s silicate mantle
Formation of Earth’s moon
Formation of Mars topographic dichotomy
Chambers et al., 2009
3
PYTS 554 – Forming Planetary Crusts II
Overview of a rocky planet



Starts as homogeneous mix of rock & iron
Molten state allows differentiation
Iron core cools and solidifies (not yet complete for the Earth)
Millions
of years
Billions
of years
~12,800
km

4
PYTS 554 – Forming Planetary Crusts II

Planets start hot

Gravitational potential energy of accreting mass
 Minimum energy delivered as velocity might be more than the escape velocity
Spread over the planet’s surface
increasing radius by ΔR

Integrate over the planets radius to get total energy delivered

Convert this energy to a temperature rise:
M P CP DT =
 Ignore cooling for now
DT =
3GM P2
E=
5RP
3GM P2
5RP
4Gpr 2
Rp = 6.28 ´10-10 Rp2
5CP

ΔT for the Earth is very large >>> melting temperature

ΔT ~ melting temperature means R~1000 km
 Objects bigger than large asteroids melt during accretion

Differentiation also releases gravitational potential energy



Amount depends on core/mantle density contrast and size of core
Typically enough to melt the body
Hf/W isotopes show differentiation essentially contemporaneous with accretion
5
PYTS 554 – Forming Planetary Crusts II

Final phase

High relative velocities
 Low gravitational focusing
 An inefficient process
 Takes ~ 100Myr

Gas has disappeared now

Jupiter and Saturn are fully formed
 Heavily affects outcome in the asteroid belt
 Determines what regions contribute the
terrestrial planet material

Final number, masses and positions
of terrestrial planets are essentially
random.
6
PYTS 554 – Forming Planetary Crusts II

Three possible impacts giant impacts to consider…

Formation of an iron-rich Mercury

Formation of Earth’s Moon

Mars Crustal dichotomy
7
PYTS 554 – Forming Planetary Crusts II
Mercury’s Abnormal Interior


Mercury’s uncompressed density (5.3 g cm-3) is much higher than
any other terrestrial planet.
For a fully differentiated core and mantle




Core radius ~75% of the planet
Core mass ~60% of the planet
Larger values are possible if the core is not pure iron
3 possibilities
 Differences in aerodynamic drag between
metal and silicate particles in the solar
nebula.
 Differentiation and then boil-off of a silicate
mantle from strong disk heating and vapor
removal by the solar wind.
 Differentiation followed by a giant impact
which can strip away most of the mantle.
8
PYTS 554 – Forming Planetary Crusts II
9
PYTS 554 – Forming Planetary Crusts II

Basic story

Mercury forms and differentiates
Proto mercury is 2.25 times the
mass of the current planet

Impactor is ~1/6 of the mass

Fast, head-on, collision needed
to strip off mantle material

 In contrast to slow oblique collisions at
Earth and Mars
 Head on collisions are less likely
10
PYTS 554 – Forming Planetary Crusts II

Impact timescale

A few hours to reform the iron rich Mercury
 Magma ocean certain

Mercury must avoid re-accreting debris

Half-life of debris is ~2 Myr
 Poynting-Robertson drag


Dynamical models suggest Mercury can
reaccumulate some small fraction of its old
silicate material
No samples means no constraints
Benz et al., 2007
11
PYTS 554 – Forming Planetary Crusts II
Formation of the Moon

Facts to consider

Moon depleted in iron & volatile substances


Bulk Earth 30% iron (mostly core)
Bulk Moon 8-10% iron (mostly in mantle FeO)

Oxygen isotope ratios similar to Earth

Moon doesn’t orbit in Earth’s equatorial plane
 Orbital solutions show that original inclination was close to 10 degrees

Angular momentum of Earth-Moon system is anomalously high
 Corresponds to spinning an isolated Earth in 4 hrs

Geochemical evidence for magma ocean
 Floating anorthosite
 Uniform age of highland material – more on this later
12
PYTS 554 – Forming Planetary Crusts II

Possible theories (that didn’t work)
 Earth and Moon co-accreted



Earth split into two pieces




Spinning so fast that it broke apart (fission)
…but the Moon doesn’t orbit in Earth’s equatorial plane
…and present day angular momentum isn’t high enough
Capture of passing body




Explains oxygen isotopes
Doesn’t explain iron and volatile depletion
Earth captures an independently formed moon as it passes nearby
Pretty much a dynamical miracle (Very hard to dissipate enough energy to capture)
Doesn’t explain oxygen isotope similarity to Earth
Current paradigm is Giant impact




Earth close to final size
Mars-sized impactor
Both bodies already differentiated
Both bodies formed at ~1 AU
13
PYTS 554 – Forming Planetary Crusts II

14
Free parameters

Late vs Early (mass of proto-Earth)
 Early accretion poses compositional problem

Mass ratio
 ~9:1 for late accretion
 ~Mars-sized impactor

Impact parameter
 Controls angular momentum of final system
 Values 0.7-0.8 Rearth work best
 Most probable impact angle is 45°
(b~0.707Rearth)

Canup, 2004
Approach velocity
 Minimum is escape velocity
 Best results for v/vesc ~ 1.1
b
PYTS 554 – Forming Planetary Crusts II
15
Canup, 2004
PYTS 554 – Forming Planetary Crusts II
16
Canup, 2004
PYTS 554 – Forming Planetary Crusts II
17

Isotopic ratios
may have
equilibrated
through vapor
cloud
Canup, 2004
PYTS 554 – Forming Planetary Crusts II
18
PYTS 554 – Forming Planetary Crusts II
19
PYTS 554 – Forming Planetary Crusts II

Most material in the
lunar disk comes from
the impacting body




Still several 1000K
Enough to remove
volatile elements and
water
Cores of bodies merge

Canup, 2004
Yellows/greens
Isotopic ratios
shouldn’t match
without re-equilibration
Temperature of material
that goes into the moon
is coolest


20
In the Earth
PYTS 554 – Forming Planetary Crusts II

Disks are 1.5-2 lunar masses

Formation of a lunar sized body is
possible in months

21
Tidal forces > self-gravity when inside
the Roche limit
 ~2.9 Rearth for lunar density material


Conservation of angular momentum




Kokubo et al., 2000
Moon ~15x times closer
Earth’s rotation ~3.9x faster (~6 hours)
Tides have removed some of this
angular momentum
Moon drifts outwards

aR ~ 2.9 Rearth
Tk ~ 7 hours
Optimum place to form moon is just
outside this limit where disk is thickest

From disk interaction
From terrestrial tides
PYTS 554 – Forming Planetary Crusts II

Timeline constraints?

Hf/W put the impact at >50Myr after CAIs

Anorthosite Sm/Nd 112 ± 40 Myr formation of
lunar crust
 Norman et al. 2003

KREEP (Zircon Pb/Pb) 150 Myr
 Nemchin et al. 2009

Whole moon Rb/Sr 90 ± 20 Myr
 Halliday 2008

Earths magma ocean gone by 163Myr
 Zircons again
22
PYTS 554 – Forming Planetary Crusts II
23
Mars: Crustal Dichotomy

Northern and southern hemispheres of
Mars are very distinct:

North






South








Low elevation
Few Craters – Young
Smooth terrain
Thin Crust
No Magnetized rock
High elevation
Heavily cratered – Old
Rough terrain
Thick crust
Magnetized rock
Dichotomy boundary mostly follows a
great circle, but is interrupted by Tharsis
No gravity signal associated with the
dichotomy boundary - compensated
Theories on how to form a dichotomy:




Giant impact
Several large basins
Degree 1 convection cell
Early plate tectonics
Zuber et al., 2000
PYTS 554 – Forming Planetary Crusts II



Despite all this the difference is only skin deep
Buried impact basins in the northern hemisphere have been
mapped
Before this burial the northern and southern hemispheres
were indistinguishable in age


Rules out Earth-style plate tectonics
Northern hemisphere is a thinly covered version of the
southern hemisphere

Mantled by 1-2 km of material (sediments and volcanic flows)
Frey et al., 200?
24
PYTS 554 – Forming Planetary Crusts II

25
Borealis basin


208E, 67N
4250-5300 km in radius

Shares slope break with at ~1.5
basin radii with other basins

Largest impact structure in the
solar system
Andrews-Hanna et al., 2008
PYTS 554 – Forming Planetary Crusts II

Hydrocode modeling of a vertical and oblique impacts
 3x1029


kms-1
J impact, 6-10
at 30-60°
No global melting – melt layer 10s of km thick within basin
Northern crust extracted from already depleted mantle
26
Marinova et al., 2008
Nimmo et al., 2008
 May correspond to Shergottites formed from a depleted reservoir 100Myr after most SNCs
PYTS 554 – Forming Planetary Crusts II
Giant Impacts make Magma Oceans



Lunar magma ocean was
probably at least a few hundred
km thick
Apollo 11 returned highland
fragments, first suggestion of
Magma ocean
Idea since extended to other
terrestrial planets

A melt has a bulk chemical composition, but no crystals

Minerals are mechanically separable crystals with a distinct composition


Terrestrial planets are dominated by silicon-oxygen based minerals – silicates
Silicate rocks are built from SiO4 tetrahedra
27
PYTS 554 – Forming Planetary Crusts II

Depending on how Oxygen is shared

Olivine
 Isolated tetrahedra joined by cations (Mg, Fe)
 (Mg,Fe)2SiO4 (forsterite, fayalite)

Pyroxene
 Chains of tetrahedra sharing 2 Oxygen atoms
 (Mg, Fe) SiO3 (orthopyroxenes)
 (Ca, Mg, Fe) SiO3 (clinopyroxenes)

Feldspars
 Framework where all 4 oxygen atoms are shared
28
PYTS 554 – Forming Planetary Crusts II

What happens when you cool a melt?


Bowens reaction series
Minerals begin to condense out in a certain order



Dense minerals sink
e.g. Olivine (Mg,Fe)2SiO4
Buoyant minerals rise
e.g. Anorthite Ca Al2Si2O8
‘Undesirable’ elements get more concentrated in remaining liquid
 Potassium (K), Rare Earth Elements (REE), Phosphorus (P)

The reverse happens when
you melt a solid

More on that in the
volcanism lectures
29
PYTS 554 – Forming Planetary Crusts II

Original planetary crusts from silicate differentiation

Calcium-rich plagioclase feldspar (anorthosite)
 Floats in an anhydrous melt – moon, mercury?
 Sinks in a hydrated melt – Earth, Mars, Venus
 Unstable at high pressures –
so sinking anorthosite is doomed

Lunar Case
Olivine and Pyroxene
 Sinks in shallow magma ocean

Undesirables form KREEP layer
 Non-uniformly distributed

Earth/Venus/Mars

Olivine rains out

Remaining composition is called

Basalt is a broad term (to be expounded upon in the volcanism lectures!)
 Variations in water content
 Variations in alkali metal content
 Variations in silica content

These are initial crusts that will be heavily modified by:
 Stripping by Giant Impacts
 Plate Tectonics
 Volcanism
30
Ultrabasic
Primative
Fe poor
Light
Less-dense
PYTS 554 – Forming Planetary Crusts II
Basic
Fe rich
Dark
Dense
31
Acidic
Evolved
PYTS 554 – Forming Planetary Crusts II

End result is a chemically distinct skin of
rock called a crust



32
10s of km thick
Density ~3000 kg m-3
Two main consequences of crustal
formation

Mantles depleted
 Upper mantle is more Olivine rich

Crusts enriched in isotopes
 The ‘undesirables’ are concentrated in the crust
 Radiogenic isotopes (heat sources ) mostly in the crust
Mantle rocks
Average
PYTS 554 – Forming Planetary Crusts II

33
The Moon has the ‘predicted’
anorthosite crust


Some resurfaced by later
basaltic flows
Unexplained:
 crustal thickness variation
 Non-uniform KREEP distribution
3.8 Ga

3.1 Ga
Mercury should have lost any
original anorthosite crust in
its giant impact



Messenger indicates lower
Ca/Si and Al/Si than the lunar
highlands
…but abundant volatile
species are a problem to
explain
Very low Fe and Ti
abundances
Nittler et al., 2011
PYTS 554 – Forming Planetary Crusts II

34
Venus rock composition



Sampled in only 7 locations by Soviet landers
Composition consistent with low-silica basalt
Exposed crust is <1 Gyr old though
Venera 14
PYTS 554 – Forming Planetary Crusts II

Earth’s crust is continuously recycled by plate tectonics and so we
don’t see any original crust


But we can see production of basaltic crust ongoing today
Characteristic stratigraphic sequence:

Gabbro
 (large grained basalt)

Sheeted dikes
 Each sheet was the wall of the inner ridge

Pillow basalts
 Blobs of basalt that are quickly quenched

Ocean sediments
 Fine-grained muds

Called an ophiolite sequence


Can be obducted onto a continental setting
Isua supracrustal belt – southern Greenland
 3.8 Ga
35
PYTS 554 – Forming Planetary Crusts II

Martian in-situ and orbital measurements


Crust dominated by basalt
With a thin weathered coating
McSween et al., 2009
36
PYTS 554 – Forming Planetary Crusts II

Hydrocode modeling of a vertical and oblique impacts
 3x1029


kms-1
J impact, 6-10
at 30-60°
No global melting – melt layer 10s of km thick within basin
Northern crust extracted from already depleted mantle
37
Marinova et al., 2008
Nimmo et al., 2008
 May correspond to Shergottites formed from a depleted reservoir 100Myr after most SNCs
PYTS 554 – Forming Planetary Crusts II

In decreasing order of severity…

Mercury – head-on, high velocity, collision
 Total planetary disruption

Earth – grazing, low velocity, collision
 Forms very large Moon
 Global magma oceans on both bodies

Mars – grazing, low velocity, collision
 Forms hemispheric dichotomy
 A baby magma ocean, no large moon

Vesta
 Distorted shape of object
 Ejected crustal and mantle samples to Earth

Giant impacts may have had other roles


Formation of Pluto’s moons
Rotation of Venus
38
PYTS 554 – Forming Planetary Crusts II

Explaining Earth’s water is a problem

Raymond et al., 2009

Best done with Jupiter and Saturn on
circular orbits
Explaining a small Mars is a problem


Best done with Jupiter and Saturn on
eccentric orbits, e ~ 0.1
Inconsistent with Nice model for later
giant planet migration
39
PYTS 554 – Forming Planetary Crusts II

Earth’s water




Possible Sources




1 Earth ocean ~ 1.4 x 1021 Kg
Estimates of Earth’s water content of ~5
oceans, about 0.1% MEarth
Inner nebula was too hot to allow water or
hydrous minerals
Adsorbed on dust grains at 1 AU
Comets
Asteroids (either ice or as hydrated
minerals)
Constraints



D/H of Earth’s water
Late veneer of highly siderophile elements
Moon is (mostly) dry
 Surface water after moon-formation
40
PYTS 554 – Forming Planetary Crusts II

41
D/H rules out comets

But only 3 Oort cloud comets have been
measured
 Condensed near Jupiter’s current position


Bulk comet might be different than its coma
Jupiter family comets might have a different D/H
 Condensed in Kuiper belt

Mars D/H matches comets
 Lack of crustal recycling?

Asteroids match Earth’s D/H





Only Carbonaceous Chondrites have significant
water
But addition of these asteroids would produce
the wrong Os isotopes
Earth has a late veneer of highly siderophile
elements (added post differentiation)
At ~0.003 of CI abundances (but in CI ratios)
Ordinary chondrites are an isotopic match
 Requires a ~1% MEarth addition after the moon forms

But late veneer and water delivery could come
from different sources
Drake, 2005
PYTS 554 – Forming Planetary Crusts II

Adsorbed onto dust grains?



Simulated adsorption onto forsterite
grains shows a few oceans can be
stored in this way
…but, not all adsorption sites would
contain water (e.g. competition from
H2)
Ordinary chondrites are not
hydrated…
Muralidharan et al., 2008
42