close

Вход

Забыли?

вход по аккаунту

?

Giant Planet Accretion and Migration: Surviving the Type I regime.

код для вставкиСкачать
Giant Planet Accretion and
Migration:
Surviving the Type I Regime
Edward Thommes
Norm Murray
CITA, University of Toronto
The Western Workshop, UWO, May 19, 2006
JPL
Gas giant formation: The core
accretion model


Gas disk lifetime sets upper limit on
gas giant formation: ~1-10 Myrs
from observations (e.g. Haisch, Lada
& Lada 2001)
The core accretion model (Mizuno
1980, Pollack et al 1996):
1.
2.

Solid core grows, ~10 MEarth
Core accretes massive gas envelope,
100+ MEarth
Observational support for core
accretion:


planet-metallicity correlation
(Gonzalez 1997, Fischer & Valenti
2003)
HD 149026 planet (Saturn mass,
~70 MEarth core; Sato et al. 2005,
Charbonneau et al 2006)
Marcy et al 2005
Planet-disk interaction
Density of planetпѓЁdisk torque
 Presence of substantial gas disk
means planet-disk interactions
important!
 Bodies in gas disk launch
density wavesпѓЁ repulsive
torque between body and inner,
outer disk
 Jupiter-mass planets open a gap,
locked into viscous evolution of
disk: “Type II” inward
migration
 Smaller bodies: no gap, outer
torques stronger: “Type I”
inward migration
Ward 1997
Geoff Bryden
Migration and accretion rates
Comparing the timescales
пѓЁ Scary result! Thus people tend to ignore/greatly reduce Type I (e.g.
Thommes, Duncan & Levison 2003, Ida & Lin 2004, Alibert et al. 2005)
But is there a way to make the worst-case scenario work...?
Accretion, no migration
Thommes & Murray 2006
Accretion + Migration
Thommes & Murray 2006
A viscously evolving disk
t=0
t=1 Myr
t=10 Myrs
Disk gas mass
Mdisk
Accretion + Migration in a
M
viscously evolving gas disk
100 AU
M30 AU
пЃЎ=10-2
Thommes & Murray 2006
Winners and losers
 Inner region:
growth too fast,
cores lost onto
star
 Outer region:
growth too slow
relative to disk
lifetime
 In between: An
annulus where
the growth rate
turns out just
right
Thommes & Murray 2006
Disk properties and core formation
 Method:
 Vary disk mass, metallicity,
пЃЎ
 For each set (MD,
пЃЎ,[Fe/H]), compute largest
protoplanet mass when 1
MJup of gas left inside 100
AU
 Results
 Disks with higher MD,
[Fe/H] do better
 There is always an
“optimal” , ~10-2-10-3;
consistent with fits to T
Tauri disks (Hartmann et al
1998)
Thommes & Murray 2006
Summary
 In the worst-case scenario of unmitigated Type I
migration:
 protoplanets in a young, massive gas disk fall onto
central star long before they can reach gas giant core
size (~10 MEarth)...
 ...but as the gas disk dissipates, a window may open
for cores to form and survive
 endgame: gas envelope accretion plays large role in
cleaning up rest of disk (cf. Lecar & Sasselov 2003)
 Predictions
 Favourable disk properties: high M(0), high [Fe/H], and
пЃЎ~10-2 - 10-3
 no giant planets (i.e. for ALMA, no gaps) in very
young, massive disks
“Dead zones” in disks
 Magnetorotational
instability (MRI) (Balbus
& Hawley 1991) leading
candidate for disk
viscosity
 MRI requires ionized disk,
to couple it to magnetic
field пѓЁcosmic rays,
stellar X-rays (near star)
 When the full vertical
column not ionized, dead
zone forms (Gammie
1996, Matsumura &
Pudritz 2003)
Gammie 1996
Disk evolution with a dead zone
 Dead zone: lower
viscosityпѓЁslower
accretionпѓЁpile-up
of gas
 Steep jumps in
surface density can
result
 How does this
affect migration...?
?
Disk torques at a surface density
jump
 Type I migration:
пЃґinner < пЃґouter, пЃґп‚µпЃ“gas
 Introducing jump in
пЃ“gas can reverse the
torque imbalance
 outer edge of a dead
zone can completely
stop Type I migration!
QuickTimeв„ў and a
TIFF (LZW) decompressor
are needed to see this picture.
Matsumura, Thommes & Pudritz, in prep.
QuickTimeв„ў and a
Cinepak decompress or
are needed to see this picture.
Thommes
A “hybrid” code: N-body+gas disk
 The N-body part: SyMBA (Duncan, Levison & Lee 1998)
 uses Wisdom-Holman (1991) symplectic method
 fast for near-Keplerian systems
 bounded energy error
 resolves close encounters
 The disk-evolution part: 1-D (azimuthally, vertically averaged)
Keplerian disk, ОЈ evolves according to
-
 -∫(dT/dr)dr applied to planet
 ...Fast! Can simulate 107 yrs in ~2 days
(Goldreich & Tremaine
1980, Ward 1997)
QuickTimeв„ў and a
Cinepak decompress or
are needed to see this picture.
Thommes 2005
Resonant exoplanets
Marcy et al. 2005
The “standard model” of core
accretion

Pollack et al (1996): 3 stages:
1. solid core accretion
2. slow gas accretion until Mgas ~
Mcore
3. runaway gas accretion

Corrections to the standard
model:


Stage 1 simplified, actually
takes longer (e.g. Thommes et
al. 2003)
Stage 2 HAS to be a lot shorter
(can be done by lowering
envelope opacity)
Pollack et al. 1996
Outline
 Background
 giant planet formation by core accretion
 migration by planet-disk interaction
 The timescale problem
 Calculations of concurrent core accretion and migration in
an evolving disk
 A way around the timescale problem
 Disk properties and the prospects for planet formation
 Summary
Документ
Категория
Презентации
Просмотров
7
Размер файла
4 882 Кб
Теги
1/--страниц
Пожаловаться на содержимое документа