SLIDE 1
For Halo and Galaxy Finding
ADVANCES IN PHASE-SPACE
Peter Behroozi, STScI
+Risa W echsler, Hao-Yi Wu, Lauren Anderson, Fabio Governato
ICTP , 5/14/15
What’s a Halo Finder?
Springel et al. (2005)
ADVANCES IN PHASE-SPACE For Halo and Galaxy Finding Peter - - PowerPoint PPT Presentation
ADVANCES IN PHASE-SPACE For Halo and Galaxy Finding Peter - - PowerPoint PPT Presentation
ADVANCES IN PHASE-SPACE For Halo and Galaxy Finding Peter Behroozi, STScI +Risa W echsler, Hao - Yi Wu, Lauren Anderson, Fabio Governato ICTP , 5/14/15 Whats a Halo Finder? Springel et al. ( 2005 ) Whats a Halo Finder? Springel et al.
SLIDE 2
SLIDE 3
What’s a Halo Finder?
Springel et al. (2005)
SLIDE 4
Where does it fit in?
N-Body Simulations Lambda CDM Springel et al. (2005), Klypin et al. (2011), …
SLIDE 5
N-Body Simulations Lambda CDM Halos + Merger Trees Behroozi et al. (2013a,b), Knollmann & Knebe (2009), …
Where does it fit in?
SLIDE 6
N-Body Simulations Lambda CDM Halos + Merger Trees Galaxies Conroy et al. (2006), Behroozi et al. (2010), Moster et al. (2010), …
Where does it fit in?
Credit: M101, Robert Gendler
SLIDE 7
N-Body Simulations Lambda CDM Halos + Merger Trees Galaxies Conroy & W echsler (2008), Behroozi et al. (2013c,d), Moster et al. (2013), … Star Formation
Where does it fit in?
SLIDE 8
N-Body Simulations Lambda CDM Halos + Merger Trees Galaxies Star Formation Gas (Cold and Hot) Metals / Outflows
Popping, PB+ (2015)
Quenching Black Hole Growth BH Merger Rates Supernovae Hosts
PB+ (prep.)
GRB Hosts
PB, R-Ruiz+ (2014)
Where does it fit in?
Peeples, PB+ (prep.) PB, Zhu+ (2015) PB, Silk+ (prep.)
Planets
PB, Peeples (2015) W atson, PB+ (prep.)
SLIDE 9
Structure Classification
SLIDE 10
Watershed Theory
For Halos Density Position (or V elocity…)
SLIDE 11
Watershed Theory
Density Position (or V elocity…) For Halos
SLIDE 12
Watershed Theory
Density Position (or V elocity…) For Halos
SLIDE 13
Watershed Theory
Density Position (or V elocity…) For Halos
SLIDE 14
Rockstar
Robust Overdensity Calculation using K-Space Topologically Adaptive Refinement
SLIDE 15
Rockstar
Robust Overdensity Calculation using K-Space Topologically Adaptive Refinement
Freely Available http://www.peterbehroozi.com/code.html Fast, Memory Efficient Small fraction of simulation runtime; massively parallel Phase Space+Time (7D), Accurate & Consistent Recovery PB, W echsler, Wu (2013) Knebe+ (2011, 2013), Onions+ (2012,2013), Srisawat+ (2013), …
SLIDE 16
Why Phase-Space?
64 66 68 70 72 74 76 78 80 Snapshot Number
- 12
- 8
- 4
4 8 X Offset From Centre of Mass [Mpc] Initial ∆v = 1000 km/s Rockstar AHF HBT VELOCIraptor SubFind
PB, Knebe, et al. in prep.
SLIDE 17
How to Find Structures?
Political Leanings Left Right Age Y
- ung
Old
SLIDE 18
How to Find Structures?
Political Leanings Left Right Age Y
- ung
Old
Berlusconi Clinton Obama Abe Hollande Merkel Cameron Renzi Putin Kim Francis Benedict XVI
SLIDE 19
The simulation is divided into FOFs for easy parallelization. For each group, particle positions and velocities are normalized by the group position and velocity dispersions, giving a natural phase-space metric.
Rockstar
Robust Overdensity Calculation using K-Space Topologically Adaptive Refinement Behroozi, W echsler, Wu (2013)
SLIDE 20
A phase space linking length is chosen adaptively such that 70% of the group’s particles are linked together.
Rockstar
Robust Overdensity Calculation using K-Space Topologically Adaptive Refinement Behroozi, W echsler, Wu (2013)
SLIDE 21
A phase space linking length is chosen adaptively such that 70% of the group’s particles are linked together. The process repeats for each subgroup: renormalization, a new linking length, and a new substructure level calculated.
Rockstar
Robust Overdensity Calculation using K-Space Topologically Adaptive Refinement Behroozi, W echsler, Wu (2013)
SLIDE 22
Once all levels of substructure are found, seed halos are placed at the deepest substructure levels and particles are assigned hierarchically to the closest seed halo in phase space.
Rockstar
Robust Overdensity Calculation using K-Space Topologically Adaptive Refinement Behroozi, W echsler, Wu (2013)
SLIDE 23
How does it work?
Behroozi, W echsler, Wu (2013)
SLIDE 24
Next Frontier
Automated Classification of Galaxies in Simulations
Credit: NGC 4676; NASA, H. Ford (JHU), G. Illingworth (UCSC/LO), M.Clampin (STScI),
- G. Hartig (STScI), the ACS Science Team, and ESA
SLIDE 25
Next Frontier
Automated Classification of Galaxies in Simulations
Credit: M101, Robert Gendler
SLIDE 26
Watershed Theory
For Spiral Galaxies Density Position (or V elocity…)
SLIDE 27
Watershed Theory
For Spiral Galaxies Density Position (or V elocity…)
SLIDE 28
Summary
Using Position + V elocity information improves stability of halo classification Rockstar Algorithm can be applied to find hierarchical structure in arbitrary-dimensional spaces Future applications to merging galaxies for IFU observations