Chemodynamical models of barred galaxies - Hervé Wozniak

Jul 10, 2006 - λ=1.4 (Kennicutt 1990). ❑ SFE = 0.1. ❑ creation of new stellar particles (remnants). ▫ feedback (instantaneous recycling):. ❑ SNII. ❑ mechanical ...
3MB taille 3 téléchargements 48 vues
Chemodynamical models of barred galaxies

Hervé Wozniak1 & Léo Michel-Dansac2 1 Centre

de Recherche Astronomique de Lyon, France 2 Instituto de Astronomia, Ensenada, UNAM, Mexico 2006 july 10th

Eskridge et al 2000 AJ 119, 536

Jogee et al. 2004 ApJ 615, L105

2006/07/10

Sheth et al. 2003 ApJ 592, L13

2

Why studying barred galaxies? „

„

Bars are the main drivers of evolution of disc galaxies between major merger events A stellar bar is able to: ‰ ‰ ‰ ‰

„

Bring gas from the disc to the central region Enhance star formation Reshape the stellar disc Homogenize distribution of elements

Strong constraints on dark matter distribution ‰

‰

Slowdown rate of bar rotation is linked to halo concentration (dynamical friction ⇒ angular momentum bar → halo exchanges) z=0 bars slow of fast rotators ?

2006/07/10

3

Observation/simulation duality Observation of stellar bars „ Length

N-body simulations (Theory) „ Corotation radius

Relationship

„

Axis ratio (ellipticity)

„

Strength

„

1 ≤RCR/Rbar≤1.7 (obs) or 2.6 (sim) ~ correlation (trend)

„

Luminosity

„

Mass

„

M/L

„

Line-of-sight kinematics

„

3D kinematics

„

Dark matter

„



„



„



2006/07/10

„

4

Comparing simulations with observations Athanassoula & Misiriotis 2002 MNRAS 330, 35 Bar lengths

2006/07/10

Mass distribution (N-body stars+DM)

5

(simple) Chemodynamical simulations „ „ „

N-body = PM scheme hydro = SPH star formation (instantaneous): ‰ ‰ ‰

„

feedback (instantaneous recycling): ‰ ‰

„ „

λ=1.4 (Kennicutt 1990) SFE = 0.1 creation of new stellar particles (remnants) SNII mechanical fraction of energy = 0.1

Maeder’s yields (1992) → metallicities cooling with solar abundances (Bohringer & Hensler 1989)

2006/07/10

6

Simulations parameters & calibration „ „ „ „ „ „

5-6 105 stellar particles (initial pop.) → ~ 1.2 106 at the end 1-5 104 SPH particles Mg / M¹ ~ 0.1 – 0.3 initial disc scalelengths : 3.5 to 5.5 kpc initial disc scaleheights: 0.5 to 1 kpc Photometric calibration ‰ ‰ ‰ ‰

SSP from Bruzual & Charlot (2000 release) Salpeter IMF 0.1 to 100 M~ Z of initial stellar pop = 0.004 initial age (4.4 to 10.4 Gyr) leads to z=0 at the end of the simulation

2006/07/10

7

10 kpc

16 kpc

Star formation rates Bar formation

2006/07/10

9

Measuring bar lengths… Ellipse-fitting methods: ---- Mass ---- B ---- K

max: used by e.g. Jogee et al. 2004 plateau: Michel-Dansac & Wozniak 2006 drop: used by e.g. Sheth et al. 2003

max

plateau

2006/07/10

Sheth et al. 2003 ApJ 592, L13

drop

10

…and ellipticities.

max ellipticity

Mass density

2006/07/10

B

B + absorption

K

K + absorption

---- Mass ---- B ---- K

11

Bar strength and ellipticity evolution

max ellipticity

---- Mass ---- B ---- K Combes & Sanders 1981 :

Qb

2006/07/10

12

Bar strength – ellipticity correlation Laurikainen et al. 2002 MNRAS 331, 880 2MASS galaxies

From MD&W 2006 simulations • Mass •B •K

2006/07/10

13

Resonances in barred galaxies ‰ ‰ ‰

„

corotation radius (Rcr) ‰

„

„

Ωb: pattern speed Ω: circular orbit frequency κ: radial epicyclic frequency

Ω (Rcr) = Ωb

inner Lindblad resonance radius Ω (Rilr) - Ωb = κ/2 ultra-harmonic resonance (Ruhr) ‰

Ω (Ruhr)- Ωb = κ/4

2006/07/10

14

Resonances in barred galaxies corotation

ILR

UHR

For each snapshot: • froze the potential • compute of orbits for several dynamical time ⇒ orbital frequencies (Ω,κ) κ ⇒ families of resonant particles ⇒ resonances location

Ωb 2006/07/10

Ω 15

Bar length and resonances evolution corotation ↔ ellipticity drop

---- Mass ---- B ---- K

UHR ↔ PA plateau

Radius (kpc)

ILR ≠ max ellipticity

Time (Gyr) 2006/07/10

16

Fast or slow bars ? Aguerri et al. 2003 MNRAS 338, 465 Rautiainen et al. 2005 ApJ 631, L129 PA plateau

2006/07/10

ellipticity drop

Rcr = 1.4 Rbar

Rcr = Rbar

17

Conclusions „

Comparison observation/simulation : ‰ ‰

„

Bar strength / ellipticity ‰

„

need to calibrate simulations use of chemodynamical simulations

spread in Qb vs ellipticity is due to evolution

Fast or slow bars? ‰

depends strongly on bar length estimator

2006/07/10

18