Chary: NOAO/NSO 50 th Anniversary 1/40 The Star Formation History of the Universe Star-formation History: <1960 : 7 (*) papers 1960-2010 : 3300 papers General Relativity: <1960: 400 1960-2010: 19000 Ranga Ram Chary Spitzer/Planck/IPAC California Institute of Technology x400 x40
40
Embed
The Star Formation History of the Universe · 2010. 5. 7. · Chary: NOAO/NSO 50th Anniversary 1/40 The Star Formation History of the Universe Star-formation History:
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.
Extragalactic Background Light ConstraintsUV estimates of SFRIR estimates of SFRSpitzer Measured Stellar Mass DensitiesGamma-ray Bursts
•We’d like to know what the ongoing star-formation rate in galaxies is;
dust unobscured and obscured•We’d like to measure the integral of the past history of star-
formation;cross consistency depends on the stellar IMF
•We’d like to know how metallicity
evolves in galaxiescross consistency with stellar mass estimates
Chary: NOAO/NSO 50th Anniversary 4/40
Extragalactic Background Light =Total Line of Sight Brightness –
Stars –
Zodiacal Light –
Interstellar Medium
Integrated Galaxy Light =Sum over light from all individually detected galaxies
IGL =< EBL
Chary: NOAO/NSO 50th Anniversary 5/40
CGRB
Blazars
300μm 3μm 300nm 0.4keV 40keV 4MeV 400MeV
Hasinger
‘00
Chary: NOAO/NSO 50th Anniversary 6/40
2.2 micron contributions
74%
14%
12% 0%
Zodiacal LightStarsCIRBISM
140 micron contributions
37%
27% 36%
Zodiacal LightCIRBISM
Unfortunately Sky Background at Infrared Wavelengths is Dominated By Zodiacal Light
CIRB (nW
m-2
sr-1)λ νIν
2.2: 22.4±6.03.5: 11.0±3.3140: 25±7240: 14±3
Chary: NOAO/NSO 50th Anniversary 7/40
2.6’
5.8μm25 hrs
24μm10.9 hrs
3.6μm25 hrs
Chary: NOAO/NSO 50th Anniversary 8/40
EBL Constraints
RC&Pope2010
Chary: NOAO/NSO 50th Anniversary 9/40
Results from the EBL
•
EBL is hard to measure and is known to poor precision
•
Equal components of optical+NIR
and FIR EBL → Dust reprocessing is important
•
IGL is lower than EBL at 1.2, 2.2, 3.6 and 60 microns → Diffuse source of radiation like cometary
dust or reionization
epoch galaxies ?
Chary: NOAO/NSO 50th Anniversary 10/40
Lyman break technique to identify star-forming galaxies
Vanzella
et al. 2007
Chary: NOAO/NSO 50th Anniversary 11/40
Hard to do spectroscopy of dusty sources
Chary: NOAO/NSO 50th Anniversary 12/40
But there is dust in different phases
Jason Marshall andIRS GTO Team
Chary: NOAO/NSO 50th Anniversary 13/40
Caputi
et al. 2007,Perez-Gonzalez et al. 2004Le Floch
et al., CE01, Franceschini
et al. 02Magnelli
et al. 2009
But phot-z
errors at z>1.4 are notoriousbecause σ(Δz/1+z)~0.03⇒σ(LIR)~3
Strong Evolution of LIRG and ULIRG population between z~0 and 1
Chary: NOAO/NSO 50th Anniversary 14/40
RC&Pope2010
Chary: NOAO/NSO 50th Anniversary 15/40
Cosmic Evolution of Extinction
Chary: NOAO/NSO 50th Anniversary 16/40
If UV slope were correct, EBL limits would be violated
Chary: NOAO/NSO 50th Anniversary 17/40
Option 1: Dust extinction in many z~4 galaxies is more consistent with SMC extinction law
H. Shim et al 2010
Chary: NOAO/NSO 50th Anniversary 18/40
Stars only
Nebular only
Stars+Nebular
Black line is for 1 Myr
old populationRed line is for 10 Myr
old populationCourtesy of Starburst99Leitherer
et al. 1999
Option 2: Nebular emission makes UV slopes redder
Top heavy IMF increases nebular emission
Chary: NOAO/NSO 50th Anniversary 19/40
Can we reliably understand the nature of galaxies dominating the SF ?
Chary: NOAO/NSO 50th Anniversary 20/40
?
Needs more data for confirmationDefinitely resolved by Herschel.LIRGs
are lower mass, ULIRGs
are higher mass
Chary: NOAO/NSO 50th Anniversary 21/40
Even for objects with spec-z, some violate upper limits or FIR photometry
Evidence for AGN ?
E. Murphy et al. 2009
Intriguingly, this happens at LIR>3E12 Lwhich is the most extreme source in the local Universe
Chary: NOAO/NSO 50th Anniversary 22/40
JD2: A z~2 LIRG. NOT a 6E11 M z~6.5 galaxy.
Are We Missing the z~2-3 LIRGs
?
Chary: NOAO/NSO 50th Anniversary 23/40
Sensitivity of Different Wavelengths to Dust Obscured Star- Formation
Mid-Infrared wavelengths are the most sensitive and least affected by confusion.However, requires large bolometric corrections.
Chary: NOAO/NSO 50th Anniversary 24/40
There are some nightmare sources!
SBS 0335−0521/40 ZHouck et al. 2004
Chary: NOAO/NSO 50th Anniversary 25/40
Wilkins et al. 2008Hopkins & Beacom
2006
An Evolving IMF at z<3 ?
Chary: NOAO/NSO 50th Anniversary 26/40
Reddy & Steidel
2009
Not so fast…..
Systematics
from: 1. Proper treatment of stellar
Remnants2. Integrating down the LF3. Dust corrections at z>2
Chary: NOAO/NSO 50th Anniversary 27/40
Results on SF History
•
Dust reprocessing well understood at z<1•
Some constraints on z>1 from 24 microns
•
Dusty SFR >> Unobscured
SF @ z<2•
At z>2 UV slope is x2 overestimate of obscuration–
Probably because dust is grey
•
No significant discrepancy with stellar mass density → no evolving IMF at z<3
Chary: NOAO/NSO 50th Anniversary 28/40
Everything Consistent ? GRB 090423 at z=8.2607.008.0
+−
•
Gamma-ray burst on 23 April 2009; highest spectroscopically
confirmed astrophysical object known; 600 Myr
after the Big Bang.
•First clue it was at high redshift
was the absence of an optical afterglow. Subsequently detected in the NIR (UKIRT/Gemini), ~0.5-1 hr after the burst.
•
Luminous (but not unusual) explosion, probably associated with the death of a massive star. 1053
ergs released within ~10s (observers frame).
•GRBs
are vital probes of metals, gas, star-formation and reionization
at early epochs since galaxies are too faint for spectroscopy (See e.g. Chary et al. 2007).
Tanvir et al. 2009, NatureSalvaterra et al. 2009
Chary: NOAO/NSO 50th Anniversary 29/40
Chary, Berger & Cowie 2007
WFC3 UDF(Bouwens)
Chary: NOAO/NSO 50th Anniversary 30/40
Could be due to GRB efficiency increasing as metallicity
decreases ?
Butler, Bloom et al. 2010
Chary: NOAO/NSO 50th Anniversary 31/40
Metallicity
increases as (stellar mass density)^0.69+/-0.17Exactly the same as dust content
Gamma-Ray Bursts Also Trace Metallicity
Evolution
Chary: NOAO/NSO 50th Anniversary 32/40
Chary: NOAO/NSO 50th Anniversary 33/40
Results from GRBs
•
Star-formation rate density is higher than from field galaxy surveys
•
Not due to dust obscuration
•
Can be explained by GRB production efficiency increases with decreasing metallicity
•
GRBs
are our ONLY tracer of metallicity
evolution especially at the faint end of the galaxy LF.
Chary: NOAO/NSO 50th Anniversary 34/40
Physical Triggering Mechanism ?
Chary: NOAO/NSO 50th Anniversary 35/40
Understanding Modes of Star-formation: Cold Flows or Minor/Major Mergers ?
Dekel
et al. 2009>10:1 mergers
Chary: NOAO/NSO 50th Anniversary 36/40
Are these starbursts or quiescent star-formation ?
Tightness of the SFR-Stellar Mass at 0.8<z<1.2seems to indicate it is quiescent star-formation
Morphological evidence is not clear –
some of theLIRGs
are spirals while some are irregular/S0s
Is this due to increasing dust content in massive galaxies ?
A starburst spans a relatively short time scale ~10% of the cosmic time within that redshiftrange. Expect a large scatter –
there is a good number of “evolved” red galaxies at z~1.
Unclear what fraction of the dust is heated by therelatively evolved stellar population rather thanyoung stars. Is this evidence that a significantfraction of MIR emission is dust heated by A stars? (Salim
et al., Bendo
et al.)Elbaz
et al. 2007Noeske
et al, Papovich
et al., Reddy et al., Daddi
et al.
Chary: NOAO/NSO 50th Anniversary 37/40
Some evidence for continuous accretion at z~4
H. Shim et al.
Chary: NOAO/NSO 50th Anniversary 38/40
But, some of those are morphological trainwrecks
Chary: NOAO/NSO 50th Anniversary 39/40
Summary
•
We have an excellent understanding of what dominates SF over half of cosmic time (z<1)
•
At z>1, uncertainties increase due to observational selection effects. But SFRD appears to be declining. Dusty galaxies still might dominate SFRD –
Herschel will be
insightful.•
At z<1, evolution appears to be due to mergers. At z>1, still up for discussion –
accretion from IGM might
contribute a fraction but 0.0 observational evidence.•
Massive galaxies appear to turn off their star-formation first (due to gas depletion or AGN feedback ?)
•
Metals and dust increase with increasing stellar mass density but at a slower rate, probably due to outflows in low mass halos.
Chary: NOAO/NSO 50th Anniversary 40/40
The Future
•
We have made immense progress in identifying star-forming galaxies to the earliest cosmic times
•
Measure EBL with precision; need to go beyond the inner solar system.
•
Measure FIR SEDs
of galaxies at z~2-4 using Herschel
•
Better spec-z in 1.4<z<2.5 range•
Get spectra of GRBs
efficiently –
best tracer of
z>6 star-formation and the first massive stars•
Get spectroscopic evidence for a cold flow before jumping on the bandwagon