NBI / STARPLAN University of Copenhagen The formation of gas dwarfs and rocky planets – a case for the new DISPATCH Code Åke Nordlund Niels Bohr Institute and Centre for Star and Planet Formation (STARPLAN) University of Copenhagen Exoplanets in Lund, May 2015 Slide 1
NBI / STARPLAN This talk is about two – related – potential break-throughs: A new, unified, scenario for A new, exa-scale ready planet formation concept for supercomputing Is it possible, perhaps even likely, How can one harness the and possible to demontrate, that immense computing power of all types of planets – gas the forth-coming exascale era (approaching 10 18 floating point giants, gas dwarfs & rocky planets form in basically the same way? ops per sec)? Very rapidly, and directly via pebble The new DISPATCH code introduces accretion in proto-atmospheres , a completely new approach to rather than via the round-about way extreme scaling via planetesimals With it, hypotheses about planet formation can be put to decisive tests Exoplanets in Lund. May 2015 Slide 2
NBI / STARPLAN So, who am I to make such bold claims? I make [extreme] supercomputer models , in e.g o solar physics – magnetic flux emergence & coronal heating o kinetic modeling – particle acceleration o star formation & protoplanetary disks – cf. ADS (Padoan et al, …) o planet formation – just starting … Hall marks: o realistic models , in agreement with observations – no ‘idealized models’ o three dimensions – no 2-D / 1-D / 0-D o full radiative transfer – no flux limited diffusion (the ‘look - alike’ method) Exoplanets in Lund. May 2015 Slide 3
NBI / STARPLAN Important recent observational findings – mostly from Kepler: The most common exoplanets are ‘gas dwarfs’ (aka ‘mini Neptunes’) o We just happen to not have any in our solar system Migration is much less pronounced than most predictions o As evidenced by the statistically very minor signature of resonances Close-in exoplanets are common, but hard to explain o Were they formed in situ , or by migration? Many exoplanet systems have Titius-Bode laws o But with different exponents Exoplanets in Lund. May 2015 Slide 4
NBI / STARPLAN The most common exoplanets are ‘gas dwarfs’ (aka ‘mini Neptunes’) [ ‘gas dwarf’ is a better name than ‘mini - Neptune’ ] We happen to not have any in our solar system – had we had them, the theory of planet formation might have taken a different track (earlier than will happen now ;-) Exoplanets in Lund. May 2015 Slide 5
NBI / STARPLAN Mean motion resonances leave only very weak Kepler signatures Exoplanets in Lund. May 2015 Slide 6
NBI / STARPLAN Other important facts/constraints on planet formation Jupiter must have once had a more massive atmosphere o Evidenced by its chemical abundance pattern Earth (and Mars) probably once had more massive atmospheres o Evidenced by Rayleigh fractionation patterns of their inert gases Planets plobably received major mass contributions from chondrules o Evidenced by similar isotopic fingerprints The asteroids must have formed at most a few 10 5 years after t=0 o Evidenced by differentiation of small bodies (revised 26 Al abundance) Exoplanets in Lund. May 2015 Slide 7
NBI / STARPLAN All of these facts can be consistently explained in a scenario where Planets – even rocky planets o form directly, via pebble accretion , and .. o where pebbles = chondrules: the result of hefty ‘thermal processing’ of dust o remaining dust = chondrite matrix, chondrules ‘dressed’ with dust rims Planet embryos are surrounded by ‘proto - atmospheres’ o atmospheres with outer BCs = disk pressure o they go away – mostly – when the disk BCs go away Proto-atmospheres make chondrule accretion much more efficient o increasing the cross sections and capture efficiencies Exoplanets in Lund. May 2015 Slide 8
NBI / STARPLAN All of these facts can be consistently explained in a scenario where Planets – even rocky planets o form directly, via pebble accretion , and .. For an early version of this scenario, see the proceedings of o where pebbles = chondrules: the result of hefty ‘thermal processing’ of IAU Symposium 276 (ÅN, 2011) dust Similar ideas have since been o remaining dust = chondrite matrix, chondrules ‘dressed’ with dust rims advanced by others, mainly in the context of gas giants Planet embryos are surrounded by ‘proto - atmospheres’ Guillot+14 • o atmospheres with outer BCs = disk pressure Johansen+15 • Levison+15 • o they go away – mostly – when the disk BCs go away Proto-atmospheres make chondrule accretion much more efficient o increasing the cross sections and capture efficiencies Exoplanets in Lund. May 2015 Slide 9
NBI / STARPLAN No planetesimals ! It’s important to understand that this scenario does not go the route via planetesimals. It assumes that There is a dominating accreter in every X% radius interval o This would explain Titius-Bode o … ‘the winner takes it all’ … inside -out The most extreme fluctuation wins (Hopkins 2013…2015), be it o just turbulence (Pan & Padoan, 2011...14) o streaming- and magneto-rotational instability (Johansen et al) o combinations (possibly gravity-assisted) of the above Rapid growth is due to pebble accretion in proto-atmospheres o Cf. recent papers by Chris Ormel et al (2015) o see also recent abstract by Hal Levison et al (2014..15) Exoplanets in Lund. May 2015 Slide 10
NBI / STARPLAN Proto- atmospheres (Hunten 1979, …, Lammer et al. 2014) Proto-atmospheres inevitably occur around objects in a gaseous disk: A normal atmosphere (e.g. Earth’s current) is hydrostatic , with essentially zero pressure outside A proto-atmosphere has a finite pressure outside o that makes a lot of difference: exponential depence on embryo mass! o the equations are exactly the same, but with a finite external BC All planet-embryos in gaseous disks have proto-atmospheres o they would in fact be humongous if the atmospheres were cold o but accretion by pebbles make them hot , which saves the day Exoplanets in Lund. May 2015 Slide 11
NBI / STARPLAN How can this be tested in a realistic setting? The range of scales involved is enormous: Scale physical relative Scale height of Earths current atmosphere 8 km 1 Scale height of a proto-atmosphere 600 km 10 2 Earth radius 6400 km 10 3 Hill radius 235 R E 10 5 Scale height of accretion disk at 1 AU, 100K 1000 R E 10 6 Distance to the Sun 23000 R E 10 7 Size of the solar system 100 AU 10 9 Size of the prestellar core 10 000 AU 10 11 50pc=10 8 AU Size of the a star forming region 10 15 The answer is Adaptive Mesh Refinement (AMR), but only if it’s limitations can be circumvented Exoplanets in Lund. May 2015 Slide 12
NBI / STARPLAN How can this be tested in a realistic setting? The range of scales involved is enormous: Scale physical relative Scale height of Earths current atmosphere 8 km 1 Scale height of a proto-atmosphere 600 km 10 2 Earth radius 6400 km 10 3 Hill radius 235 R E 10 5 Scale height of accretion disk at 1 AU, 100K 1000 R E 10 6 Distance to the Sun 23000 R E 10 7 Size of the solar system 100 AU 10 9 Size of the prestellar core 10 000 AU 10 11 50pc=10 8 AU Size of the a star forming region 10 15 The answer is Adaptive Mesh Refinement (AMR), but only if it’s limitations can be circumvented Exoplanets in Lund. May 2015 Slide 13
NBI / STARPLAN How can this be tested in a realistic setting? The range of scales involved is enormous: Scale physical relative Scale height of Earths current atmosphere 8 km 1 Scale height of a proto-atmosphere 600 km 10 2 Earth radius 6400 km 10 3 Hill radius 235 R E 10 5 Scale height of accretion disk at 1 AU, 100K 1000 R E 10 6 Distance to the Sun 23000 R E 10 7 Size of the solar system 100 AU 10 9 Size of the prestellar core 10 000 AU 10 11 50pc=10 8 AU Size of the a star forming region 10 15 The answer is Adaptive Mesh Refinement (AMR), but only if it’s limitations can be circumvented Exoplanets in Lund. May 2015 Slide 14
NBI / STARPLAN How can this be tested in a realistic setting? The range of scales involved is enormous: Scale physical relative Scale height of Earths current atmosphere 8 km 1 Scale height of a proto-atmosphere 600 km 10 2 Earth radius 6400 km 10 3 Hill radius 235 R E 10 5 Scale height of accretion disk at 1 AU, 100K 1000 R E 10 6 Distance to the Sun 23000 R E 10 7 Size of the solar system 100 AU 10 9 Size of the prestellar core 10 000 AU 10 11 50pc=10 8 AU Size of the a star forming region 10 15 The answer is Adaptive Mesh Refinement (AMR), but only if it’s limitations can be circumvented Exoplanets in Lund. May 2015 Slide 15
NBI / STARPLAN Other AMR codes, with part of the solution AREPO and GIZMO use unstructured and meshless representations, respectively. Their representation has the important advantage (over for example RAMSES), to respect Galilean invariance; i.e., their results are the same, independent of any bulk motion of the system under study. Main disadvantage: speed! Exoplanets in Lund. May 2015 Slide 16
Recommend
More recommend
