Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.


Master's thesis defense

No description

Michael Küffmeier

on 15 February 2016

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Master's thesis defense

of the
40 pc
≈ 8.25

Zooming in on Star Formation and Protoplanetary Disks
How to model Protoplanetary Disks?
Global disk and shearing box models
Consider environment
Åke Nordlund
(GMCs, Zoom-in, Development)
Troels Haugbølle
(Low mass SF, Zoom-in, Development)
Paolo Padoan
(Galactic Fountains)
Aris Vasileiadis
(Transport of Short-Lived Radio Isotopes)
Michael Küffmeier
(Analysis and Visualization of Zoom Runs)
and their main roles
Michael Küffmeier
University of Copenhagen
zoom simulations
Why do we care?
"Everything is connected"
Planets form in ...
...Protoplanetary Disks form in ...
...Prestellar Cores form in ...
...Fragments of Giant Molecular Clouds
How to do it? - Strategy
Forward Modeling
Separation of Scales
Ensembles are used
Separation of Scales - Example:
The Group
Small scale:
Consider only the next 30 minutes
Predictable that everyone will stay in the room
Only need local conditions:
Initial: Everyone is in the room
Boundary: I locked the door
Medium scale:
Consider this evening/tomorrow
More difficult, but roughly predictable: Most of you are going home, sleep, and go back to the office
For more detailed prediction, more information is needed family, hobbies, etc.
Large scale:
Consider predictions for years
Extremely difficult/complicated, maybe the NSA could do it
single model represents only
one possible outcome
Applied to Star Formation
Power laws:

Larson's velocity law:
Time scales get longer with larger spatial scales
"Things are mostly the same at different scales", only occur at different spatial/temporal scales
Different events on small scales, while larger scales evolve only a bit
Representative values are given by ensemble averages
In Practice
Start to drive supersonic turbulence to establish density contrast
40 pc box in RAMSES (AMR, ideal MHD)
Large scales determine small scale dynamics
Observational constraints
Most class I or class II objects have Keplerian Disks
(Haisch et al. 2001)
Also more and more evidence that disks larger than ~50 AU form already within the first 100 kyr of Star Formation
(cf. Lindberg et al. 2014)
Magnetic braking suppresses formation of disks
cf. Machida et al. 2012, Commerçon et al. 2012,
Seifried et al. 2011
cf. Commerçon et al. 2010, Machida et al. 2014
Local molecular cloud core collapse models
Length of this box is 16 000 AU
Pre-stellar Core I
Pre-stellar Core II
16 000 AU
Accretion processes
Variety in stellar environment leads to:
Different accretion histories of individual stars, different magnetization

disks form in any case, because of
chaotic magnetic braking
Impact on Disk Characteristics
Angular momentum transport due to magnetic braking no need for radial transport by MRI
Disks are inhomogeneous, thick and flared
External Magnetic Field Evolution
5 kyr after Star Formation
100 kyr after Star Formation
Hour glass shape
Thick toroidal structure
Also magnetic towers and outflows
asymmetric jets and outflows occur
bent because anchored to large scales
smaller minimum cell size, higher outflow speeds
up to 100 km/s for minimum cell size of 0.016 AU
Accretion through channels!
also seen by other groups, cf. Seifried et al. 2013
Short replenishment times
less than 100 kyr
Very dynamic evolution
NO static (standard) accretion disk
Optional Points
Magnetic bubble at peak of accretion
Dissipation effects in ideal as well as including resistivity terms --- turbulent reconnection and null point evolution
Magnetically driven disk cleaning, potential alternative explanation for gap opening
Implementation of radiative transfer
Influence of refinement settings
Magnetically driven disk clearing
For one case, the disk is cleared by break down of thick toroidal magnetic field
(about 1 mG)
(timestep 0.4 kyr)
Collapse and quick reformation of thick toroidal field
Potential explanation for gap opening
Ab initio simulation of the formation of protoplanetary disks, from 40 pc down to 0.016 AU covering 9 orders of magnitude
Small scales are anchored in larger scales
Each star evolves differently, ensemble is needed
Disks form due to chaotic magnetic braking
Disks are inhomogeneous, dynamic and replenishment times are short
Thank you for your attention!
Turn on statistical star formation to allow supernovae driving
Turn on gravity and star formation by creating sink particles
Zoom experiments: refinement level

16, 22 and 29; i.e.: minimum cell size 126 AU, 2 AU and 0.016 AU
Magnetic Bubbles
Also seen by
Seifried et al. 2011
, but
not with

Possibly avoided/reduced by including resistivity
More than 1 G; magnetic pressure causes rebound of gas
Adapted from J. B. Simon (http://jila.colorado.edu/~jasi1566/MRI_Turbulence_in_Protoplanetary_Disks.html)
Global disk models give constraints on processes in the entire disk
Shearing boxes are useful to investigate microphysical processes
Mostly periodic boundary conditions
(what goes out to the right, comes in on the left)

Assumption of an isolated and time independent object: no interaction with the disk environment, and
no accretion
Ok, this is the point where I don't know what you're talking about...
Concept of adaptive mesh refinement in 2D
level 1
level 2
level 3
level 4
Length of individual cell
level of refinement
length of the box
Continuity equation
Momentum equation
Energy equation
Induction equation
Gauss law for magnetism
Consider only the principle
This corresponds to a conservative form
BUT they have an average overall velocity
Collect particles to
MANY particles move through space with individual velocities
Altogether: We use
adaptive mesh refinement (AMR)
and evolve the equations for
ideal magnetohydrodynamics (MHD)
idealized global disk model
shearing box
Replenishment time
Average time of the gas staying in the disk before being accreted to the star
Classical approach - Standard Accretion Disk (SAD)
The gravitational field is determined by central object
The disk is axisymmetric
The disk is geometrically thin and optically thick
Hydrostatic balance holds in the vertical direction
No disk winds, no external torques
Basic idea:
Transport angular momentum radially outwards with viscosity parameter
Magnetic braking - heuristically in a nutshell
1) Magnetic fields thread the environment around a star
2) Disk forms <=> magnetic field lines becomes toroidal since magnetic field lines are coupled to the gas
3) That causes braking
magnetic field
Somehow similar to a string puppet
Length: 250 AU
Variety in refinement settings around the star leads to:
Better refinement distribution around the sink; better resolved structure in and around the disk
Length of the boxes: 250 AU
Lower accretion to the sink particle; disk mass scales roughly with square root of accretion rate
Higher disk masses and larger disk radii due to:
Distribution of cells refined to maximum level 22 and density distribution (≈ 60 kyr after star formation)
Local cloud core collapse models
"Hyper-"global simulations/zoom simulations
length: 40 pc ≈ 8.25 million AU
length: 4000 AU
length: 250 AU
Starting from first principles, considering large scale environment
Going from 40 pc box down to smallest cell size of 0.016 AU
length: 16 AU
Corresponds to 9 orders of magnitude
Turbulence avoids magnetic braking catastrophe
(Seifried et al. 2013)
, also misalignment of B-field
(Joos et al. 2012)
Consider idealized initial and boundary conditions and simulate core collapse
Example: Machida et al. 2014
Motivation for modeling protoplanetary disks and different computational approaches
Ideas behind zoom simulations and setup
Magnetic field evolution and its effects
Disk formation and characteristics
Variability in star formation
Useful for understanding impact of different parameters, but
no statistical information
Used by many groups to study earliest stages formation and magnetic braking
Magnetic braking
Formation of bipolar outflows widely seen observationally
Reproduced in turbulence-free ideal models
Problematic in turbulent simulations
Seifried et al. 2013
Example: Machida et al. 2014
external magnetic fields
Magnetic towers collimate outflows/jets
Magnetic null points - sign posts of magnetic dissipation
Magnetic null points occur and are signs for changes in the topology of magnetic field structure
Consequence of wrapped up magnetic fields
Notice: the entire range is present also for the smallest scales
Length of all boxes: 250 AU
box length 16 AU
box length 250 AU
Larger scales provide boundary conditions for smaller scales
Troels Frostholm
(Radiative Transfer)
Søren Frimann
Tommaso Grassi
Christian Brinch
People not involved in my project, but using the code
No interaction with the universe:
Length of the box: 250 AU
Full transcript