This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Modelling multiwavelength SEDs – tools for galaxy formation models
Plan:* Modelling SEDs - GRASIL- characteristics, aims and limitations- Fitting observed SEDs- Effects of different SED treatmens* Application to SAMs: GALFORM, ABC, MORGANA +GRASIL* Modelling SEDs with Artificial Neural Networks * SEDs for SPH: GRASIL3D
Laura Silva - INAF Trieste
Gian Luigi Granato, Andrew Schurer (INAF); Cedric Lacey, Cesario Almeida, Carlton Baugh, Carlos Frenk (ICC); Olga Vega (INAOE); Fabio Fontanot, Alessandro Bressan (INAF),
Pasquale Panuzzo (CEA)
Multi- SED modelling – ingredients & aims
*Stellar pop. synthesis
*SFR(t)+Mgas(t),Z(t) analytical, chemical evolution or galaxy formation models
* UV/optical attenuation and IR emission
T
SSP dttZtTLtSFRTL0
))(,()()(
Semi-empirical: attenuation curve for LIR
+ IR shape. Pros: non time consuming – analysis of large data sets. Cons: not great predictive power
Theoretical: Explicit computation of radiative transfer and dust emissionPros: broader interpretative/predictive power. Cons: time consuming
Modelling UV to radio SEDs with GRA(phite)SIL(icate)
Star- forming MCs
Diffuse dust
Extincted stars
1) Realistic and flexible SED modelling
Stars and dust in a bulge (King profile) + disk (double exponential)
Dust: big grains, very small grains and PAHs. Emission is appropriately computed for each component
Stars are born within MCs and gradually escape as a function of their age age-dependent extinction
UV-to radio SEDs2) Reasonable computing time
Radiative transfer exactly solved for opt thick MCs, with approximation in the cirrus (real bottle-neck)
Presence of symmetries
3 dusty environments: dense (star forming Molecular Clouds), diffuse (cirrus) ( clumping of stars and dust), dusty envelopes of AGB stars
Best fitting models:•Age-dependent extinction due to star forming Molecular Clouds and Cirrus (stellar age stratification in the disk wrt dust)
•Sequence of models with increasing dust content in the cirrus and age for thin-thick disk separation tthin=25-200 Myr
Sample of GALEX NUV-selected late type galaxies (Buat+, Iglesias-Paramo+)
MW
SMC •No age-dependent extinction•Sequence of models with increasing dust content (1m polar opt depth =0.05-6.4)•MW and SMC dust composition•2175A bump within NUV
UV Attenuation in spiral galaxies – role of age-dependent extinction (Panuzzo+2007)
Meurer e
t al 1
999
UV-bright s
b
SED analysis of ULIRGs (Vega+2008)
Sample of 30 nearby ULIRGs w MIR to radio data vs large grid of SF+GRASIL+AGN tori models
SF histories from the Semi-Analytical Model for galaxy formation MORGANA - SED by GRASIL (colored) and empirical [attenuation curve with C&F + slab] (hatched)
Fontanot, Somerville, Silva+09
Different treatments predict different SED for the same SFR(t):Attenuation
SAM + GRASIL
|||||||||||SAM+[C&F00 + slab]
Fontanot, Somerville, Silva+09
Fontanot, Somerville, Silva+09
MORGANA+template
MORGANA+GRASIL
(average SEDs for
low-z and high-z mock catalogues)
Different treatments predict different SED for the same SFR(t): IR
SF histories from the Semi-Analytical Model for galaxy formation MORGANA - SED by GRASIL (black) and templates (colored)
Different treatments predict different SED for the same SFR(t): IR
Effects of dust assumptions on SED (Schurer+09)
MW-type
Ellipticals
Irregular
Z from chem. model
Mdust/Mgas(t)
M*=10^12
M*=10^11
M*=10^10
Z
Z
Z
Representative SF for Spirals (MW-type), Ellipticals and Irr + evolution of C- and Si- based dust with assumptions on dust production (evolved stars, SN ejecta) and distruction efficiencies constrained by chemical abundances & dust depletion (Calura, Pipino & Matteucci 08)
model + MW ext
model + QSO ext
Mdust Z + MW ext
MW ext curve
QSO ext curve (Maiolino+04)
Young Elliptical model vs Balmer-break galaxies (Wiklind et al 2008)
MW ext curve
QSO ext curve (Maiolino+04)
Young elliptical model vs SHADES sources (Clements et al 2008)
model + MW ext
model + QSO ext
Mdust Z + MW ext
Computing SEDs in Semi-Analytical galaxy formation Models• SAM: DM with gravity-only N-body or MC, baryons with analytical recipes – compare with widest range of observed galaxy properties
• Associate to each mock galaxy its “real” SED but:complexities in treating radiative effects - unknown dust properties - computing time fundamental issue for cosmological volumes
• SAMs with theoretical SED:GALFORM+GRASIL(Granato+00,Silva+01,Baugh+05,Lacey+08,09) Anti-hier.BarionicCollapse+GRASIL(Granato+04,Silva+05,Lapi06) MORGANA+GRASIL(Monaco+07,Fontanot+07,09)
• Outputs: simulated catalogues of galaxies at different z slices; SFR(t), Mgas(t), Z(t), morphology, scale radii for stars & gas
Semi-empirical treatment: fix v (L or f(Mgas, Z)) + dependence + uniform distrib. of stars and dust in a 1D slab + IR templates
Aim: get downsizing within hierarchical assembly of DM to explain high-z massive galaxies & ell with SAM:*cooling gas in big halos at high-z start vigorous SF without setting in a disk*SFR promotes the development of SMBH from a seed, feedback of the QSO on the host to possibly quench SF
SCUBA 850 m
MAMBO 1200 m
model
data
K Band counts and z distribution
All sph
Passive sph
modelobserved
K20 SURVEYmass range required by sub-mm
counts
Extremely Red Objects (R-K)>5
passive
active
Modelling SEDs with Artificial Neural NetworksAlmeida+09, Silva+09
• Aim: computing SEDs with GRASIL but much faster (now: several minutes) Exploit the Millennium Simulation – a mock galaxy catalogue requires millions runs Improve on RT approximations Fast search for best fit parameters for large data sets
Lacey+09
L IR > 10^11 Lo L IR > 10^12 Lo S(100m)>2mJy
Spectral variance for a GALFORM + GRASIL catalogue
* Mathematical algorithms for data analysis, introduced to replicate the brain behavior: learn from examples
* It works!
• SEDs : complex, non-linear, high dimensionality and large variance functions of some galaxy properties
Why ANN:
Input: parameters determining the SED
Output: SED
ANNalgorithm(black box)
The ANN is trained to predict the SED from controlling parameters using a suitable precomputed training set (many sets of known input-output)
ANN & SED: 2 methodsGeneral use - very fast (Silva+09): input = physical quantities determining the SED of MCs and Cirrus – one single trained net for any application
dust:polar, equatorial, homog - ~ measure concentration of dust
R*/Rdust: ~relative position of * and dust
Hardness of radiation field: ~ MIR to FIR ratio
ANN vs GRASIL - Examples of single SEDs
M51
M82
ANN vs GRASIL with ABC SAM– randomly extracted SEDs
ANN vs GRASIL - ABCmock galaxies making
submm counts
ANN vs GRASIL – GALFORM z=0 catalogue
ANN vs GRASIL for ABC – comparison for galaxy counts
70m 100m 160m
350m250m 500m
SED and SPH galaxy models: GRASIL3D A.Schurer 09 PhD theses
Aim: exploit the spatial information for stars and gas in hydro simulations of galaxy formation and of observed images – requires no symmetries
GRASIL->3D: •generalised to an arbitrary geometry through a cube grid in which stars and gas particles output by the SPH are distributed• Gas in each cell divided in SF molecular clouds and cirrus (if young stars are present and gas density > threshold)• Intrinsic stellar SED in each cell, with young stars within MCs• Radiation field in each cell due to all other cells
Preliminary tests: z = 2, comparison to SCUBA galaxies
SUMMARY
• Multi-wavelength modelling as a tool to quantitatively interpret observations – make predictions and constrain galaxy formation models
• Different treatments predict different SEDs for the same SFR(t)-> necessity of a reliable computation of the SED for proper interpretations of observations and predictions of galaxy formation models
• The treatment of dust reprocessing of UV/optical in the IR requires a proper computation – time cosuming
• For large cosmological applications: promising solution with ANN