Dwarf spheroidal galaxies and the physical

properties of dark matter

Mark Wilkinson

University of Leicester

Gerry Gilmore (Cambridge) Jan Kleyna (Hawaii)Andreas Koch (UCLA) Wyn Evans (Cambridge)Eva Grebel (Heidelberg)Rosemary Wyse (JHU)Justin Read (Zurich)Vasily Belokurov (Cambridge)Dan Zucker (Cambridge)

Collaborators

Local Group satellites

Grebel (1999)

Satellites and dark matter

Moore

Hot Warm Cold

• Low luminosity, gas-poor satellites of Milky Way and M31

•

• No well-defined tidal radii

• Individual stellar velocities measurable with ≥ 4m telescopes

•

• Core radii

• Inferred mass-to-light ratios in range

Dwarf spheroidal galaxies

Odenkirchen et al. (2001)

L = 3× 104L! − 2× 107L!

σ0 ∼ 7− 12 km s−1

r0 ≈ 130− 500 pc

3− 300 M!/L!

Velocity dispersion profilesDraco Ursa Minor Sextans

0 2 4

Carina

0 2 4

Leo I

0 2 4

Leo II

0 5 10 15

Omega Centauri

0 5 10 15

Mass follows light

Gilmore et al. (2007)

Velocity dispersion profiles II

Walker et al. (2007)

dSph dispersion profiles generally remain flat to large radii

Are dSphs in equillibrium?

!"#

!$#

%##

%!#

&'())*+,)-!./

!.0!.#!0#0.#.0

!$#

!$0

!1#

!10

%##

2&'(3))*+,)-!./

45678)59:-)85;:5<);:-=5>(?))*58(,:>/

@5-= A?-=

Koch et al. (2007)

• Velocity gradients are signature of tidal disturbance

• No dSph shows evidence of significant velocity gradient in inner regions ⇒ bulk mass estimates are robust

dSphs: the case for cored haloesJeans equations give simple relation between kinematics, the light distribution and the underlying mass distribution

We can either:

1. Assume a parameterised mass model and velocity anisotropy and fit dispersion profile

or

2. Use Jeans equations to determine mass profile from projected velocity dispersion profile and a fit to the light distribution

M(r) = −r2

G

(1ν

d νσ2r

d r+ 2

βσ2r

r

)

β(r)M(r)

Fitting dSph dispersion profiles: Leo I• Assume either NFW halo

( ) or more general profile ( )

• Best-fit dispersion obtained for cored profiles with roughly isotropic velocity dispersions

• Significant velocity anisotropy not favoured (but not excluded)

• Enclosed mass in both cored and cusped haloes

! !"# !"$ !"% !"& '!

(

'!

'(

)*+,-./

!**+,0*1*2/

34)5

"*6*!!"(

"*6*!

"*6*!"(

"*6*'"789):

!

(

'!

'(

;<=

"*6*!!"(

"*6*!

"*6*!"(

"*6*'

Koch et al. (2007)

∼ 8× 107M!

ρ ∝ r−1

ρ ∝ r−α

Density profiles from Jeans equations: assumptions

10 20 30 40 500.01

0.1

1

10

100

R (arcmin)

Draco

0 10 20 30 400

5

10

15

20

R (arcmin)

1 5 10 500.01

0.1

1

10

R (arcmin)

Carina

0 10 20 300

5

10

15

R (arcmin)

• Masses:

• Conclusion: dSph kinematics are consistent with cored haloes

Density profiles from Jeans equations

3− 8× 107 M! =⇒ M/L = 13− 240 M!/L!

r (kpc)

ρ(M

!pc

−3

)

Ursa Minor

Draco

LeoII

LeoI

Carina

Sextans1/r

10−1 10010−3

10−2

10−1

100

Gilmore et al. (2007)

• Majority of current data consistent with cored dark matter distributions and a mass scale (interior to light) of

• Mean dark matter density

• NFW halo:

Global trend of dSph haloes

!!"!!#!!$!%!&!$

$

!$!

!$#

!$'

()

*(+,- ./0/1)..+..*(+,- .0

23450

6(7

89:/4;<

=03;4:

85>?@/03

A437;4

,90B

,90BB

C;DBE

C;DBB

6(4

F00/9<

Mateo et al. (1998), Wilkinson et al. (2006), Gilmore et al. (2007)

∼ 3× 107M!

ρ(r < 10 pc) ≈ 60 M! pc−3(2 TeV/c2cm−3)

! 0.1 M! pc−3(5GeV/c2 cm−3)

Size distribution of stellar systems

Gilmore et al. (2007)

The next step - resolving the cusp/core issue with the VLT

! !"# $ $"#!

%

&

'

(

$!

$%

)*+,-./

!*012*+,3*2!$/

*

*

4156*!*721851-7.

492-*!*721851-7.

4156*!*5:;7:0

492-*!*5:;7:0

!!" !#" !$" " $" #" !"!"%&

!"%'

!"%(

!"%#

"

"%#

)*+,-*./0*-

!$1

.2*3,45*!*2*36-71*8*2*3,45

• Degeneracy between velocity anisotropy and inner density profile can be broken using ~500 radial velocities in the inner 0.2 kpc

• Program to apply this to Carina dSph currently underway at VLT

Conclusions

• Dwarf spheroidal galaxies are valuable laboratories for testing the properties of dark matter

• No kinematic evidence that tides have inflated central velocity dispersions of dSphs

• Current kinematic data are consistent with high M/L ratio, cored dark matter distributions - mass scale ( ):

• Mean dark matter density:

• In cusped halo:

• More data being obtained to place more robust constraints on density profiles

∼ 3× 107M!r ! 400 pc

ρ(r < 10 pc) ≈ 60 M! pc−3(2 TeV/c2cm−3)

! 0.1 M! pc−3(5GeV/c2 cm−3)