1. Flow-Driven Formation of Molecular Clouds: Insights from Numerical Models The Cypress Cloud, Spitzer/GLIMPSE, FH et al. 09 Lee Hartmann Javier Ballesteros-Paredes Adrianne Slyz Julien Devriendt Andreas Burkert

2. What sets the star formation efficiency? If all the molecular gas in the Galaxy collapsed on its free-fall time, the star formation rate would be ~20 times higher than observed. The Classical Problem of Star Formation: Molecular clouds supported against collapse for many free-fall times by turbulence and/or magnetic fields. Star formation: slow equilibrium process. Traditional solution: Cited evidence: - long cloud lifetimes - existence of starless clouds - turbulent linewidths - magnetic fields 3Myr @ 100 cm -3 characteristic timescale:

3. The Points to be Made: (2) Molecular clouds are dynamic ( = not in equilibrium). They are collapsing during their formation. (1) Gravity is a long-range force. Thus, global gravity rules. Filaments are a natural consequence of global gravity. (3)“Turbulence” in molecular clouds is driven by gravity. Turbulent support does not exist. (4)Star formation is “rapid”, but inefficient. The SFE is set by rapid fragmentation during the cloud’s formation (and stellar feedback). What sets the star formation efficiency?

4. Why rethinking our picture of star formation? Taurus-Perseus, 12 CO Ungerechts & Thaddeus 1987 Perseus, 12 CO Ridge et al. 2006 Taurus, 12 CO Goldsmith et al. 2008 Molecular clouds have (“turbulent”) structure. What sets the star formation efficiency?

5. <10 Myr: local star-forming regions Hartmann et al. 01, Ballesteros-Paredes & Hartmann 07 The Classical Problem of Star Formation: What sets the star formation efficiency? “Long” cloud lifetimes? Not in local star-forming regions!

6. Why Look at Molecular Cloud Formation? Maddalena-Thaddeus Cloud, 13 CO Lee et al. 1994 Core, 24  & CS; Megeath et al. 2009 Most clouds form stars. Stellar age spreads are small (1-3 Myr). Thus, star formation is rapid. stellar ages Hartmann 03 What sets the star formation efficiency? Starless clouds?

7. Cep OB2: supernova, H II region-driven bubbles ~ 10 Myr- old cluster: supernova/ winds 50 pc 100  m IRAS dust emission 1 Myr-old stars ~ 4 Myr-old cluster, H II region Extragalactic view: (only100 pc) 10 Myr “age spread”; H II, H I, CO (H 2 ); is this a single cloud?

8. Rapid onset of star formation requires specific cloud properties: free-fall time of a spherical cloud of uniform density: - Freefall time independent of radius. - Linear density perturbations will be swept up in global collapse. Burkert & Hartmann 04 - Massive stars form only rarely (if at all) in “isolation”. Chu & Gruendl 08, Lamb et al. 09 The rapid onset of star formation requires non-linear density seeds during cloud formation. Physical Constraints Flow-Driven Cloud Formation 3Myr @ 100 cm -3

9. thermal instability The Physical Regimes of Cloud Formation Elmegreen 07 warm neutral gas :T = 5000-8000 K, n = 1…3 cm -3 cold neutral/molecular gas:T 100 cm -3 warm ionized gas :T = 10000 K, n < 1 cm -3 Flow-Driven Cloud Formation

10. Fluid Dynamics of Cloud Formation Large-scale flows: - spiral arms - expanding/colliding shells - galaxy mergers Processes & Agents: FH et al 08b - shocks & shear flows fragmentation, turbulence - radiative losses/thermal instability fragmentation, strong compression - gravity fragmentation, collapse - magnetic fields we’ll get them later  ≈ 1  < 0   1 WIM WNM CNM Thermal Equilibrium curve: P  n  log T = 4 log T = 1 Flow-Driven Cloud Formation

11. Two uniform, identical flows no assumption about turbulence colliding head-on at interface expanding shells, spiral arms with large-scale geometric perturbation mimicking unavoidable shear in non-periodic domain. allowing global gravitational modes Heating and cooling to model WNM  CNM. No stellar feedback. Hydro and MHD models. Fixed-grid simulations. 44pc A Numerical Experiment: Methods: Proteus FH et al. 04, 07, 08 Athena Stone et al. 08 Flow-Driven Cloud Formation

12. 44pc n = 3 cm -3 v = 9 km s -1 3D at 256 x 512 x 512 19 < log N [cm -2 ] < 23 FH et al. 08a blue/green : thermal fragmentation; red/yellow : local collapse; filament : global collapse Cooling, Gravity & Geometry Flow-Driven Cloud Formation

13. The “rapid” formation of molecular clouds and stars without self-gravity 22 pc FH & Hartmann 08 with self-gravity contours: HI color : CO Global gravity increases CO formation. Flow-Driven Cloud Formation N(filament) @ 10 Myr ~ 10 22 cm -2 but

14. Global vs local free-fall time FH & Hartmann 08 Thermal fragmentation:  small filling factors  short free-fall times (not coherent)  “presetting” local collapse Densest regions form stars, while the envelope might never be relevant for star formation. Goldsmith et al. 08 Flow-Driven Cloud Formation

15. The Role of Turbulence: Flow-Driven Cloud Formation “Turbulence” is (at least partially) driven by gravity. The bulk of the energy is on the largest scales. There is no such thing as “turbulent support”. Turbulence in cold gas is not supersonic hydrodynamically.

16. Clouds are not in equilibrium: Flow-Driven Cloud Formation The virial parameters are of order unity. Global gravity starts dominating the cloud’s evolution.

17. Magnetic Fields FH et al. 09 Collapse of dense regions, support of diffuse envelope. 44 pc field in equipartition with flow energy Taurus, Goldsmith/Heyer 08 collapsing cores supported, diffuse envelope FH et al. 09 Flow-Driven Cloud Formation

18. The Major Points to be Made: (1)Gravity is a long-range force. Thus, global gravity rules. Filaments are a natural consequence of global gravity. (2)Molecular clouds are dynamic ( = not in equilibrium). They are collapsing during their formation. (3)Turbulence in molecular clouds is driven by gravity.Turbulent support does not exist. The Points to be Made

19. The Major Points to be Made: (4)Star formation is “rapid”, but inefficient. The SFE is set by rapid fragmentation during the cloud’s formation (and stellar feedback). Elmegreen 07 Gritschneder et al. 09 The Points to be Made

20. The Major Points to be Made: (2) Molecular clouds are dynamic ( = not in equilibrium). They are collapsing during their formation. The Points to be Made (1) Gravity is a long-range force. Thus, global gravity rules. Filaments are a natural consequence of global gravity. (3)Turbulence in molecular clouds is driven by gravity. Turbulent support does not exist. (4)Star formation is “rapid”, but inefficient. The SFE is set by rapid fragmentation during the cloud’s formation (and stellar feedback).

21. END