• Save
Semi-Analytic Modeling: Creation of the Far-IR Populations
Upcoming SlideShare
Loading in...5
×

Like this? Share it with your network

Share

Semi-Analytic Modeling: Creation of the Far-IR Populations

  • 926 views
Uploaded on

Talk given at the US SPICA workshop at JPL, Wednesday November 1, 2006.

Talk given at the US SPICA workshop at JPL, Wednesday November 1, 2006.

More in: Business , Technology
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
    Be the first to like this
No Downloads

Views

Total Views
926
On Slideshare
925
From Embeds
1
Number of Embeds
1

Actions

Shares
Downloads
0
Comments
0
Likes
0

Embeds 1

http://www.techgig.com 1

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide

Transcript

  • 1. Semi-analytic Modeling: Creation of the Far-IR Populations Andrew Benson, Caltech
  • 2.
    • Modeling galaxy formation
    • Black holes
    • Substructure
    • Star formation
    • Application to sub-mm galaxies
    • Dust extinction & emission
    • Radio emission
    • Results
    • Number counts, dN/dz
    • Conclusions
    Talk Overview Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 3. GALFORM Model GALFORM Model Gravitational collapse Dark matter and gas distributions Gas cooling rates Star formation, feedback Dynamical friction Luminosities, colors Positions and velocities Star formtn. rate, ages, composition Structure & Dynamics Morphology State of the Art: Semi-analytic Models Collaborators Carlos Frenk, Shaun Cole, Cedric Lacey, Carlton Baugh, Richard Bower, John Helly, Rowena Malbon, Cesario Almeida ( ICC, Durham, U.K. ) Martin Stringer ( Oxford University, UK/Caltech ) Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 4. How Do We Model Galaxy Formation? Cole et al. 2000 Combination of simulations, analytic results and recipes with parameters Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 5. Tracking the Growth of Black Holes Rowena Malbon et al. 2006
    • Black holes grow by:
    • Cold gas accretion in
    • galaxy mergers
    • Mergers of black holes
    • Kauffmann & Haehnelt 2000
    • Cattaneo et al. 2005
    Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 6. Fundamental Plane of Ellipticals Cesario Almeida et al. 2006 Comparison of predicted sizes of local bulge dominated galaxies with SDSS analysis by Bernardi et al. 2005 Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 7. Physical Model: Star Formation Gravitational collapse Dark matter and gas distributions Gas cooling rates Star formation, feedback Dynamical friction GALFORM Model GALFORM Model Supernovae energetics/dynamics Molecular cloud collisions Pressure-induced star formation Galactic fountain Multi-phase interstellar medium Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 8. Need for “Complicated” Models? Mass to light ratio Total group luminosity Variation of M/L with total group luminosity shows how the efficiency of galaxy formation should depend on halo mass. Galaxy formation most efficient Effectiveness of feedback processes and variation in gas cooling time within haloes of different mass drive change in M/L Eke et al. 2004, 2005 Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 9. The Challenge of (sub-)mm Galaxies Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 10. The Challenge of (sub-)mm Galaxies SCUBA image of HDF More star formation at high-z? Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 11. The Challenge of (sub-)mm Galaxies *Population of sources missed by Lyman-break dropout & UV imaging *Possibly more star formation at high redshift than previously thought *Inferred SFRs huge ~ 1000 Msun/yr! *Is all emission due to starburst or is some from an AGN? *Is a SCUBA source an elliptical galaxy in formation? *Massive galaxies in place at high-z? How can SCUBA sources be accommodated in hierarchical models? Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 12. Modeling Dust Extinction & Emission *Naïve model: assume dust temperature *Physically inconsistent! *Dust temperature should be determined by thermal equilibrium between heating and cooling of grains *With the bolometric luminosity and dust mass as parameters, and with the dust in thermal equilibrium, Which gives : Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 13.
    • Physical model for dust grains, chosen to reproduce local ISM extinction law
    • Mixture of graphite & silicate grains, with distribution of grain sizes
    • Includes PAHs (polycyclic aromatic hydrocarbon molecules)
    Modeling Dust Extinction & Emission
    • Assume dust/gas proportional to gas metallicity
    • Optical depth for dust depends on both dust mass and galaxy radius
    Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 14.
    • Free-free radiation from HII regions ionized by young stars
    Model for Radio Emission (Bressan, Silva & Granato 2002)
    • Synchrotron radiation from relativistic electrons accelerated in supernova remnants – assume const frac of SN energy radiated
    Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 15. Modeling Dust Extinction & Emission
    • Complete star formation history of galaxy, including starbursts triggered
    • by galaxy mergers.
    • 2. Scale lengths of the disk and bulge components, calculated by conserving
    • angular momentum and applying conservation of energy.
    • 3. Metallicity and cold gas mass: dust mass.
    Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 16. Modeling Dust Extinction & Emission
    • Complete star formation history of galaxy, including starbursts triggered
    • by galaxy mergers.
    • 2. Scale lengths of the disk and bulge components, calculated by conserving
    • angular momentum and applying conservation of energy.
    • Metallicity and cold gas mass: dust mass.
    • A spectro-photometric model to compute dust extinction and emission.
    Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 17. Modeling Dust Extinction & Emission
    • GRASIL : Silva et al. 1998
    • Emission from stars
    • Extinction by dust in two
    • components: clouds & diffuse
    • Computes temperature at
    • each location in galaxy
    • applying thermal eqm.
    • Composite dust spectrum
    Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 18. Modeling Dust Extinction & Emission
    • Complete star formation history of galaxy, including starbursts triggered
    • by galaxy mergers
    Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 19. Examples of Predicted SF Rates Star formation rate GREEN: total RED: Starbursts BLUE: Quiescent Disks Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 20. Example SEDs from CDM Model Quiescent spiral Ongoing burst dust stars Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 21. Model SEDs Compared to Observations M51 (spiral) M82 (starburst)
    • GRASIL model can reproduce observed SEDs of local galaxies (Silva etal 1998, Bressan etal 2002)
    Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 22. Cosmic Background from Galaxies no dust bursts quiescent total
    • integrated background slightly too high c.f. COBE obs in sub-mm
    • compatable with TeV gamma-ray absorption in mid-IR
    • somewhat too low c.f. observed number counts in optical/NIR
    TeV absn Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 23. Standard High-z Predictions 850 micron counts Lyman-break luminosity function at z=3 Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 24. Number Counts in near- & mid-IR Spitzer 3.6  m Spitzer 8  m
    • 3.6  m : counts dominated by stellar emission
    • 8  m : contribn from PAH emission becomes signif at bright fluxes
    no dust bursts quiescent total Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 25. Changes Made to Improve Predictions
    • Change to a constant star formation timescale, rather than one that scales
    • with the dynamical time
    • 2. Minor mergers trigger starbursts in gas rich disks
    Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 26. Changes Made to Improve Predictions
    • Change to a constant star formation timescale, rather than one that scales
    • with the dynamical time
    • Minor mergers trigger starbursts in gas rich disks
    • Use a flat IMF in starbursts:
    • more energy output in UV by high mass stars
    • more energy absorbed by dust
    • more dust to prevent heating to too high a temperature
    Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 27. Predictions with a Flat IMF in Starbursts 850 micron counts Lyman break LF z=3 Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 28. Which Changes Drive Agreement? Use standard IMF in bursts Switch off minor merger bursts Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 29. Predicted/Observed 850  N(z) Baugh et al. (2005) Chapman et al. (2003) Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 30. Predictions at Other Wavelengths 8 micron counts and N(z) : dust & PAHs start to dominate Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 31. Predictions at Other Wavelengths 160 micron number counts and redshift distribution Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 32. Predictions at Other Wavelengths 24 micron number counts and redshift distribution Accurate modelling of PAHs essential Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 33. Predictions at Other Wavelengths Discrepancy with inferred Photo-z n(z) at 24 microns Sources brighter than 83 micro Jy. Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 34. Evidence in Support of Top-Heavy IMF Model with top-heavy IMF matches metal abundances in ICM Nagashima et al. 2005 Type I & Type II SN Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 35. Number Counts at 24  m Accurate modelling of PAH emission is crucial Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 36. Number Counts in Far-IR 70  m 160  m Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 37. Predicted dN/dz at 3.6  m Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 38. Predicted dN/dz at 24  m Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 39. Predicted dN/dz at 70  m Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 40. Conclusions
    • Galaxy formation modeling:
    • (Necessarily) complicated
    • Can provide a very diverse set of predictions
    • Application to IR/sub-mm/mm
    • Requires detailed treatment of dust
    • GRASIL model provides this
    • Makes matching the data difficult!
    • Seems to require a top-heavy IMF in starbursts
    • Galaxy catalogs available at www.galform.org
    • More extensive catalogs available soon......
    Creation of the Far-IR Populations SPICA Workshop, November 2006
  • 41.  
  • 42. Evolution of stellar mass function
    • high-mass end evolves by factor ~10 from z=0 to 3
    • predicted evoln from z=0 to ~1 may be too rapid c.f. obs
    • caveat: variable IMF complicates comparison with obs
    Modeling Galaxy Formation IR/sub-mm/mm Sack Lunch, September 2006
  • 43. Physical Model: Star Formation
    • Galaxy formation models rely on empirical rules
    • Galaxy formation models rely on empirical rules
    • Star formation
    • Feedback
    • Good physical understanding of these is emerging
    • Use cosmological data (0.01-100 Mpc) to constrain star formation models (1-10pc)?
    • Galaxy formation models rely on empirical rules
    • Star formation
    rules used in G ALFORM
    • Galaxy formation models rely on empirical rules
    • Star formation
    • Feedback
    • Galaxy formation models rely on empirical rules
    • Star formation
    • Feedback
    • Good physical understanding of these is emerging
    Reionization at z=17: Can WMAP be correct? Jodrell Bank, October 2005
  • 44. Formation of a Galaxy in G ALFORM
    • Model predicts full dynamics of forming galaxy as a function of time
    • Model predicts full dynamics of forming galaxy as a function of time
    • Need a movie!
    • Stars
    • Model predicts full dynamics of forming galaxy as a function of time
    • Need a movie!
    • Stars
    • Dark matter
    • Model predicts full dynamics of forming galaxy as a function of time
    • Need a movie!
    • Stars
    • Dark matter
    • Full halo
    • Model predicts full dynamics of forming galaxy as a function of time
    • Need a movie!
    • Stars
    • Dark matter
    • Full halo
    • Zoomed region
    • Model predicts full dynamics of forming galaxy as a function of time
    • Need a movie!
    • Stars
    • Dark matter
    • Full halo
    • Zoomed region
    • Watch for growth of the galaxy
    • Small dark matter halos form first
    • Small dark matter halos form first
    • Merge to produce larger halos
    • Small dark matter halos form first
    • Merge to produce larger halos
    • Results in halos-within-halos
    • Process repeats
    • Small dark matter halos form first
    • Merge to produce larger halos
    • Results in halos-within-halos
    • Model predicts full dynamics of forming galaxy as a function of time
    • Need a movie!
    Reionization at z=17: Can WMAP be correct? Jodrell Bank, October 2005
  • 45. Why do we need to do better?
    • Tests of dark matter require knowledge of galaxy formation
    • Observational data sets are already way ahead of us, and getting better.....
    • Understanding galaxy formation would be really interesting!
    • Explaining data after the fact (fitting, interpretation...)
    • Explaining data after the fact (fitting, interpretation...)
    • Making predictions
    Sand, Treu & Ellis
    • Tests of dark matter require knowledge of galaxy formation
    • Tests of dark matter require knowledge of galaxy formation
    • Observational data sets are already way ahead of us, and getting better.....
    Tangential arc Radial arc Elliptical galaxy Role of Theory Galaxy Formation for the Next Decade University of Pittsburgh, January 2005
  • 46. Building the Luminosity Function
    • How do we make the luminosity function of galaxies?
    • How do we make the luminosity function of galaxies?
    • Start simple and add physics....
    • How do we make the luminosity function of galaxies?
    • Start simple and add physics....
    • Dark matter
    Benson et al. (2003) Dark matter halo masses: luminosity ∝ mass Cole et al. (2001)
    • How do we make the luminosity function of galaxies?
    • Start simple and add physics....
    • Dark matter
    • Cooling
    Benson et al. (2003) G ALFORM
    • How do we make the luminosity function of galaxies?
    • Start simple and add physics....
    • Dark matter
    • Cooling
    • Merging & Jeans
    Benson et al. (2003) Benson et al. (2003)
    • How do we make the luminosity function of galaxies?
    • Start simple and add physics....
    • Dark matter
    • Cooling
    • Merging & Jeans
    • Feedback
    • How do we make the luminosity function of galaxies?
    • Start simple and add physics....
    • Dark matter
    • Cooling
    • Merging & Jeans
    • Feedback
    • Conduction?
    Benson et al. (2003)
    • How do we make the luminosity function of galaxies?
    • Start simple and add physics....
    • Dark matter
    • Cooling
    • Merging & Jeans
    • Feedback
    • Conduction?
    • Superwinds?
    Benson et al. (2003) Galaxy Formation for the Next Decade University of Pittsburgh, January 2005
  • 47. Global star formation history Dynamical time scaling Fixed timescale Baugh et al. 2005
  • 48. Modelling dust extinction and emission
    • Complete star formation history of galaxy, including starbursts triggered
    • by galaxy mergers.
    • 2. Scale lengths of the disk and bulge components, calculated by conserving
    • angular momentum and applying conservation of energy.