OotM 1.iii: Moon-forming impact scenarios
Sep. 26th, 2013 10:12 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
kaberettConference. Notes. Please feel encouraged to ask questions. There's about twenty of these; I am sure you will learn the drill. ;)
F ROBIN CANUP -- MOON-FORMING IMPACT SCENARIOS
"Canonical impact":
- oblique, low-velocity, Mars-sized impactor
-- results in: iron-poor, 1.5-2xlunar-mass disc, angular momentum comparable to that in current Earth and Moon
Problems:
- Earth and Moon have identical oxygen compositions. This requires E & impactor to have similar oxygen compositions, which is fine if the impactor forms near Earth... we thought. Which would also account for low velocity. But Mars' offset is SIXTY TIMES largest difference in Earth-Moon compositions. Ergo impactor unlikely to have same O-composition as target.
- Pahlevan & Stevenson (2007) found that a typical impactor is likely to be as different delO from its planet as the Earth and Mars
Solutions:
- impactor only tiny contribution?
- equilibration? Mixing between disk & Earth. Need at least 100 years delay before Moon accumulates, with lunar material processed through inner disk
Can an impact directly produce a disc + planet with like composition?
-NO, assuming post-impact angular momentum L comparable to that in current Earth-Moon L_EM and a Mars-like impactor composition.
-- impactor origin of disc dates back a LONG way; strongly resisted effects of resolution, equation of state, or grid-based vs particle-based hydrocodes
- head-on, higher-velocity hit-and-run impacts can tolerate larger contributions, but...
- BUT assumption that L must be L_EM may be incorrect! Evection resonance of Cuk & Stewart (2012) -- criticised earlier.
-- issues: narrow range of tidal parameters; results vary with tidal model; tehrmal, rotational state of Moon important (talk by Wisdom at satellite meeting)
- YES, if L > L_EM. If high-velocity, head-on impact into fast-spinning Earth, or oblique impact between two similarly sized objects, each having ~half the Earth's mass - but both require evection resonance bring angular momentum back down to current value
Half-Earth impactor (Canup 2012):
- large impactor strongly affects the composition of both the final disc and the planet
- if identical (size-wise) get no difference in compositions
- successful cases: have impact-angle 25-45 degrees, broad range of speeds, with or without pre-impact rotation, but require impactor mass at least 40% of Earth mass
Canonical vs high-AM impacts:
- canonical is good match for angular momentum, bad for composition (require equilibration)
- high-AM good for composition, require evection for angular momentum
Half-Earth impactor:
- requires >40% impactor - but this is less probable. Final impact on 20% of large planets managed this in low-resolution simulations (Agnor 1999)
- Raymond et al (2009) one in six final planets experienced this kind of late, large impact
-- an impact this size much less likely than Mars-sized impactor, but may be as much as 10% (though even smaller if incorporate Grand Tack...)
Impact into fast-spinning Earth:
- requires Earth day of <2.7 hours at time of impact
- N-body planet accretion models find very fast rotation rates, but they systematically overestimate these
-->N-body simulations cannot be used to assess probable rotation rates...
- erosive impacts do not effectively spin-up target
- accretionary impacts do, and these typically also create moons
- as moons tidally evolve, slow planet's spin
- this kind of rapid rotation required needs a half-Earth impactor to spin up Earth, followed by smaller, lunar forming impact into fast-spinning Earth, followed by spin being slowed down by Moon's capture into evection. But this means the moon formed by half-Earth impactor musn't slow down Earth's spin much else wouldn't get the necessary fast-spinning...
- spin requirement introduces strong limitation on impact history before Moon-forming event
- this requires SO many steps that it's probably less plausible than both Half-Earth and Canonical impacts
Conclusions:
- current scenarios can account fo rEarth and Moon masses, the system angular momentum, lunar iron depletion, and the Earth-Moon oxygen similarity
-- but all require system to be modified after the Moon-forming impact - they are NOT simple
-- FSE scenario requires particular events BEFORE the Moon-forming impact as well as after
- other compositional constraints may not be satisfied by all models
-- e.g. high-AM impacts must avoid altering Si, W composition of Earth's mantle relative to that of the disc as Earth accretes impactor's core
- important to better understand:
-- likelihood of equilibration and angular momentum removal by evection
-- is there _any_ differnce in the Earth-Moon oxygen compositions? Greater precision could help rule out some scenarios
-- expected impact conditions from planet formation models
- is agreed that Venus DelO17 is most important!
Discussion:
- if we end up creating a prior moon (FSE), what happens to it?
- W tells you about silicate reservoir (Hf-W systematics); Si relates to PT of core formation (experimentally)
-- so if we have Half-Earth collision, what do we get? V different PT conditions, what does the Si fractionation look like? Should be v different to its prior partitioning.
-- half-Earth suggestion gets very nicely around this problem - can let Si re-equilibrate, maybe? Or would that take too long relative to how much the Moon spins the Earth down? Would need cores to merge quickly and completely without chemically equilibrating w/ most of Earth's mantle, sadly.
- but maybe it *wasn't* simple... maybe we *do* need something complex/unlikely here?
- generally moon accumulation time very short compared to time between impacts; but could we accrete a moon over several impacts? Hasn't be quantitatively modelled, but... probably won't efficiently incorporate material from multiple impacts into final moon?
F ROBIN CANUP -- MOON-FORMING IMPACT SCENARIOS
"Canonical impact":
- oblique, low-velocity, Mars-sized impactor
-- results in: iron-poor, 1.5-2xlunar-mass disc, angular momentum comparable to that in current Earth and Moon
Problems:
- Earth and Moon have identical oxygen compositions. This requires E & impactor to have similar oxygen compositions, which is fine if the impactor forms near Earth... we thought. Which would also account for low velocity. But Mars' offset is SIXTY TIMES largest difference in Earth-Moon compositions. Ergo impactor unlikely to have same O-composition as target.
- Pahlevan & Stevenson (2007) found that a typical impactor is likely to be as different delO from its planet as the Earth and Mars
Solutions:
- impactor only tiny contribution?
- equilibration? Mixing between disk & Earth. Need at least 100 years delay before Moon accumulates, with lunar material processed through inner disk
Can an impact directly produce a disc + planet with like composition?
-NO, assuming post-impact angular momentum L comparable to that in current Earth-Moon L_EM and a Mars-like impactor composition.
-- impactor origin of disc dates back a LONG way; strongly resisted effects of resolution, equation of state, or grid-based vs particle-based hydrocodes
- head-on, higher-velocity hit-and-run impacts can tolerate larger contributions, but...
- BUT assumption that L must be L_EM may be incorrect! Evection resonance of Cuk & Stewart (2012) -- criticised earlier.
-- issues: narrow range of tidal parameters; results vary with tidal model; tehrmal, rotational state of Moon important (talk by Wisdom at satellite meeting)
- YES, if L > L_EM. If high-velocity, head-on impact into fast-spinning Earth, or oblique impact between two similarly sized objects, each having ~half the Earth's mass - but both require evection resonance bring angular momentum back down to current value
Half-Earth impactor (Canup 2012):
- large impactor strongly affects the composition of both the final disc and the planet
- if identical (size-wise) get no difference in compositions
- successful cases: have impact-angle 25-45 degrees, broad range of speeds, with or without pre-impact rotation, but require impactor mass at least 40% of Earth mass
Canonical vs high-AM impacts:
- canonical is good match for angular momentum, bad for composition (require equilibration)
- high-AM good for composition, require evection for angular momentum
Half-Earth impactor:
- requires >40% impactor - but this is less probable. Final impact on 20% of large planets managed this in low-resolution simulations (Agnor 1999)
- Raymond et al (2009) one in six final planets experienced this kind of late, large impact
-- an impact this size much less likely than Mars-sized impactor, but may be as much as 10% (though even smaller if incorporate Grand Tack...)
Impact into fast-spinning Earth:
- requires Earth day of <2.7 hours at time of impact
- N-body planet accretion models find very fast rotation rates, but they systematically overestimate these
-->N-body simulations cannot be used to assess probable rotation rates...
- erosive impacts do not effectively spin-up target
- accretionary impacts do, and these typically also create moons
- as moons tidally evolve, slow planet's spin
- this kind of rapid rotation required needs a half-Earth impactor to spin up Earth, followed by smaller, lunar forming impact into fast-spinning Earth, followed by spin being slowed down by Moon's capture into evection. But this means the moon formed by half-Earth impactor musn't slow down Earth's spin much else wouldn't get the necessary fast-spinning...
- spin requirement introduces strong limitation on impact history before Moon-forming event
- this requires SO many steps that it's probably less plausible than both Half-Earth and Canonical impacts
Conclusions:
- current scenarios can account fo rEarth and Moon masses, the system angular momentum, lunar iron depletion, and the Earth-Moon oxygen similarity
-- but all require system to be modified after the Moon-forming impact - they are NOT simple
-- FSE scenario requires particular events BEFORE the Moon-forming impact as well as after
- other compositional constraints may not be satisfied by all models
-- e.g. high-AM impacts must avoid altering Si, W composition of Earth's mantle relative to that of the disc as Earth accretes impactor's core
- important to better understand:
-- likelihood of equilibration and angular momentum removal by evection
-- is there _any_ differnce in the Earth-Moon oxygen compositions? Greater precision could help rule out some scenarios
-- expected impact conditions from planet formation models
- is agreed that Venus DelO17 is most important!
Discussion:
- if we end up creating a prior moon (FSE), what happens to it?
- W tells you about silicate reservoir (Hf-W systematics); Si relates to PT of core formation (experimentally)
-- so if we have Half-Earth collision, what do we get? V different PT conditions, what does the Si fractionation look like? Should be v different to its prior partitioning.
-- half-Earth suggestion gets very nicely around this problem - can let Si re-equilibrate, maybe? Or would that take too long relative to how much the Moon spins the Earth down? Would need cores to merge quickly and completely without chemically equilibrating w/ most of Earth's mantle, sadly.
- but maybe it *wasn't* simple... maybe we *do* need something complex/unlikely here?
- generally moon accumulation time very short compared to time between impacts; but could we accrete a moon over several impacts? Hasn't be quantitatively modelled, but... probably won't efficiently incorporate material from multiple impacts into final moon?