1. Multi-wavelength Approach to joint formation and evolution of Galaxies and AGNs Fabio Fontanot Max-Planck-Institute fuer Astronomie, Heidelberg Lubiana, 25/03/08

3. Outline Introduction to the problem of joint Formation of Galaxies and AGNs Theoretical perspective Observational Constraints Original Results Assembly of Massive Galaxies Evolution of the AGN population

4. Galaxy Formation and Evolution

5. 1. Baryonic gas falls in the gravitational potential of Dark Matter Halos 2. Baryonic gas is shock-heated to the virial temperature

6. 3. Radiative Cooling puts gas toward the center 4. Star Formation begins in disk-like structure

7. Dark Matter Halos Merger Tree TIME

8. Tidal Stripping Dynamical Friction Merging 5. Interaction of galaxies with the enviroment: instabilities modify galactic structures (bulge formation)

9. Galactic Winds Infall Feedback 6. Thermal processes in the baryonic gas Stellar

10. Active Galactic Nuclei

11. AGNs & Quasars Compact and luminous sources (L~10 46-49 erg/s) Accretion of gas onto a Supermassive Black Hole (10 6-9 Msun) at the center of galaxies Strong Connection with host galaxy formation and evolution (feedback, energy transfer) Padovani & Urry 1995

12. AGN – Host Galaxy connection Marconi & Hunt 2004

13. Observational Constraints

14. “Downsizing” Archeological Stellar populations in in massive galaxies are older than those in low-mass galaxies Massive galaxies are more metal rich than low- mass counterparts (Gallazzi+ 2005) Star formation timescales are shorter in massive galaxies (Thomas+ 2005) Stellar Mass Assembly Massive galaxies already in place at high-z (Cimatti+ 2006, Conselice+ 2007) Star Formation Activity Specific star formation rate declines more rapidly for massive galaxies (Panther+ 2007; Zheng+ 2007)

15. Space density of brighter AGNs peaks at higher redshift with respect to fainter ones Most massive BH accreted their mass faster and at higher redshift with respect to low-mass ones (Shankar+ 2004) Anti-hierarchical behavior of baryons

16. Zheng et al. 2007

17. MORGANA Model for the Rise of GAlaxies aNd Agns (Monaco, Fontanot & Taffoni, 2007)

18. 1: Complexity Mass flows Outside the integration disc instabilities minor and major mergers tidal stripping and disruption quasar winds

19. 2: Cooling & Infall hot polytropic gas in hydrostatic equilibrium equilibrium computed at each time-step gas is cold within the cooling radius the cooling radius is a dynamical variable that takes into account the hot gas from feedback

20. Viola+ 2008 MORGANA Cooling Simulations Gadget2 SPH code with entropy-conserving integration 60000 DM particles and 60000 gas particles inside the virial radius Static DM halo with NFW profile Gas profile in hydrostatic equilibrium Radiative cooling switched on Classical Cooling

22. 3: Feedback Stellar Feedback Stars provide both thermal and kinetic energy to cold gas (by Starlight and/or SNe explosions) Improved modeling (Monaco, 2004) with two phase treatment of star forming ISM

24. 3: Feedback Stellar Feedback Stars provide both thermal and kinetic energy to cold gas (by Starlight and/or SNe explosions) Improved modeling (Monaco, 2004) with two phase treatment of star forming ISM Kinetic feedback Velocity dispersion of cold clouds σ cold = σ 0 t * -⅓

25. 3: Feedback QSO feedback Accretion on central BH Energy Input

26. 1. Black Hole (BH) seed in every model galaxy 2. Creation of Gas Reservoir following instabilities (Granato+ 2004) 3. QSO shining & Feedback

27. 3: Feedback QSO feedback Accretion on central BH Energy Input QSO shining is able to change the physical conditions of stellar feedback in galaxies (Monaco & Fontanot, 2005) Triggering of galactic winds (“QSO Mode”?)

29. 3: Feedback QSO feedback Accretion on central BH Energy Input QSO shining is able to change the physical conditions of stellar feedback in galaxies (Monaco & Fontanot, 2005) Triggering of galactic winds (“QSO Mode”?) Feedback from Radio Jets Bringing energy from the center to the external regions Quenching of the cooling flows (“Radio Mode”)

30. 4: Diffuse Stellar Component Monaco, Murante, Borgani, Fontanot, 2006

31. Hopkins 2004 Cosmic Star Formation Rate

32. Stellar Mass Function

33. Fontana+ 2006

34. The effect of stellar feedback and quasar winds on the AGN population (Fontanot, Monaco, Cristiani & Tozzi 2006)

35. Hard X-ray and Optical LF

36. Space Density Evolution

37. Effect of Kinetic Feedback

38. Black Hole – Bulge Relation

39. Evolution of the Black Hole – Bulge Relation Peng+ 2006

40. The assembly of massive galaxies in hierarchical cosmology (Fontanot, Monaco, Silva & Grazian 2007)

41. Spectrophotometric Codes GRASIL (Silva+ 1998) Includes the effect of age-selective extinction (younger stellar populations are more affected by dust extinction) Computes dust emission in infrared regions Salpeter IMF

42. Redshift Distribution Cimatti+ 2002

43. Redshift Distribution

44. K-band LFs Pozzetti+ 2003 Cirasuolo+ 2006

45. SCUBA counts

46. Downsizing? MORGANA Predictions GOODS-MUSIC data

47. Conclusions Models based on Lambda CDM cosmology are able to reproduce the properties of AGN and massive galaxies We are able to reproduce the anti-hierarchical behavior of black hole growth Winds are needed Kinetic stellar feedback We are able to reproduce the early assembly and late almost-passive evolution of massive galaxies Stellar feedback Improved modeling of cooling We are not able to reproduce the observed downsizing trend of stellar mass assembly

48. Metallicity

49. The High-z QSO Luminosity Function (Fontanot et al., 2007)

50. Motivations  Quasars are luminous but rare sources  Large area surveys vs Deep survey  Bright end vs Faint end  Faint end of Luminosity Function  Measure QSO contribution to the UV background (Madau et al., 1999)  Constraints on the mechanisms responsible of the joint formation of supermassive black holes and host galaxies

51. GOODS Project  Study Galaxy Formation and Evolution over a wide range of cosmic lookback times (Giavalisco et al., 2004)  Multiwavelenght survey  Two fields centered on HDFN and CDFS  total area 320 sqarcmin

52.  Optical data from ACS (B 435, V 606, i 775, z 850 )  Selection Criteria (Cristiani et al., 2004)  Magnitude Limit 22.45 < z 850 < 25.25  Color Criteria tested on template spectra (Cristiani & Vio, 1990)  (i-z<0.35)∩(V-i<1.00)∩(1.00<B-V<3.00)  (i-z 3.00)  (i-z 0.80)∩(B-V>2.00)  (i-z 1.90) Selection of optical candidates

53.  Quasar selected with 3.5<z<5.2  Also included Ly-break and Seyfert Galaxies

54. Matching with X-ray observations  1202 optically selected candidates  557 in HDFN + 645 in CDFS  Match with Chandra surveys  Alexander et al., 2003  Giacconi et al., 2002  16 Final candidates  10 in HDFN + 6 in CDFS

55. X-ray Matching  Estimate of Visibility (Vignali et al. 2003)  Any z > 3.5 x-ray source must harbour an AGN  Type I QSOs with M 145 < −21 up to z ~ 5.2 GOODS -S #6

56. Spectroscopic Follow-up  50 LBGs out of optically selected candidates  Results: QSO candidates (Vanzella et al., 2004)  3 low-z galaxies  12 QSOs with 2.6 < z < 5.2  2 QSOs with z > 4  QSO at z = 5.186 (Barger et al. 2001)  QSO at z = 4.76 (Vanzella et al. 2004)

57. High-z LF  Faint QSOs  GOODS observations (Cristiani et al., 2004)  Bright QSOs  SDSS Quasar Data Release 3 (DR3QSO: Schneider et al. 2005)  Key Issues  Understanding systematics, selection effects and completeness  Reproducing survey features

59. Predicting QSO color evolution  Define a Statistical Sample of QSOs  High completeness redshift interval  2.2 < z < 2.25  High quality QSO spectra from SDSS  Sample of 215 QSOs  Building up template library  Computing restframe spectra  Fitting continuum  Simulating high redshift objects  Computing Statistical Properties

60. Choosing Redshift Interval

61. Results: Color Diagrams

62. Results: Color Evolution

63. Computing LFs  Analytical form for LF  Compute expected number of QSOs  Simulate magnitudes in photometric systems  Mock SDSS and GOODS catalogues  Apply selection criteria  Mock SDSS and GOODS selected catalogues  Compare observed and simulated objects  Define chi square estimator  Evaluate agreement between data and LF

64. Results: LFs FAINT END BRIGHT END

65. Results

66. Completeness