Transcript Capture of Irregular Moons by 3

Capture of Irregular Satellites
during Planetary Encounters
David Nesvorny
David Vokrouhlicky
(SwRI)
Alessandro Morbidelli
(CNRS)
 Cassini image of Phoebe
Irregular Satellites
 95 known objects: 54 at Jupiter, 26 at Saturn,
9 at Uranus, 6 at Neptune (excluding Triton)
(Gladman, Sheppard, Jewitt, Holman and others)
 1-km to 340-km diameters
 Colors ranging from ‘gray’ to ‘light red’ (Grav et al.)
 Irregular satellites have large, eccentric and
predominantly retrograde orbits
 Origin distinct from the one of regular moons
(which formed by accretion in a circumplanetary disk)
Origin of Irregular Satellites
 Capture from the circumsolar planetesimal disk
(aerodynamic gas drag, planet’s growth and expansion
of its Hill sphere, etc.)
 All have one important drawback: formed IR satellites are
dynamically removed later when planets migrate in the
planetesimal disk (e.g., Beauge et al. 2002)
 In the Nice model (planets migrate, Jupiter & Saturn cross 2:1,
excited orbits of Uranus & Neptune stabilized by dynamical friction):
any original populations of irregular satellites are removed
during encounters between planets (Tsiganis et al. 2005)
New model for Capture
 We propose a new model:
‘Irregular satellites were captured during
planetary encounters when background
planetesimals were deflected into bound
orbits around planets as a result of 3-body
gravitational interactions’
Capture during Planetary Encounters
 We performed 50 new simulations of the Nice model,
~20 successful runs produced correct planetary orbits
Nice model:
example simulation (seed1)
Neptune
Uranus
Saturn
Jupiter
2:1
Capture during Planetary Encounters
 We performed 50 new simulations of the Nice model,
~20 successful runs produced correct planetary orbits
 Planetary orbits and state of the planetesimal disk were
recorded during every planetary encounter
State of the planetesimal disk recorded
at the last encounter of the seed1 run
Most orbits
beyond ~30 AU are
dynamically cold
Encounter happens at
~19 AU
Excited orbits in the
encounter zone:
<e>~0.2, <i>~10o
Capture during Planetary Encounters
 We performed 50 new simulations of the Nice model,
~20 successful runs produced correct planetary orbits
 Planetary orbits and state of the planetesimal disk were
recorded during every planetary encounter
 Typically several hundred planetary encounters
Planetary encounters
Distance
In the seed1 run:
219 encounters between
Uranus and Neptune
Speed
3 encounters between
Saturn and Neptune
1-3 km/s encounter speeds
Capture during Planetary Encounters
 We performed 50 new simulations of the Nice model,
~20 successful runs produced correct planetary orbits
 Planetary orbits and state of the planetesimal disk were
recorded during every planetary encounter
 Typically several hundred planetary encounters but not
enough disk particles to record captures directly
 Bulirsch-Stoer integrations, 3 million objects (clones of
original disk particles) were injected into the encounter
zone at each recorded encounter
Capture during Planetary Encounters
 We performed 50 new simulations of the Nice model,
~20 successful runs produced correct planetary orbits
 Planetary orbits and state of the planetesimal disk were
recorded during every planetary encounter
 Typically several hundred planetary encounters but not
enough disk particles to record captures directly
 Bulirsch-Stoer integrations, 3 million objects (clones of
original disk particles) were injected into the encounter
zone at each recorded encounter
 Our model accounts for the encounter sequence where
satellites are captured, removed or may switch between
parent planets
Capture during Planetary Encounters
Will show in the following
the results for the last 22
encounters between Uranus
and Neptune in the seed1
run
# of satellites captured at Neptune in
the last 22 encounters in seed1
Generations of satellites
Late
generations
captured during early
planetary encounters
do not contribute much
to the final population
~320 stable satellites
Early
generations
captured around Neptune
in this experiment
(out of 3 million test particles)
~10-7-10-8 capture probability
per one particle in the disk
Orbit distributions of captured objects in
the last 22 encounters in seed1
Satellites of Uranus
Wide range of inclinations
and eccentricities
Semimajor axis values up
to ~0.25 AU
Satellites of Neptune
Comparison with orbits of known
irregular moons
Satellites of Uranus
A good agreement for Uranus
Two IR satellites of Neptune,
S/2002 N4 and S/2003 N1,
have a~0.32 AU
Satellites of Neptune
Comparison with SFD of known
irregular moons
Jupiter
Saturn
35 Earth masses, Bernstein et al.
SFD of present Kuiper belt, & our
capture efficiency
 Planetary encounters produce
Uranus
Neptune
more small irregular satellites
than needed, their SFD is steeper
 Indicates that the SFD of the
planetesimal disk may have
been shallow during planetary
encounters
Conclusions
 Planetary encounters in the Nice model remove
pre-existing irregular satellites and create large
populations of the new ones
 Shallow SFD of planetesimals at 10-30 AU
at the time when encounters happened;
constraint on timing (early vs. LHB Nice models)
 Results consistent with spectroscopic obs. of
IR moons that show no clear correlation between
color and heliocentric distance
Captures via Exchange Reactions
 Observed large fraction of binaries in Kuiper Belt
 Exchange reactions suggested by Agnor &
Hamilton (2006) as an attractive model to capture
Neptune’s Triton; proposed by H. Levison for
irregular satellites
 We have studied exchange reactions for
irregular satellites via numerical simulations of
the late phase of planet migration and
via millions of scattering experiments
Distribution of encounter speeds between
planets and planetesimals
 Speeds typically a few km/s
 To capture by exchange,
orbit speed of the binary needs
to be comparable or larger than
the encounter speed
 Requires large, planetary-sized
mass of the binary
Orbits of objects captured by
exchange reactions
2 Mars-mass primary
and several million
encounter experiments
We varied binary’s
semimajor axis, inclination
and orientation of its orbit
relative to the target plane
Encounters taken from
migration runs
Good capture efficiency
but produced orbits have
large e or small a
Conclusions
 Exchange reactions during binary-planet
encounters require a planetary-sized primary
 Captured objects have very large eccentricities
and/or small semimajor axis values
 Requires additional mechanism that can
expand captured orbits
(e.g., at Neptune, captured and tidally-evolving Triton may
scatter stuff around, Cuk & Gladman 2005)