Simon White and the Seven Dwarfs

1
Simon White and the Seven Dwarfs

• Matthew G. Walker University of Michigan
• mgwalker_at_umich.edu
• Collaborators
• Mario Mateo University of Michigan
• Edward Olszewski University of Arizona,
  Steward Observatory
• Rebecca Bernstein University of Michigan
• Oleg Gnedin University of Michigan
• Bodhisattva Sen University of Michigan, Dept.
  of Statistics
• Xiao Wang University of Maryland, Baltimore
  County
• Jayanta Kumar Pal University of Michigan,
  Dept. of Statistics
• Michael Woodroofe University of Michigan,
  Dept. of Statistics
• Jim Joyce University of Michigan, Dept. of
  Philosophy

Astrophysical Probes of the Nature of Dark
Matter March 23, 2007, UC Irvine
2
(No Transcript)
3
(No Transcript)
4
A New Era in dSph SpectroscopyLarge Telescopes
Multi-object Spectrographs

• VLT GIRAFFE, UVES,
• Keck DEIMOS
• WHT WYFFOS
• MMT Hectochelle
• Magellan MMFS

See work by Mateo, Olszewski, Walker et
al. Gilmore, Wilkinson, Kleyna, Read et
al. Majewski, Munoz, Sohn et al. Martin et al.
5
Magellan MMFS

• 256 fibers over 30 arcmin field
• Magnesium Triplet 5140-5180 Angstroms at
  resolution 20000-25000
• /- 2 km/s velocities for V20.5 star in 2.5
  hours
• Fe/H to /- 0.2 dex
• Up to 750 spectra per night

6
Velocity and Spectral Line Strength
7
Mg Line Strength vs. Velocity
8
Maps New Magellan Sample
9
Maps New MMT Sample
Draco
Leo I
Leo II
10
Summary of Kinematic Data
Ntot obtained by adding in published samples by
Mateo et al. (1991, 1993, 1998) Munoz et al.
(2006) Kleyna et al. (2002, 2004) Hargreaves et
al. (1996) Vogt et al. (1995)
11
Things Worthy of Mention, Before I Ignore Them
Radial variations in stellar populations (e.g.,
Harbeck et al. 2001 Tolstoy et al. 2004
Battaglia et al. 2006)
Tidal Effects (e.g., Munoz et al. 2006, Sohn et
al. 2006, Mateo et al. 2007)
Kinematic substructure (e.g., Kleyna et al. 2003,
2004 Walker et al. 2006)
Also, See Talks by Majewski, Gilmore, Read,
Munoz,
12
Velocity Dispersion Profiles
13
Velocity Dispersion Profiles
14
Velocity Dispersion Profiles
NFW constant anisotropy M_vir 108 109
M_sun ß -0.5
15
Velocity Dispersion Profiles
NFW constant anisotropy M_vir 108 109
M_sun ß -0.5 Core ? a 1(r/r0)3/2-2 V_max
15 25 km/s rcore tens of pc Truncated NFW ?
a r-1exp(-r/rb) (Kazandzitis et al. 2004) M_tot
108 109 M_sun rb 300-500 pc NFW variable
anisotropy (Osipkov-Merritt) M_vir 108 -109
M_sun ra ? 8
16
Nonparametric Mass Estimation (Wang et al. 2005)

• Assumptions
• Spherical symmetry
• Dynamical equilibrium
• Velocity isotropy
• Parametric model
• Mass follows light
• Jeans Equation
• where
• Estimate f(r) and µ(r) separately
• f(r) as a step function, recover from star count
  data
• M(r) as a cubic spline subject to shape
  restrictions

17
Mass Profiles
18
One Size Fits All?
Solid lines are NFW-like (Navarro et al. 2004
Gnedin et al. 2004) with Mvir (2-4) x 109
M_sun ß-0.3 (Leo I has ß -1, Leo II has ß0.3)
19
M(rlt750pc) vs. Luminosity
20
M(rlt750pc)/Luminosity vs. Luminosity
21
Summary

• New spectra of 6000 dSph targets, 4900
  members
• Flat velocity dispersion profiles
• ltv2gt1/2 10 /- 3 km/s
• Despite span of several magnitudes in baryonic
  mass
• M(rlt750pc) 108 M_sun

22
(No Transcript)
23
Dwarf Spheroidal Galaxies
Dra
Car
Leo II
For
Leo I
UMi
Sex
Sgr
CMa
UMa
Boo
Scl
24
Mass Profiles (sorted by model)
25
(No Transcript)
26
(No Transcript)
27
Equilibrium vs. Tidal Disruption

• Tidal disruption simulations
• Apparent rotation about minor axis
• Major axis aligned with proper motion vector
• Non-decreasing velocity dispersion profile

Read et al. (2005)
Piatek Pryor. (1995)
28
Equilibrium vs. Tidal Disruption Streaming
Walker et al. (2007)
29
Equlibrium vs. Tidal Disruption Streaming
Walker et al. (2007b)
30
(No Transcript)
31
(No Transcript)
32
Dwarf Spheroidal (dSph) Galaxies

• Most galaxies are dwarfs
• 15 dSph satellites of the Milky Way Why not
  more?
• L 105-7 Lsun
• ancient stars How did dSphs form stars at
  all?
• intermediate/young stars How did dSphs
  retain/recapture gas?
• Stellar speeds of 10 - 20 km/s
• R 500 3000 pc Dark Matter or tidal
  disruption?
• Virial theorem ? M/L 10 1000
• Resolved Stellar populations
• CMDs SFH?
• Individual stellar spectra Mass Profiles?
• Substructure?

Klypin et al. 1999
33
Mass Profiles Non-parametric Mass Estimation
Wang et al. (2007)
34
(No Transcript)
35
(No Transcript)
36
(No Transcript)
37
(No Transcript)
38
Kinematic Substructure Velocity Dispersion
Surface
Walker et al. (2006b)
39
Fe/H Metallicity
40
(No Transcript)
41
(No Transcript)
42
(No Transcript)
43
(No Transcript)
44
(No Transcript)
45
Dark Matter From Simulations
Center for Cosmological Physics (U. Chicago)
Klypin et al. 1999
46
Dwarf Spheroidal Galaxies
Dra
Car
Leo II
For
Leo I
UMi
Sex
Sgr
CMa
UMa
Boo
Scl
47
Dependence on Sample Size
N100 stars
N400
N700
N1000
48
Distinct Kinematic Populations in
Sculptor(Tolstoy et al., 2005)
49
Equilibrium Models 3 Jeans Analysis
Strigari et al. (2006)
50
(No Transcript)
51
Equilibrium Models 4 Nonparametric Models

• Spherical symmetry
• Equilibrium
• Parameterized moments of DF
• Velocity isotropy
• Mass follows light
• Luminous Dark component (dynamically coupled)
• E.g. Wang et al. (2005)
• Mass nondecreasing with r
• Density decreases with r

52
Equilibrium Models 4 Nonparametric Models
v_disp(r)
M(r) and v_circ(r)
53
Non-Equilibrium Models Tidal Disruption
Walker et al. (2006)
54
Issues
55
Issue 1 Foreground Contamination

• Improve decontamination algorithms (using stellar
  positions and foreground model)
• Washington photometry
• Model the DF with foreground component

56
Issue 2 Distinct Stellar Populations
Tolstoy et al. (2004)

• Need to determine metallicity information from
  MMFS spectra
• Calibrate RGB photometry

57
Issue 3 dSph Substructure

• Complicates equilibrium analysis
• Interesting in its own right
• Origin and evolution of dSphs
• Survival of substructure constrains halo models
• Disruption
• Dynamical friction

Coleman et al. (2004)
58
Velocity Dispersion Surface
Walker et al. (2006)
59
Recovering f(r) from its Projection

• Let
• projected density gS(s) relates to 3-D density by
• Let
• We estimate GS directly from star counts
• Treat f as step function
• for

60
Substructure
Coleman et al. (2004 2005)
61
(No Transcript)
62
(No Transcript)
63
(No Transcript)