Lecture 22; Nov 13, 2017 High-Redshift Galaxies III - Galaxy Evolution - Galaxy “Main Sequence”: Relevance - Star Formation Law, Gas Fractions - First Light and Reionization Reading: Chapter 9 of textbook Continue working on your final project (presentations due Nov 15/20). • We are happy to meet with you to discuss remaining issues, questions etc. • You have 12-15 minutes (practice!) • An electronic version of your final paper is due by Dec 6, 4:30 pm
51
Embed
High-Redshift Galaxies III - Cornell University · High-Redshift Galaxies III ... at z=1-3 compared to present day ... , the first stars and galaxies form and begin to ionize the
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
Lecture 22; Nov 13, 2017
High-Redshift Galaxies III- Galaxy Evolution- Galaxy “Main Sequence”: Relevance- Star Formation Law, Gas Fractions- First Light and Reionization
Reading: Chapter 9 of textbook
Continue working on your final project (presentations due Nov 15/20).
• We are happy to meet with you to discuss remaining issues, questions etc.• You have 12-15 minutes (practice!)
• An electronic version of your final paper is due by Dec 6, 4:30 pm
In class presentationsNov 15/20: Final presentation (15 minutes incl. questions)
Wed/Mon: Final Presentations (aim for min. “1 min per slide”, so <~10 slides)try to prepare to speak freely, use your summary to memorize key issues
Email us a copy by morning (make sure we get it!) or bring your own laptop to avoid unexpected technical issues
I use a Mac, so if you use another OS to prepare, a pdf may be safer in case
2015151515202020
?
First galaxies
“Epoch of galaxy assembly”
Present day
CIB vs. Star Formation HistoryThe dust-obscured fraction of star formation in the universe is significant!UV/optical studies miss substantial fraction of cosmic star formationÞNew phenomenon: distant, very IR-luminous (observed-frame submm) galaxies
The (sub)mm is a key wavelength regime to understand galaxy evolution!
ALMA Deep Fields
ALMA Deep Fields
• The “K-correction” is the correction we apply to the observed flux of an object of a given SED that accounts for its redshift.
• The K correction at the mm and submm at λ > 250 µm, which yields a flux density that is almost independent of redshift.
“Negative” K-correction
• At high z the SFR was >10x higher than today (Lilly et al. 1996, Madau et al. 1996)
• Did it drop because there were fewer merger-driven starbursts?
Þ insufficient, only ~20% of galaxies at z~0.75 are visibly interacting (but: lower limit due to morphological classification)
• Did it drop because “quiescent” disk galaxies today form fewer stars?
Þ yes, but why? (fgas, discuss later)
Why did the global SFR decrease towards the present epoch?
Star formation => stellar mass• The rate at which stars
form over time must correspond to the buildup of the stellar masses of galaxies
• Individual galaxies can grow in mass by
- forming new stars
- coalescence (merging) of pre-existing bits
• Merger rate?
- of order 1 major merger since z~1 for massive galaxies
• At present: galaxy mass function is– “Schechter function”– most stars in Mgal=1010.5Msun
• At earlier epochs:– Define M*-limited sample, independent of SFR
(which brightens galaxies)à near-IR selection is needed
• Results:– Galaxy mass function looks similar 0<z<4– “characteristic mass” was only slightly lower
at high-z– Co-moving density was considerably lower
• Most stars were always in the most massive galaxies!– At least for z<4, since when 95% of all stars
formed
e.g. Marchesini et al. 2009
Galaxy Mass Function at Earlier Epochs
Connecting the BzK, LBG and DRG PopulationsSF Density Contributions
LBG
Distribution of M>1011M� galaxies
LBG DRG
• Reddy et al. (2005) - sBzK and LBGs are (largely) identical populations
• Kong et al. (2006) - Clustering of sBzK and pBzK galaxies is identical, suggesting same population and SF is simply transient
• van Dokkum et al. (2006) - LBGs constitute only 17% of massive galaxies!
The star formation “main sequence” of galaxies
Daddi et al. 2007, Noeske et al. 2007
The general high-redshift galaxy population:BzK, BX/BM, LBG-selected galaxies (“typical”/“normal”), SMGs (“starbursts”)
- There appears to be a relation between SFR and M* for actively star-forming galaxies, a “main sequence” (MS) of star formation
- “passive” galaxies fill the triangular region below- Merger-driven “starbursts” deviate from the MS (few times higher SFR)- The normalization appears to evolve with redshift towards higher SFR
(caution: there are some mergers/starbursts on the MS, but they are a minority)
Physical relevance of the star formation “main sequence” of galaxies
- where disk galaxies are located at high z
- 90% of cosmic star formation out to z~2 occurs on main sequence
- MS galaxies have typically high duty cycles, are mostly not starbursts (but they can have high SFRs)
e.g., Adelberger et al. 2004, Noeske et al. 2007, Daddi et al. 2007, Foerster Schreiber et al. 2009, 2012, Rodhigiero et al. 2011, Wuyts et al. 2012
Herschel/PACS
Evolution of the Main Sequence
Comparison of 25 studies in the literature (Speagle et al. 2014 - undergrad student):- MS is constrained out to z~6- When putting all on a common calibration scale, remarkable agreement- Width of the MS: remarkably narrow, 0.2 dex- Slope and normalization of the MS: both are likely time-dependent:
SFR(M*, tcos) = (0.84 ± 0.02 - 0.026 ± 0.003 × tcos)log M* - (6.51 ± 0.24 - 0.11 ± 0.03 × tcos), where tcos is the age of the universe in Gyr
Specific SFR:sSFR = SFR/M*
Galactic star formation in equilibrium with cosmic accretion & outflows
e.g., Keres et al. 2005, 2009, Guo et al. 2009, Oppenheimer & Dave 2006, Dekel et al. 2009, Dave et al. 2010
Star Formation Law at High Redshift
- High redshift galaxies follow the same SF law as nearby galaxies
- SF law may have two sequences, “quiescent”/disk vs “starburst”/merger-driven
- difference in SF efficiencies is reflected in different gas depletion times (<100 Myr vs. 1 Gyr)
- physical mechanisms are the same, but high-z galaxies are high on both axes
- main complication: conversion factor aco to obtain Mgas is difficult to measure at high z
Daddi et al. 2010, Genzel et al. 2010
Carilli & Walter 2013 ARAA; after Magdis et al. 2012
const.?
Even “main sequence” galaxies(defined as typical SFR/M*(z))
show 10-30x higher gas fractionsat z=1-3 compared to present day
Ø Increased SF history driven by high gas fractions of galaxies(not by extreme merger rates)
Ø Star formation is elevated, but underlying physics are similar
Ø Evolution at z>3 poorly known
Metallicity and Conversion Factor aCO
- Local universe: dwarfs are high on SF law due to metallicity effects, resulting in high aCO
- High redshift: metallicity evolution, even massive galaxies can have sub-solar metallicity
Þ can mimick different SF efficiency. In principle, running aCO could unify SF “sequences”
Þ likely to contribute to the scatter in the SF law, but unlikely to result in single tdepletion
Genzel et al. 2012
cold gas history of the universe
connection:
star formation law (Mgas vs. SF rate)
Star formation law: SF history of the universe is a reflection of the cold gas history of the universe (gas supply)
Ø Studies of galaxy evolution are shifting focus to cold gas (source vs. sink)
WSFR WM(gas)
Problem: populations at high-z so far are highly selected (IR, radio, UV/optical luminosity)Ø may miss cold gas rich, quiescent galaxy populations
Solution: complete census of molecular gas, the fuel for star formationi.e. a molecular deep field (at the same time: continuum deep field)
H2 density
SFR density
COLDz Molecular Deep Field: A JanskyVLA Large Program
CO luminosity function is only measured well at z=0 to date.
Ø Require very deep, wide-band line survey over substantial cosmic volume to measure CO luminosity function at high redshift
Ø CO à H2 mass: Obtain “Cold Gas History of the Universe”
à observe CO J=1-0
Only possible with the fully upgraded Karl G. JanskyVLA
Ø VLA Large Program (PI: Riechers)
R. Pavesi, PhD Thesis project
Big Bang f(HI) ~ 0
f(HI) ~ 1
f(HI) ~ 10-5
History of Normal Matter (IGM ~ H)
0.4 Myr
13.8 Gyr
Recombination
Reionization
z = 1100
z = 0
z ~ 6 to 120.4 – 1.0 Gyr
Djorgovski/Caltech
• Once the H-atoms form (380,000 yrs), the universe is filled with neutral gas which should emit in the 21 cm line (ground state, spin-flip transition)
• The universe continues to expand and cool; the gas remains neutral in the absence of any source of ionizing radiation.
• “Dark Ages”: there are no sources of “light”.
• Once the first objects formed, the hydrogen was reionized!ÞThe “Epoch of Reionization”
A. z>200: TCMB = TK= TS by residual e-, photon, and gas collisions. No signal.
B. z�30 to 200: gas cools as Tk ≈ (1+z)2
vs. TCMB ≈ (1 + z), but TS = TK via collisions => absorption, until density drops and TS àTCMB
C. z�20 to 30: first stars => Ly-α photons couples TK and TS
=> 21-cm absorption
D. z�6 to 20: IGM warmed by hard X-rays => TK > TCMB. TS coupled to TKby Ly-a. Reionization is proceeding => bubble dominated
E. IGM reionized
Signal: HI 21cm Tomography of IGM
z=12 9
7.6
§ DTB(2’) ~ 20 mK
§ SKA rms ~4 mK
§ Pathfinders: rms ~ 80 mK
ÞRequires SKA (>2023-30)
Furlanetto, Zaldarriaga et al. 2004
Reionization• After recombination, the universe was neutral• At z~20 – 30, the first generation of galaxies and mini quasars formed• At z~6 – 15, the UV radiation from the first generation objects ionized
most of the HI in the universe– The neutral fraction of the universe changed from 1 to 10-5 (phase
transition in ionization state)– The temperature of the IGM electrons changed from CMB
temperature to 104 K (phase transition accompanied by temperature change)
– IGM becomes transparent to UV radiation, the universe is like a giant HII region (temperature change accompanied by opacity change)