Open problems in terrestrial planet formation Sean Raymond Laboratoire d’Astrophysique de Bordeaux …with audience contributions welcome!
Jan 28, 2016
Open problems in terrestrial planet formation
Sean RaymondLaboratoire d’Astrophysique de Bordeaux
…with audience contributions welcome!
How did the Solar System form?
• Simulations can roughly reproduce the masses and orbits of Earth and Venus (O’Brien et al 2006; Kenyon & Bromley 2006; Chambers 2001; Agnor et al 1999; Raymond et al 2006)
• Biggest problem: Mars’ small size (Wetherill 1991)
• Accretion process strongly dependent on giant planets (Levison & Agnor 2003; Raymond et al 2004)
• Goal: Reproduce inner solar system – Constrain Jup, Sat’s orbits at early times– Test relevant physics
Constraints• Masses, orbits of terrestrial planets
– Mars’ small mass is a mystery (Wetherill 1991, Chambers 2001)
– Very low eccentricities (O’Brien et al 2006)
• Structure of asteroid belt– Separation of S, C types– No evidence for remnant embryos (gaps)
• Accretion timescales from Hf/W, Sm/Nd– Earth/Moon: 50-150 Myr (Jacobsen 2005; Touboul et al 2007)
– Mars: 1-10 Myr (Nimmo & Kleine 2007)
• Water delivery to Earth– Asteroidal source explains D/H (Morbidelli et al 2000)
– Other models exist (Ikoma & Genda 2007; Muralidharan et al 2008) Str
onge
r C
onst
rain
ts
Dust (µm)
Planete-simals (~km)
Cores Embryos
Earth-sized
planets
104-5 yrs 105-7 yrs 107-8 yrs
dust sticking
Grav. collapse (cm - m)
Runaway growth
Oligarchic growth
Gas giants
Late-stage accretion
QuickTime™ and aTIFF (Uncompressed) decompressorare needed to see this picture.
QuickTime™ and aTIFF (Uncompressed) decompressorare needed to see this picture.
Runaway gas accretion
Initial conditions for late-stage accretion
• Planetary embryos (aka protoplanets) form by runaway and oligarchic growth: ~Moon-Mars sized (~105-6 yrs) (Kokubo & Ida 1998, Leinhardt & Richardson 2005)
• Late-stage accretion starts when local mass in embryos and planetesimals is comparable (Kenyon & Bromley 2006)
Kokubo & Ida 2002
Ecc
entr
icit
y
Semimajor Axis (AU)
(Giant planets must form in few Myr, so they affect late stages)
Key factors for accretion
1. Giant Planets (Levison & Agnor 2003)
– Formation models predict low eccentricity– Nice model: Jup, Sat closer than 2:1 MMR
during accretion (Tsiganis et al 2005; Gomes et al 2005)
• Perhaps in chain of resonances (Morbidelli et al 2007)
2. Disk Properties (Wetherill 1996, Raymond et al 2005)
– Total mass ~ 5 Earth masses inside 4 AU (Weidenschilling 1977; Hayashi 1981)
– ∑ ~ r-1.5 (MMSN) or perhaps more complex (Jin et al 2008; Desch 2007)
QuickTime™ et undécompresseur Cinepak
sont requis pour visionner cette image.
Nice model 2 (J, S in 3:2 MMR)
• No Mars analogs
• Embryos in asteroid belt– Inconsistent with
observed structure if embryo Mars-mass or larger
Nice model 2 (J, S in 3:2 MMR)
QuickTime™ and aTIFF (Uncompressed) decompressor
are needed to see this picture.
• No Mars analogs
• Embryos in asteroid belt– Inconsistent with
observed structure if embryo Mars-mass or larger
Nice model 2 (J, S in 3:2 MMR)
QuickTime™ et undécompresseur Cinepak
sont requis pour visionner cette image.
Eccentric Jup, Sat (e0=0.1)
Eccentric Jup, Sat (e0~0.1)
• Strong secular resonance (6) at 2.2 AU
• Mars consistently forms in correct configuration
• Earth and Venus are dryQuickTime™ and a
TIFF (Uncompressed) decompressorare needed to see this picture.
Inconsistent with Kuiper Belt structure
–no migration of giant planets possible (Malhotra 1995, Levison & Morbidelli 2003)
Influence of giant planets
QuickTime™ and a decompressor
are needed to see this picture.
Raymond, O’Brien, Morbidelli, & Kaib 2009
Influence of giant planets
Raymond, O’Brien, Morbidelli, & Kaib 2009
QuickTime™ and a decompressor
are needed to see this picture.
QuickTime™ and a decompressor
are needed to see this picture.
Hard to form low-e, highly concentrated terrestrial planet systems
Mars
• Small Mars forms naturally if inner disk is truncated at 1-1.5 AU (Agnor et al 1999; Hansen 2009)
• Can reproduce all 4 terrestrial planets if embryos only existed from 0.7-1 AU (Hansen 2009)
Hansen 2009
Other effects
• Gas disk effects:– Type 1 migration (McNeil et
al 2005; Morishima et al 2010)
– Secular resonance sweeping (Nagasawa et al 2005; Thommes et al 2008)
• Collisional fragmentation (Alexander &
Agnor 1998; Kokubo, Genda)
QuickTime™ and a decompressor
are needed to see this picture.
Morishima et al 2010
Jin et al (2008) disk
• Assume MRI is effective in inner, outer disk but not in between
• At boundary between low, high viscosity, get minimum in density
• Occurs at ~1.5 AU– Explanation for Mars’
small mass?
QuickTime™ and aTIFF (Uncompressed) decompressor
are needed to see this picture.
Jin et al (2008)
Summary• No tested configuration of Jup, Sat reproduces all
constraints (Raymond et al 2009)
– Closest is eccentric Jup, Sat but Earth is dry and JS not consistent with Kuiper Belt
• Including gas disk effects doesn’t solve the problem (Morishima et al 2010)
• Hard to reproduce Mars’ small size– Strong constraint on Jup, Sat’s orbits at early times– Was there just a narrow annulus of embryos? (Hansen 2009)
• What’s missing?– Secular resonance sweeping during disk dispersal (Nagasawa et al
2005, Thommes et al 2008)
– Something else?
Recent progress
• Morishima et al 2008, 2010• Raymond, O’Brien, Morbidelli, Kaib 2009• Hansen 2009• Thommes, Nagasawa & Lin 2008• O’Brien, Morbidelli & Levison 2006• Raymond, Quinn & Lunine 2006• Kenyon & Bromley 2006• Nagasawa, Thommes & Lin 2005• Kominami & Ida 2002, 2004• Chambers 2001• Agnor, Canup & Levison 1999
Initial conditions
• Start of chaotic growth phase (Wetherill 1985; Kenyon & Bromley 2006)
• Equal mass in 1000-2000 planetesimals and ~100 embryos (5 ME total)– Embryos is Mars’ vicinity
are 0.1-0.4 Mars masses
• Integrate for 200 Myr + with Mercury (Chambers 1999)
QuickTime™ and aTIFF (Uncompressed) decompressor
are needed to see this picture.
QuickTime™ and a decompressor
are needed to see this picture.
MarsLow-ecc.
Ast. belt
Form. time
Earth Water
Current JS
Eccentric JS
Nice model 1
Nice 1 eccentric
Nice model 2
Jin disk
Cases
• Current Jup, Sat• Jup, Sat with e0~0.1
– e ~ current values after accretion
• Nice Model 1: Jup 5.45 AU, Sat 8.12 AU, e0=0
• Nice Model 2: Jup, Sat in 3:2 MMR, low-e
• Disk: ∑~r-1 and r-1.5
– Little difference
• Disk from Jin et al (2008)– Dip in ∑ at ~1.5 AU
QuickTime™ and a decompressor
are needed to see this picture.