feb07 am

Views:
 
     
 

Presentation Description

No description available.

Comments

Presentation Transcript

Dwarf Galaxies and Missing Satellites: Reionization Issues: 

Dwarf Galaxies and Missing Satellites: Reionization Issues Ann Martin Astro 620 February 7, 2007

Proposed Solutions to the Missing Satellite Problem: 

Proposed Solutions to the Missing Satellite Problem Very numerous subhalos in N-body sims add up to andlt; 10% of the total halo mass Proposed solutions include: Changing the character of inflation Changing the character of the dark matter (warm dark matter, self-interacting dark matter, strongly annihilating dark matter) … which introduce their own cosmological problems Or some other process occurring between z ~ 1000 and z = 0.

AKA the “Visible Substructure Problem”: 

AKA the 'Visible Substructure Problem' Missing satellite problem is a mismatch between predictions and observations. The problem can thus be solved if Dwarf galaxies cannot form or survive to z=0 Dwarf galaxies are difficult to detect (few stars and little HI gas) by z=0 Dwarf galaxies are 'invisible' (depleted of baryons) by z=0 Dark matter halos predicted by CDM are still permitted in these scenarios

Evidence for Reionization: 

Evidence for Reionization Appears to have occurred between 6 andlt; z andlt; 11 Lower limit: No Gunn-Peterson trough in the spectra of quasars at that redshift Upper limit: optical depth out to CMB last scattering surface Image credit: SDSS

Process of Reionization: 

Process of Reionization ½ Myr after the Big Bang: Universe full of neutral atoms (hydrogen) Gravitational attraction allowed structures to begin to condense First young stars and AGN form Their UV radiation ionizes the surrounding hydrogen, and the effect quickly spreads throughout the filaments and voids.

Process of Reionization cont’d: 

Process of Reionization cont’d ½ Myr after the Big Bang: UV radiation ionizes the surrounding hydrogen, and the effect quickly spreads throughout the filaments and voids. Most of the Universe’s hydrogen photoionized by z ~ 7 Gas in halos heats to 104 Kelvin, and stays there until the UV flux decreases

Outline: Effect of Reionization on Dwarf Galaxies: 

Outline: Effect of Reionization on Dwarf Galaxies Reionization provides a natural explanation for the observed discrepancy and its occurrence near Vcirc ~ 30 km s-1 Mechanisms for Disruption of Formation/Evolution of Dwarf Galaxies Photoevaporation, inhibition of collapse, prevention of cooling Related ideas: inhibition of star formation; supernova feedback Why do we still see dwarfs today? Major theoretical stumbling blocks Progress: the state of the art Dwarf Galaxies in Voids (Hoeft et al. 2006) Dwarfs in 'Local Groups' (Benson et al. 2002)

Mechanisms: 

Mechanisms How does reionization disrupt the formation and evolution of dwarf galaxies? Galaxies form in two stages of collapse: Dark matter halos Cooling of baryonic gas within halos If the gas can cool efficiently (to below the halo virial temperature), then it will lose pressure support and can form a galaxy.

Mechanisms: 

Mechanisms How does reionization disrupt the formation and evolution of dwarf galaxies? Photoevaporation Heating of gas 'boils' it out of the shallow gravitational potential wells of the host halo Inhibition of collapse Pressure of IGM prevents further gas infall for haloes up to 30 km/s Prevention of cooling By reducing the fraction of baryons in neutral atoms, the UV background has an impact on the ability of the gas to cool These mechanisms preferentially affect small halos.

Spectrum of UV Radiation: 

Spectrum of UV Radiation Radiation field, specific intensity per unit frequency: with Lyman limit frequency nL Spectral index a=1.8 for quasar spectra and a=5 for stellar spectra. Many studies find results that are insensitive to the magnitude of the flux.

Photoevaporation: 

Photoevaporation Ionized gas has a thermal velocity of order 10 km s-1 Dark matter haloes with an escape velocity lower than this thermal velocity will lose baryons at reionization Pressure force (kBÑ[rT/m]) competes with gravitational force (rGMr-2) Ratio of forces is roughly kBT to mGMr-1, or T to Tvir. Thus, for T andgt; Tvir, baryonic gas is expelled from halos. These smaller objects will not be destroyed by internal UV sources, since they tend to be unable to collapse andamp; form stars at earlier epochs (Barkana andamp; Loeb 1999)

Photoevaporation: 

Photoevaporation Left panel: Unbound fraction as a function of circular velocity. Solid curves: spectral index a = 5. Dashed curves: spectral index a=1.8 Right panel: Total fraction of gas in the Universe as a function of redshift. Solid: a=1.8. Dotted: a=5. (Loeb)

Inhibition of Collapse: 

Inhibition of Collapse Ionized IGM has high pressure: TIGM ~ 104 K Prevents further gas infall for haloes up to 30 km s-1 Barkana andamp; Loeb 2006: Later, a decline in the UV background flux could allow collapse andamp; star formation (observational test of reionization scenarios)

Prevention of Cooling: 

Prevention of Cooling UV photons ionize gas, reducing the number of neutral atoms. Thus, collisional excitation is not available for cooling. This inhibits the formation of galaxies but, even more, should (for a time) keep stars from forming in the galaxies that do survive or that collapse before reionization.

Prevention of Cooling: 

Prevention of Cooling Solid line: cooling rate as a function of temperature with no ionizing radiation. Dashed lines andamp; dotted lines: cooling curves for various UV fluxes Dot-dashed lines: rate of heating by photoionization for a=1 (upper) and a=5 (lower) Image: Efstathiou (1992)

Prevention of Cooling: 

Prevention of Cooling Peaks in solid curve: collisional excitation of HI and HeII. Peaks cannot occur if gas is highly ionized. Shows that modest flux values (as used in Efstathiou’s 1992 study) can produce 'large reductions in the cooling function.' Image: Efstathiou (1992)

Other Mechanisms from the Era of Reionization: 

Other Mechanisms from the Era of Reionization Star Formation Difficulties The smallest halos may not be able to ever cool efficiently enough to form any stars at all; predictions andamp; local observations set the limit at vvir ~ 10 km s-1 (Shaviv andamp; Dekel) Supernova Feedback Could remove a significant fraction of gas from dwarf systems, again preferentially destroying the lowest-mass systems Supernova explosions (Pop III?) create 'outflows' that impart kinetic energy to gas If these outflows are energetic enough, the gas can fully escape from the shallow potential well of the host galaxy However, this leaves the question of why any dwarfs survive until z=0, so that reionization is preferable in many minds (Bullock et al. 2000)

Why do we still see dwarfs today?: 

Why do we still see dwarfs today? Various mechanisms appear to propose 'complete destruction' of sub-populations Observations tell us that dwarfs with a variety of circular velocities survive to z=0 Dark matter halos may be substantially more massive than previously thought. Models find that the observable satellites (i.e. those that survive, with some baryons and some stars at z=0) are those that accrete a substantial amount of gas prior to the reionization epoch (Bullock et al. 2000)

Why do we still see dwarfs today?: 

Why do we still see dwarfs today? Thin bars: all surviving dark matter halos Thick line: velocity function of observable halos (zformation andgt; zreion) Triangles: velocity function of MW/M31 satellites Total number of observable systems is roughly 10% the total dark halo abundance, because most galaxies don’t form until after reionization. Dwarfs visible today correspond 'to rare, high peaks in the initial density field.' Image: Bullock et al. 2000 Velocity functions may make more sense than luminosity functions.

Major Theoretical Stumbling Blocks: 

Major Theoretical Stumbling Blocks UV photon escape fraction fesc, the fraction of ionizing photons from stars that are unable to escape from their galaxies Major uncertainty Expected to change with both redshift and galaxy properties Having the wrong escape fraction could change the intensity of the UV flux by orders of magnitude. Studies typically use anything 5% andlt; fesc andlt; 60% (Loeb 2006)

Major Theoretical Stumbling Blocks: 

Major Theoretical Stumbling Blocks Simulation resolution Overestimation of effects? Poor mass resolution makes it difficult to accurately trace the amount of gas loss (via photoevaporation and supernova blowout) Some halos in simulations don’t form any stars at all, and are thus taken to be 'invisible halos,' but it’s very difficult to resolve star formation. Our understanding of star formation Is our crude model of star formation even applicable to tiny dwarfs? The very faintest galaxies in models/simulations often contain only 103 Msun. in stars. Entire galaxies can be less massive than a single molecular cloud (Benson et al. 2002)

Results: Dwarf Galaxies in Voids (Hoeft et al): 

Results: Dwarf Galaxies in Voids (Hoeft et al) 'Voids' in CDM cosmologies do contain structures and halos Less massive than dense counterparts Predictions of small halos No observational evidence for a large number of dwarfs in voids UV heating could explain both the Local Group and the void discrepancies Observational tests of reionization scenario Theory predicts that reionization should be completed in dense areas first, and only then propagate out to voids Void dwarfs test the epoch of reionization and the effect that reionization has on the growth of structure.

Results: Dwarf Galaxies in Voids (Hoeft et al): 

Results: Dwarf Galaxies in Voids (Hoeft et al) Results A characteristic mass for baryon-poor simulated galaxies: Mc(z=0) = 6.5 x 109 h-1 Msun At reionization, halos with mass M andlt; Mc(zreion) are unable to cool or collapse further However, they find that their simulated dwarfs are still able to form stars after reionization Faint-end slope of luminosity function improves . . . But the problem is not solved Further work will need to include additional physical feedback processes.

Results: Dwarf Galaxies in Voids (Hoeft et al): 

Results: Dwarf Galaxies in Voids (Hoeft et al) Baryon fraction as a function of total galaxy mass Open circles: no UV Open squares: no thermal feedback from stars Xes: 'impulsive' heat (i.e. heat added phenomenologically between time steps) Note: mass resolution puts a lower limit on the size of the halos they consider here.

Results: Dwarf Galaxies in Voids (Hoeft et al): 

Results: Dwarf Galaxies in Voids (Hoeft et al) Sample mass accretion history in cold condensed gas for several halos Numbers indicate the dark matter mass at z=0 measured in 1010 h-1 Msun. 'Jumps' indicate mergers The 'little guys' flatten out after reionization

Results: Dwarf Galaxies in “Local Groups” (Benson et al): 

Results: Dwarf Galaxies in 'Local Groups' (Benson et al) GALFORM semi-analytical model Good success: Find 'no conflict between the Cold Dark Matter hypothesis and a low abundance of satellites in the Local Group' Luminosity function flattens out at faint end Find two epochs of re-heating (ionization of HI and HeII) with suppression of star formation after each Galaxies brighter than L* survive mostly unaffected.

Results: Dwarf Galaxies in “Local Groups” (Benson et al): 

Results: Dwarf Galaxies in 'Local Groups' (Benson et al) Consider many physical processes, including dynamics: Formation of dark halos Merging of dark halos Shock heating of gas in halos Radiative cooling of gas in halos Collapse of cold gas Star formation UV photons produced by galaxies/quasars; total UV flux at z is determined by number of sources at z. Galaxy mergers, with dynamical friction and tidal stripping Chemical enrichment Luminosity evolution of stellar populations Can therefore consider not just a mass function of survivors, but characteristics of galaxies

Results: Dwarf Galaxies in “Local Groups” (Benson et al): 

Results: Dwarf Galaxies in 'Local Groups' (Benson et al) Create 'Local Groups': large galaxies with many small satellites Galaxies that survive until z=0 did tend to be those that formed while the Universe was still neutral Galaxies with small circular velocities are preferentially destroyed by photoionization The remaining survivors are still numerous but not in conflict with observations: They are understandably difficult to observe Detection would require very, very deep imaging, or statistical analysis of overdensities of stars HI is a possibility but many had 105 Msun or less in neutral gas

Results: Dwarf Galaxies in “Local Groups” (Benson et al): 

Results: Dwarf Galaxies in 'Local Groups' (Benson et al) Velocity Function per central galaxy Dotted curve: previous model from 2000 Dashed curve: Benson et al. model without photoionization Solid curve: Benson et al. standard model Dot-dashed curve: model with fesc = 10% Photoionization is the main effect (dominates tidal effects) log (N (andgt; Vc) per central galaxy) log (Vc, km s-1) 1 1.5 2

Future Work: 

Future Work Further study of the star formation histories of observed dwarf galaxies. New observational tests for the redshift of reionization? Can evidence for environmental differences be found? Correlation of the properties of observed dwarfs with the predictions of simulated/modeled dwarfs. Increased mass andamp; spatial resolution within simulations and phenomenological models.

References: 

References Barkana, R., andamp; Loeb, A. 2006. astro-ph/0611541 Barkana, R., andamp; Loeb, A. 1999. astro-ph/9901114 Benson, A.J. et al. 2002, MNRAS: vol 333, p. 177 and p. 156 Bullock, J.S., Kravtsov, A.V., andamp; Weinberg, D.H. 2000, ApJ, 539, 517 Efstathiou, G. 1992, MNRAS, 256, 43P-47P Grebel, E.K. andamp; Gallagher III, J.S. 2004, ApJ, 610, L89 Hoeft, M. et al. 2006, MNRAS, 371, 401 Loeb, A. 2006. astro-ph/0603360 Shaviv, N.J., andamp; Dekel, A. astro-ph/0305527

authorStream Live Help