a new zeal for globular clusters · milky way gcs datasets: cohen etal. 1998, puzia etal. 2002,...
TRANSCRIPT
A New Zeal for Globular
Clusters
Duncan Forbes, Swinburne
Jean Brodie, UC Santa Cruz
Globular Clusters andGalaxy Formation
Duncan Forbes
Swinburne University
The Fellowship
• Robert Proctor
• Jay Strader
• Soeren Larsen
• Lee Spitler
• Mike Beasley
• Trevor Mendel
(next talk)
Collaborators
Half of this conference,
plus back home at Swinburne…
• Robert Proctor
• Trevor Mendel
• Michael Pierce
Two Towers
Most (perhaps all) large galaxies reveal two
distinct sub-populations of globular clusters.
V-I = 0.95 1.15 [Fe/H] = -1.5 -0.5
Bimodal or not Bimodal?
Bimodal Unimodal
Unimodal
Yoon etal. 2006
NGC 5128 Bimodal !
Spitler et al. 2007
Strader et al. 2007
NGC 4472 Bimodal !
The return of the King
(Profile)
Barmby et al. 2007
HST/ACS imaging of GCs in M31
Globular
Cluster
Formation
When ? How ?
Ongoing accretion
Recent mergers
Early formation
Forbes etal 2003
Ongoing GCAccretion
Sgr Dwarf Galaxy
M54 (dwarf nucleus)
+ 5 GCs + Whiting 1
Well fit by a closedbox model
Pal12, Ter7 [!/Fe] ~ 0
M54 [!/Fe] ~ +0.3(Pritzl et al. 2005)
Forbes et al. 2004 Fellhauer NZ conf.
Whiting-1
Canis Major
4-6 GCs
+ 2 old open clusters
(not BH176, Pal1)
Different age-metallicity relation forthe in situ GCs ?
Forbes et al. 2004
Huxor et al. 2005
Peng NZ conf.
UCDs
Willman1
dwarfs
FFs
M31
Half-light radius vs Magnitude
Milky Way GCs
Datasets: Cohen etal. 1998, Puzia etal. 2002,
Schiavon etal. 2005
!~50 MW GCs with S/N ~ 100
Test SSP models using a multi-line fitting
technique to Lick indices
Compare results to `known’ values
Alpha
elements
Pritzl etal 2005
SN II vs SNIachemicalenrichmenthistory fromhigh resolutionspectra
Metal-poor halo stars
Arnone etal 2005
Hierarchical Merging vs
Early Collapse
“This [early collapse] view has been
tenaciously resistant to challenges.”
De Lucia, Springel, White et al. (2005)
Galaxy
Formation
Merging of darkmatter halos iswell understoodbut details ofbaryon physicsless understood
Dissipative Collapse
Forbes etal 2005
Recent Mergers
To form an Ellipticalgalaxy GC system, newmetal-rich GCs (redafter ~2Gyrs) arerequired to form. Thenumber of metal-poor(blue) GCs is unchanged.
Ashman & Zepf 1992
Blue GCs vs galaxy mass
Spitler NZ conf
Blue GCs vs environment
Virgo
NGC 4365
AMR
"
Brodie etal 2005
Redshift 0 1 2 5
Galaxy age
NGC 4365 spectrum
"
Leo Group NGC 3379 AMR
"
Pierce, Beasley, Forbes etal 2005
NGC 821
NGC 4649 colours
Forbes etal 2005
Canis Major
Half-light radiusof Canis Major GCsare smaller thantypical Milky WayGCs
Forbes et al. 2004
Spiral+dwarf with Keck
• March 2003
• 1 hr long-slit spectrum
• Spiral velocity = 43,433 +/- 99 km/s
• Dwarf velocity = 43,445 +/- 225 km/s
• Redshift = 0.15
• Distance = 2 billion light years
• Spiral galaxy mass = 1012 solar masses
• Dwarf galaxy mass = 1010 solar masses
Dwarf properties
Reff = 1 kpc
!eff = 23 V mag/sq arcsec
!o = 21 V mag/sq arcsec
Mv = -16.0
Recent starburst
Twisted isophotes
=> dwarf ellipticalwith ~10 GCs
Forbes etal 2003
dIrr ->
dE ->
dSph ?CM
NGC 4649 AMR
"
Pierce, Bridges, Forbesetal 2006
NGC 1052 vs NGC 1316
Both are recentwet mergers:
Gas and dust
Fine structure
2-3 Gyr old stars
Forbes etal 2001! = N1052, " = N1316
Multi- GC systems
"
Strader et al. 2005
Both red andblue sub-pops arecoeval at~12 Gyr old.
Most GCformation inmassiveellipticalshappened atz > 2, not inrecentmergers. Blue Red
Old age isochrones
Young Elliptical NGC 1052 AMR
"
Pierce, Brodie,Forbes etal 2005
Wet…or Damp Mergers ?
V-I
g-z
Forbes etal 2007
Red
Red
Early-type Galaxies
ACS/Virgo Galaxies
Dry Merger
Galaxy Bimodality
MV ~ -19.5, M#~3x1010 Mo, Mhalo~6x1011Mo
Luminosity function, Colour-magnitude,Star formation rates, bulge/disk ratio,X-ray emission, stellar age, AGNemission, M/L ratio
Globular Cluster Specific Frequency
Transition from hot accretion flows to coldaccretion and SN winds (Dekel)
Virgo galaxies
For low massgalaxies
M/L ~ M -0.67
For high massgalaxies
M/L ~ M +0.5
Forbes 2005
Bekki NZ conf
Virgo galaxies
Bekki, Yahagi & Forbes 2006
SN in 1996
Durrell etal 1996
Tag Team Handover
GC-galaxy properties
Brodie & Strader 2006
Similar slope tostar-galaxyrelation.
Providesconstraints onhierarchicalmerging.
MP GCs formingat very high z“knew” the galaxyto which theywould ultimatelybelong
See also Peng et al 2006
Need blue GCtruncation atz ~5 to givebimodalcolours.
Reionization?
Hierarchical
GC
Formation
Beasley, Baugh, Forbes etal 2002
Blue TiltA mass-metallicityrelation for blueGCs.
Due to self-enrichment in darkmatter mini-halos?
BCGs: Harris et al2006
Summed dEs:Mieske et al 2006
Sombrero: Spilteret al 2006
Spirals: Forde etal 2007 Strader et al. 2006 Jordan NZ conf
Early GCFormation
Forbes, Brodie &Grillmair (1997)
+ Beasley et al (2002)
Need blue GCtruncation at z >5to give bimodalcolors.
Reionization?
#CDM…
N-body simulationssuggest dark matterhalos “sitting” on top oflarge overdensitiescollapse first (z>10)
The majority of thesehalos combine to formmassive galaxies at z=0
The Biasing
Scenario
• “Biasing” - halos on top oflarge over densities collapsebefore those on periphery
• Reionization truncates
metal-poor GC formation
• Halos that collapse first producemetal-poor GCs of higher metallicity
• First-collapsing halos produce the metal-poor subpopulationof the most luminous galaxies
• More distant halos survive independently to become dwarfgalaxies
Biased Hierarchical MergingMust simultaneously accommodate:
1) MP GC metallicity – galaxy Lrelation
2) Some merging since z~2MP GC relation was different at
high z• Merging was biased – direct
result of, and strong endconstraint on, hierarchicalstructure formation
• MP relation rules out mergersand accretion but only in thelocal universe for structureforming at present day
Brodie & Strader2006
Blue GCs vs galaxy mass
Reionization• If blue GC formation is truncated by reionization GC
surface density distributions probe epoch &(in)homgeneity of reionization
• Blue GC surface density distributions for galaxies ofdifferent masses and environments
Moore et al (2005), Diemand et al (2005) Bassino et al (2006)
MilkyWay Fornax
Conclusions• Although ongoing accretions and recentmergers contribute, most GC formation isearly (z > 2, age > 10 Gyrs), i.e. massive galaxiesformed early
• GC SN bimodality mirrors galaxy bimodality
• GCs provide the best observable link betweenbaryons and dark matter
• GC scaling relations constrain hierarchicalgrowth: red GCs trace bulge formation, bluestrace DM halos at earliest times + reionization
LMC/SMC
Cloud collision rate inLMC, and hence GCformation, is dramaticallyenhanced 3.6 Gyrs agowhen LMC/SMC reachclose pericentre.
Bekki et al. 2004