Motivation Q: WHY IS STAR FORMATION SO INEFFICIENT? Ṁ M gas / dyn. Log SFR. Kennicutt Log. gas / dyn

Transcription

1 Motivation Q: WHY IS STAR FORMATION SO INEFFICIENT? Ṁ M gas / dyn Log SFR Kennicutt 1998 Log gas / dyn

2 Motivation Q: WHY IS STAR FORMATION SO INEFFICIENT? Moster 2009 No Feedback 10% of baryons Log( Stellar/Halo Mass ) Number Density Stellar Feedback SMBH Feedback Observed Log( Halo Mass ) M stars [M ]

3 Feedback is the Key! SO WHAT S THE PROBLEM? Gravity Hydro + Ideal MHD Good! Chemistry / cooling physics Yes! Not bad Standard (in Galaxy Formation): Couple SNe (~1e51 erg/sn) as heating /thermal energy and then cheat : Turn off cooling Force wind by hand ( kick out of galaxy) Feedback Huh? No Feedback Piontek & Steinmetz SNe Heating Alone Cooling Turned Off

4 Stellar Feedback: How Can We Do Better? High-resolution (~1-10 pc), molecular/metal cooling (~10 K), SF at nh > 100 cm -3 Energy/Mass/Momentum/Metal Injection: SNe (II & Ia) Stellar Winds (O & AGB) Photoionization (HII) & Photoelectric Cosmic Rays? (also MHD, anisotropic conduction, diffusion)

5 (Stellar) Feedback Does: generate galactic outflows (sets galaxy stellar masses) disrupt GMCs (sets GMC lifetimes/integrated SF efficiency ; difference between bound cluster+association) self-regulate ISM/galaxy-scale collapse (sets KS law on >kpc scales) enrich ISM (metals, dust?) Feedback Does Not: regulate instantaneous collapse/sf efficiency in GMCs/cores/dense gas (KS law on <<kpc scales) power turbulence in contracting clouds alter the GMC/cloud IMF (number you see affected by lifetimes) alter the high-mass end (Salpeter slope) of the IMF (except to stop runaway accretion) alter low-mass end of the IMF (preventing fragmentation), and (more indirectly) highmass end (limiting runaway accretion) determine phase structure of ISM (hot/cold balance, not necessarily morphology) reionize the Universe (H, He from QSOs) power all of the ISM-scale turbulence (gravity!) determine (most) morphology of the ISM (gravity+supersonic turbulence) affect galaxy or core/young star clustering (gravity alone; independent of SF efficiency) power ISM turbulence

6 Cloud Properties WHERE FEEDBACK MATTERS GMC SF Efficiencies GMC MF

7 ISM Properties WHERE FEEDBACK MATTERS

8 ISM Properties WHERE FEEDBACK MATTERS turbulent velocity dispersion radial/vertical velocity dispersion No FB all feedback Only radiation No radiation Scale height (h/r) Toomre Q Time (Gyr)

9 Kennicutt-Schmidt relation emerges naturally ISOLATED GALAXIES no feedback gas / dyn 0.02 gas / dyn with feedback PFH, Quataert, & Murray, 2011a

10 How Efficient Are Galactic Super-Winds? WHAT MECHANISMS DRIVE THEM? S. Muratov et al., in prep energy scaling momentum scaling

11 How Efficient Are Galactic Super-Winds? WHAT MECHANISMS DRIVE THEM? S. Muratov et al., in prep depends on where you measure it! and not one-to-one!

12 How Efficient Are Galactic Super-Winds? AND WHAT MECHANISMS DRIVE THEM? Massive High-z Disk PFH, Quataert, & Murray, 2011c Dwarf Starburst with feedback no SNe or stellar winds no radiation pressure no feedback

13 Sub-Grid Is Not Enough WE NEED TO DO BETTER! PFH, Keres, et al. (arxiv: ) Proto-MW: Gas Temperature: Insert Winds By Hand (Sub-Grid) Following Full Feedback

14 Quenching: Don t Trust Models that Don t Do Stars Right SMALL GALAXIES BECOME BIG GALAXIES Decoupled Winds (Sub-Grid) Following Explicit Feedback

15 Dwarf Metallicities are Also Surprising DEPENDS ON DETAILS OF INFLOW-OUTFLOW INTERACTIONS Illustris sims Oppenheimer/Dave sims SAMs Bathtub Models Xiancheng Ma Outflows suppress new infall of pristine material? Metal-rich gas preferentially re-accretes in fountains?

16 Winds ( Kinematic Feedback ) BLOW OUT SOME CORE MASS (AND POWER TURBULENCE) Clark Can we just multiply the CMF by ~0.3? S. Offner Non Wind Wind See also Li ea 2010, Hansen ea 2011 Krumholz ea 2012, ApJ, 545, 46

17 Feedback from Radiation WITHOUT RADIATIVE TRANSFER THERE ARE TOO MANY BDs No RT

18 My Wish List onion structure of massive GMCs (mapping ionization fronts, X-ray regions, breakout, different shells) for ~1e3 GMCs with different ages, masses onion for galaxies: fountain/cgm (outflow/inflow) phases+kinematics (Mdot, v) as function of radius, for ~1e4 galaxies with z~0-6, M*~1e4-1e12 census of mechanisms: stellar winds (young+agb), SNe (I & II, hypernovae?), radiation pressure ( trapping factors for IR/lines?), photo-heating, cosmic rays (???) inflow rates, not just outflow