Evolutionary Population Synthesis models

Download Report

Transcript Evolutionary Population Synthesis models

Advanced Lectures on Galaxies (2008 INAOE): Chapter 4
Evolutionary Population
Synthesis models
Divakara Mayya
INAOE
http://www.inaoep.mx/~ydm
What do we try to synthesize?
Observed quantities (spectrum,
colors, Luminosity etc.) from a
region of a galaxy which consists of
Stars: emit light
Dust : absorb and re-radiate
Gas : ionize and re-radiate
In general the three components
are mixed even for parsec size
regions such as the Super Star
Cluster R136.
Mayya
What do we try to synthesize?
The aim is to obtain the ages and masses of all important
stellar groups the in a given region, by comparing the observed
quantities with the theoretically computed quantities.
The region in study may be as simple as an old
globular cluster (GC) or as complex as a starburst
in an interacting galaxy such as the Antennae.
GCs are relatively simple --- all the stars are
of the same age, hardly any gas and dust
Starburst systems are complex --- Age spread
- Metallicity spread
- In-homogenous dust distribution
- Underlying background
Mayya
The Simplest model
Simple Stellar Populations (SSP)
or Instantaneous Bursts (IB)
Stars:
Total Stellar Flux = Number of living stars * Flux of each star
- all the living stars have the same age and metallicity
- mass distribution is power-law (Salpeter IMF)
Dust: Correct the observed fluxes using a derived extinction assuming
foreground dust model and an extinction curve (Cardelli et al. 1989)
Gas: Add the fluxes calculated from photo-ionization models
for an HII region to the synthesized stellar fluxes (Osterbrock’s text)
SSP: Basic equations and Ingredients
Mayya
SSP: Basic equations and Ingredients
Mayya
SSP: Basic equations and Ingredients
Mayya
SSP: Basic equations and Ingredients
Stellar Evolutionary tracks (Isochrones)
- Geneva
- Padova
Stellar Atmospheric models
- Kurucz (LTE) models
- Observed stellar spectra
Uncertainties:
-Mass-loss rates?
-Rotation?
Uncertainties:
-non-LTE effects?
-Hot star models
Mayya
SSP: The method
Isochrone Interpolation schemes
Mayya
SSP: The method
Effect of rotation: rotating (_____)
non-rotating (---)
Mayya
SSP: The output
1. Nebular Lines
2. Continuum band luminosity
Mayya
SSP: The output
3. Colors and equivalent widths
- U-B, B-V, V-K etc.
- EW(Ha), EW(Hb) etc.
4. Selected spectral features
- CaT from RSGs
- Broad 4686 from Wolf-Rayet
5. Radio continuum
- Thermal flux from HII region
- Non-thermal flux from SNRs
6. Far-infrared continuum in dusty galaxies
- Bolometric luminosity
7. Mechanical energy
- Power from stellar winds and SN explosions
Mayya
(Class II) SSP: Results
Discussion of the paper Sec. 3: Dependence of SSP evolution
with input parameter, comparison with observations etc.
Mayya
(Class III) SSP: observable phases
1. Nebular ( < 6 Myr)
: Emission lines
2. Wolf-Rayet (3-5 Myr)
: HeII 4686 broad spectral feature
3. Red Supergiant (7-20 Myr): Calcium Triplet in absorption
4. A-star (50-500 Myr)
: Balmer lines in absorption
5. Intermediate (0.5-2 Gyr) : Balmer and CaII H and K line ratios
6. Old population (>2 Gyr): 4000 Ang break and other Lick indices
Mayya
SSP: spectral evolution
Mayya
Continuous Star formation (CSF) vs IB:
Ionizing photons
Mayya
CSF vs IB: Magnitude
Mayya
CSF vs IB: colors
Mayya
CSF vs IB: SED
Mayya
Deriving Age and Mass: diagnostic diagrams
1. Color vs Color
: age/extinction
2. Magnitude vs Color: age/extinction and mass
3. EW(Ha) vs Color : age and extinction
4. Spectral fitting
: age and extinction
5. Lick Indices
: age/metallicity
Mayya
CSF vs IB: RSG features
Mayya 1997
Mayya
The real case:star formation history of
starburst nuclei
Mayya
The real case: star formation history of
starburst nuclei
Mayya
Star formation history of M82 disk
Mayya et al. (2006)
Mayya
Other applications: SFR
Kennicutt 1998
Mayya
Other applications: Galaxy
formation and evolution
1. Fossil analysis (MOPED)
2. Integrated approach (GRASIL)
Mayya