Modeling the Extended Structure of the Carina Dwarf Spheroidal

1
Modeling the Extended Structure of the Carina
Dwarf Spheroidal
Ricardo Muñoz (University of Virginia)

• In collaboration with
• Steven Majewski (University of Virginia)
• Kathryn Johnston (Wesleyan University)

2
Goal of this talk

• Readdress the question (one more time) to what
  extent Galactic tides may be affecting dSph
  galaxies
  
• Carina in particular, most compelling case for
  tidal stripping besides Sgr.
  

• More specifically
• Are one-component, mass-follows-light models
  really inappropriate?
  
• Have they really been ruled out?
• Under what circumstances (if any) can they
  reproduce the salient properties of Carina?
  

3
Motivation

• MW-satellite interactions have been modeled (many
  times) to try and understand how they would
  affect the internal kinematics of dSphs.
  
• Weve learned that

• If satellite is destroyed or near destruction,
  then artificial inflation of M/L is observed
  (Kroupa 1997).

• If satellite retains a significantly bound core,
  the central velocity dispersion does reflect the
  instantaneous mass content (Oh, Lin Aarseth
  1995, Piatek Pryor 1995).

• Velocity gradient (apparent rotation) is main
  indicator of tidal stripping (e.g., Piatek
  Pryor 1995, Read et al. 2006)

4
Motivation
Muñoz et al., 2006
Carina
Sculptor
Westfall et al. 2006
Draco
Muñoz et al. 2005
Ursa Minor
Muñoz et al. 2005
Leo I
Sohn et al. 2007, ApJ, in press
Sag.
Majewski et al. in prep. Ibata et al. (1997).
rlim
5
Motivation

• Not what it is expected from one-component,
  mass-follows-light model not being tidally
  disrupted!

Gilmore et al. 2007
6
Motivation

• Problem with flat velocity dispersion profiles?
• Tidal stripping can cause them, but

• Apparently no clear signs of tidal stripping in
  dSphs? For example regular shapes, no tidal
  tails, lack of rotation
  
• then need second mechanism.

• Solve Jeans equations and get underlying mass
  distribution using some assumption for
  anisotropy
  
• extended dark matter halos.

7
The case of Carina

• Over the last years there has been significant
  effort to measure RVs of stars at large radii.

• To date Carina is system with most extended
  radial coverage (both photometric and
  spectroscopic) and observations indicate some
  degree of tidal stripping.

8
Carina
Muñoz et al. 2006
9
Our Simulations
-Milky Way Miyamoto-Nagai 1975 disk, Hernquist
spheroid and logarithmic halo. Parameters from
Law, Johnston Majewski 2005
-Satellite 105 particles. Single-component
(mass-follows-light), Plummer distribution.
10
Our Simulations

• Wide exploration of parameter space
• Mass 1 x 106 to 3 x 108 Mo
• Scale length 90, 194, 280 and 480 pc
• Orbit Nearly Circular to 0.9 eccentricity

Satellite is required to have current RV of
Carina and be at same distance and position.
200 Simulations for Carina, 100 for Leo I
11
Results for Carina
Muñoz, Majewski Johnston, in prep.
12
Results for Carina
-Best Models Orbital parameters similar to the
ones found by Piatek et al. (2003) from proper
motion measurements.
13
Results for Carina

• Best Model properties
• - Rather eccentric orbit
• (apoperi 10015 kpc)
• - Initial mass 3.8 x 107 MSUN
• - Mass Loss Rate 10 orbit-1

M/L 40 (M/L)SUN In agreement with result
s
from Core-Fitting technique!!
14
Very little stripping can cause flat/rising
velocity dispersion profile
Result for Leo I

• Best Model properties
• - Very eccentric orbit
• (apoperi 40013 kpc)
• - Current mass 4.0 x 107 MSUN
• - Mass Loss Rate 2 Gyr-1

M/L 5 (M/L)SUN In agreement with resu
lts
from Core-Fitting technique!!
Sohn et al. 2007, ApJ, in press
15

• Mass-follows-light?

• Johnston and Bullock simulations

When DM halo is stripped to the point that
stars are stripped, then system behaves like MFL.
16
Lack of rotation can be deceiving
Other General Results
No velocity gradient?
17
Lack of rotation can be deceiving
Other General Results
18
Lack of rotation can be deceiving
Other General Results
19
and appearances can be deceiving
Other General Results
20
and appearances can be deceiving
Other General Results
21

• Conclusions

• A dark matter-dominated, MFL system undergoing
  tidal stripping provides a very good match to all
  the observed properties of Carina (and Leo I).
  Premature to rule out MFL

• Very little tidal stripping can greatly affect
  velocity dispersion profile. Remove outliers
  (Klimentowski et al. 2006)

• Apparent rotation not convenient as a tidal
  disruption indicator (if limited to King radius)

• Sensitivity of current photometric and
  spectroscopic surveys may not be enough to find
  signatures of tidal stripping. MORE
  DATA

22

• Questions?