Padova 033D Spectrography
3D Spectrography:3D Spectrography:
II - The tracersII - The tracers
Morphology: distribution of each component
Dynamics: kinematics via the emission or absorption lines
Line strengths: allow to study stellar populations
Padova 033D Spectrography
The different tracers: The different tracers: GasGas
90% H, 10% He
Neutral, ionized, molecular
H
He
Poussière
5 109 0.1 – 10 100 - 1000
100 - 1000 103 - 104 10 000
1 – 5 109 105 - 106 103 - 105 10
5 107 40
HI
HII
H2
Dust
Mass Cloud TDensity
Msun Msun (K)cm-3
Orion
H
He
Dust
Padova 033D Spectrography
HI GasHI Gas
Hyperfine transition line at 21 cm
Rare transition, but very abundant gas
Aligned poles(higher energy)
Opposed poles (lower energy)
Padova 033D Spectrography
HI Gas = Radio astronomyHI Gas = Radio astronomy
Padova 033D Spectrography
HI Gas - CartographyHI Gas - Cartography
Padova 033D Spectrography
HI GasHI Gas
Velocity profiles
Sofue et al.
Padova 033D Spectrography
HI GasHI Gas Position-Velocity diagram
Padova 033D Spectrography
HI Gas - KinematicsHI Gas - Kinematics
NGC 253 – HI Observations
Koribalski et al.
Padova 033D Spectrography
Ionized gas: HIonized gas: H
Spectrum in the visible
Padova 033D Spectrography
Ionized gas: HIonized gas: H Comparison HI / H
Padova 033D Spectrography
Ionized gas: HIonized gas: H
Velocity map
Khoruzhii et al.
Padova 033D Spectrography
StarsStars
Absorption lines
Deconvolution:
G = S* LOSVD GG = S*S* LOSVDLOSVD
LOSVD :
Line Of Sight
Velocity Distribution
Deconvolution:
G = S* LOSVD GG = S*S* LOSVDLOSVD
LOSVD :
Line Of Sight
Velocity Distribution
LOSVD
template
galaxy
Calcium triplet
V [km/s]
[ang]
Padova 033D Spectrography
StarsStars Problems due to population differences
(template mismatching)
Deconvolution: G = ii Si* LOSVDi
GG = ii SSi* * LOSVDLOSVDi
Different populations=
Different kinematics
Padova 033D Spectrography
LOSVDs and LOSVDs and kinematicskinematics
Many different methods for deconvolving:– Direct pixel fitting– Fourier fitting– Cross-correlation techniques– Fourier Quotient Correlation method– Others…
Fittings LOSVD moments:– Gauss-Hermite moments
(van der Marel & Franx 93, ApJ 407, 525
Gerhard 93, MNRAS 265, 213)
Padova 033D Spectrography
LOSVDs and LOSVDs and kinematicskinematics
LOSVDs of NGC 5582
Halliday et al., 2001, MNRAS, 326, 473
Padova 033D Spectrography
LOSVDs and LOSVDs and kinematicskinematics
Halliday et al., 2001, MNRAS, 326, 473
Padova 033D Spectrography
How to determine the age and composition of a galaxy?
• Assume 1 age and uniform composition.
• Assume same laws of physics as in a globular cluster.
• Stellar evolution: artificial HR diagram
• Find matching spectra
• Add these spectra composite galaxy spectrum
• Repeat previous steps for different ages/metallicities
• Determine best fit
Padova 033D Spectrography
The Lick System of Indices
Determining age and metallicity in practice
• Determine strengths of absorption features• Correct them for velocity broadening of the galaxy• Compare them with theoretical line strengths
Padova 033D Spectrography
Stellar population modelsStellar population models
Vazdekis (1999) models atLick resolution (~9 Å FWHM) based on Jones (1999) library
[MgFe52]=(Mgb x Fe5270)^0.5
Padova 033D Spectrography
Age & metallicity Age & metallicity for Fornax galaxiesfor Fornax galaxies
Kuntschner 2000, MNRAS, 315, 184
Padova 033D Spectrography
Aperture spectroscopy
Velocity, velocity dispersion …
Padova 033D Spectrography
Long-slit spectroscopyLong-slit spectroscopy
Kinematical profiles
Padova 033D Spectrography
We obtain a spectrum at each position
Integral field spectroscopyIntegral field spectroscopy
Padova 033D Spectrography
And each spectrum gives:
IFU spectroscopyIFU spectroscopy
FluxVelocityLine StrengthDispersion
Padova 033D Spectrography
H
V
Mgb Fe5270
NGC 3384 S0 NGC 3384 S0 (cluster)(cluster)
Padova 033D Spectrography
Line-strength maps – N3384Line-strength maps – N3384
No H gradient Strong Mgb in centre Fe peaks in centre
Restricted wavelength range
de Zeeuw et al. 2002, MNRAS, 329, 513
Padova 033D Spectrography
3D Spectrography:3D Spectrography:
Adaptive 2D BinningAdaptive 2D Binning
Padova 033D Spectrography
Photometry binningPhotometry binning
NGC4342WFPC2
Cappellari 2001: Efficient MGE fitting method
Padova 033D Spectrography
Spectroscopy 1D-binningSpectroscopy 1D-binning
IC1459
Major axis kinematics
Cappellari, Verolme et al. 2002
Padova 033D Spectrography
The SAURON test data:The SAURON test data:NGC 2273NGC 2273
Result of multiple pointings:• irregular domain• vertical S/N jumps
S/N mapReconstructed image
Barred Sa galaxy
Padova 033D Spectrography
2D-binning requirements2D-binning requirements Topological: partition the plane without holes or
overlapping bins Morphological: bins as compact or “round” as
possible Uniformity: minimal S/N scatter
Padova 033D Spectrography
Tiling of the planeTiling of the plane
20 10 0 10 20
20
10
0
10
20
10 0 10
10
0
10
10 0 10
10
0
10
Towle 2000
Penrose tiling
Padova 033D Spectrography
2D-binning by QuadTree 2D-binning by QuadTree decompositiondecomposition
Regular cells but: • large S/N scatter • border problems
2x
Satisfies Topological and Morphological requirements
Padova 033D Spectrography
Voronoi TessellationVoronoi Tessellation
Satisfies Topological requirement ONLY
Definition: each point in a bin is closer to its generator than to any other point
Padova 033D Spectrography
Taking pixels into accountTaking pixels into account
1D case: growing bins along the slit
2D analog: growing bins around the bin baricenter
Padova 033D Spectrography
Centroidal Voronoi Centroidal Voronoi TessellationTessellation
All Topological, Morphological and Uniformity requirements satisfied!Cappellari & Copin 2002
It is the perfect solution in the case of Poissonian noise and many pixels.
Padova 033D Spectrography
Voronoi TesselationVoronoi Tesselation2D-binning for2D-binning for
NGC 2273NGC 2273
• Small S/N scatter• Compact bins• No border problems
Padova 033D Spectrography
NGC 2273 NGC 2273 stellar mean velocity field stellar mean velocity field
2D-binned velocity Not binned
Padova 033D Spectrography
What to keep in mindWhat to keep in mind
Ionized gas and stars are (easily?) traceable via emission and absorption line spectra.
Derivation of the distribution, kinematics and line strengths.
Again, a two-dimensional spatial coverage is often critical for the scientific interpretation
More importantly: it is the link between all these tracers which allows us to really understand the physical status of these objects, leading to a theory of their formation and evolution.
Further readings:– Galactic Astronomy, Binney & Merrifield,
Cambridge University Press