Dwarf Dark Matter Halos in CDM cosmology : small scales, big problems

1
Dwarf Dark Matter Halos in CDM cosmology small
scales, big problems

• Oleg Gnedin (Ohio State University)

and the CDM community
2
Open Questions in Dwarf Galaxy Formation

• Is the missing satellites problem still a
  problem? (can we
  reproduce dwarf galaxy numbers and spatial
  distribution?)
• What can we learn about star formation in the
  young universe from nearby dwarfs? (do we
  understand stellar populations, star formation
  histories?)
• Do present-day dwarf galaxies resemble
  high-redshift galaxies? (how big and extended are
  dark matter halos of dwarf galaxies now?)
• What other tests of cosmology can be done using
  dwarf galaxies? (are central regions of dwarf
  halos cuspy or cored? can stellar feedback
  change that?)
• New opportunities for probing dark matter halos

3
Primordial gas can cool in all halos where T
104 K
Vc 10 km/s 100 km/s
HI
HeII
H2
(figure from Barkana Loeb 2001)
4
The Missing Satellites Problem
Kravtsov, OG, Klypin (2004)
5
Model of Dwarf Galaxy Formation Dynamics Gas
Physics Kravtsov, OG, Klypin 2004

• halos accrete gas while isolated (filtering
  mass approximation)
• gas distribution is exponential with scale
  radius ? ?
• tidal truncation is determined by maximum of
  Ftid
• Schmidt law of star formation with a threshold
  d?/dt ? ?g1.4 for ?g 5 M? pc-2
• if Itid 100 Gyr-2, starburst mode up to 50
  of gas converted into stars
• stellar populations are heated by external
  tidal field
• Disclaimer the model is based on the locally
  observed star formation law and can be modified
  to explore different scenarios at high redshift

6
Tidal interactions truncate and thicken disks
Tidal stripping is determined by Tidal heating
is determined by
in clusters of galaxies OG 2003
7
Three types of satellite halos
8
Our Answer to the Missing Satellites Problem
Kravtsov, OG, Klypin (2004)

• Subsequent results in agreement with
  observations
• dSph vs. dIrr radial distribution, including a
  dSph (Tucana) galaxy 1 Mpc away
• anisotropic distribution on the sky (Zentner,
  Kravtsov, OG Klypin 2005)

9
Star formation histories of dwarf galaxies

• All of the luminous satellites formed (a few)
  stars before the reionization (z 6)
• afterwards, some were massive to retain their
  gas and kept forming stars (episodically) dIrr
  dSph
• some were not massive enough and lost all of
  their gas due to photo-evaporation fossil dSph

Model predictions agree qualitatively with the
varied star formation histories of the Local
Group dwarfs
Dolphin et al. 2005
10
Dwarf irregular galaxies would keep forming stars
in the regions where the gas density rises above
the threshold (more stochastic and centrally
concentrated with time)
?
?crit
r
11
Morphological mix of the Local Group galaxies
based on vrot / ?
12
Draco dSph small and truncated?
circular velocity
velocity dispersion
(Wilkinson et al. 2004)
13
Center of Fornax dSph core or cusp?
Strigari, Bullock, Kaplinghat, Kravtsov, OG,
Abazajian, Klypin, in prep.
GC
Goerdt, Moore, Read, Stadel, Zemp (2006)
globular clusters in Fornax would have spiraled
into the center by dynamical friction is there is
a cusp, but not if there is a core
14
Maximum Stellar Feedback
OG Zhao (2002)

• Simulations of galaxy formation within CDM
  predict density cusps in centers of all dark
  matter halos ? ? r 1
• Observations of dark matter-dominated LSB
  galaxies indicate constant density cores (based
  on fitting rotation curves)
• Can powerful outflows resolve the problem of dark
  matter cusps?

Maximum expansion of a dark matter halo is
achieved when feedback is very efficient mass
of young stars is negligible gas
removal timescale is shorter than the dynamical
time
if Rd ? rs heating proportional to binding
energy Md/Rd if Rd ?? rs energy input
proportional to disk mass Md
disk size Rd is a measure of the angular momentum
of the baryons ? 0.05, 0.01 (-3??), 0.0025
(-6??)
15
Stellar feedback cannot convert dark matter
cusps into cores
Dark matter density increases in response to disk
formation Got Contra? http//astronomy.ohio-state
.edu/ognedin/contra/
16
Galactic satellites align along the major axis
of DM halo

• Zentner, Kravtsov, OG Klypin (2005)
• triaxial halo is a generic prediction of CDM
  models
• close to the disk, baryons affect the halo
  shape
• at large distances, the halo is expected to be
  triaxial

17
HyperVelocity Stars as Probes of Halo Shape
most likely ejected from the Galactic Center with
1000 km/s (three-body interaction with the
supermassive black hole) all
discovered in 2005, probe halo potential to 75
kpc
SDSSJ090745024507 Vr 853
km/s SDSSJ093321441705 Vr 708
km/s HE0437-5439 Vr 723
km/s SDSSJ091301305120 Vr 603
km/s SDSSJ091759762238 Vr 543 km/s
18
Measuring proper motions of HVS can determine
the shape and orientation of the dark matter
halo
HVS1
HVS2
OG, Gould, Miralda-Escude, Zentner 2005
19
Constraints on the halo axis ratios
from one HVS there is distance-triaxiality
degeneracy
From two HVS strong constraints and distance
determination
20
Summary

• The abundance, spatial distribution, and star
  formation histories of dwarf galaxies in the
  Local Group can be reproduced using local star
  formation laws and dynamical evolution of CDM
  halos.
• Type I. Some dwarfs were significantly more
  massive at z2 and had extended star formation
  histories, unaffected by reionization (dIrr
  dSph)
• Type II. Some (dSph fossils) only formed stars
  before reionization
• Type 0. Most satellite halos remained forever
  dark (HVC?)
• Draco dSph appears to be significantly
  truncated
• Fornax dSph appears to have a very large core,
  based on GCs (?)
• Stellar feedback cannot turn cusps into cores
• Hypervelocity stars open a new window to probe
  halo structure