Nearfield Cosmology from the Andromeda galaxy and subgroup
Near-field Cosmology from the Andromeda galaxy and subgroup Scott C. Chapman Io. A, University of Cambridge With: R. Ibata, M. Irwin, G. Lewis, A. Ferguson, N. Tanvir, N. Martin, A. Mc. Connachie, J. Penarrubia, M. Collins, D. Trethewey
Outline n The M 31 outer disk n The M 31 outer halo: The “first” galaxy n (later: ) Dwarf galaxy satellites of M 31 (and the Milky Way) n n Is And-XII a “true” fossil? Minimum DM halo mass?
Context: Hierarchical Galaxy Formation - Little galaxies merge to make big galaxies … - How/when are the galaxy components assembled? Big Bang … Cosmic Microwave Background … … Galaxy Formation and Evolution … Fossil Records today! Local galaxies (MW, Andromeda) are ideal laboratories to study archeology.
Bullock et al. (2005) Model/Approach: 1. Construct accretion histories for Milky-Way type halos using semianalytic “merger tree”. 2. For each accreted system, model its previous star formation history based on expected mass growth history 3. Embed stars in the center of accreted dark matter halo. 4. Follow evolution within the (growing) host halo
Bullock+2005, Font+2006 Predictions: 1. ) Substructure in halo. 2. ) Chemically distinct outer halo. Observational Requirements: 1. ) Spatial coverage. 2. ) Radial velocities. 3. ) Chemical distribution. 4. ) Ages ? ?
Bullock+, Font+, Johnston+ model is our best current prediction for MW/M 31. (Changing with Aquarius -- Springel et al. 2008)
Observational Tests: Local Galaxy Archeology (Milky Way, Andromeda, satellites) Dissecting the history of a galaxy by digging up its stars 1 by 1: Near-Field cosmology -where are missing satellites -are DM profiles universal? (cuspy NFW? ) -DM: extent, nature, spatial distribution -how were MW and M 31 constructed: typical disks? -role of accretion in formation of halo, disk, bulge? -stars maintain birth statistical pattern -chemical evolution proceeds in 1 direction
Imaging and Spectroscopic study of Andromeda -- M 31
M 31 (M 33) Fossil Record of Galaxy Formation: Using the Keck 10 m / DEIMOS spectrograph: … dissect components & piece together the evolutionary history (Ibata+ 04, 05, 06; Chapman+ 04, 05, 06; Mc. Connachie+04, 06) Classical (Palomar) view of M 31 Modern (wide-field CCD) view of M 31: a train wreck! (Irwin+05) 6 degrees (12 full moons) 100 kpc
Fine structure of simulated galaxies age in Gyr thin disk thick disk mass fraction spheroid Abadi, Navarro, Steinmetz & Eke 2003
Building the Spheroid (Bullock+06)
M 31 Kinematics and Metallicity Experiment Ibata et al. (2004, 2005); Chapman et al. (2005, 2006, 2008) Imaging Surveys: INT 2. 5 m widefield survey of M 31 CFHT/Mega. Cam halo survey of entire M 31 halo out to 150 kpc Initiated w/ Keck/DEIMOS: Sept 2002 Status Nov, 2008 Number of Nights : 21 (70% usable) Number of Fields : 75 Limiting I-mag < 22 Candidate M 31 Spectra: ~14, 000 (and ~6000 Milky Way foreground stars) Data products: ~5 -10 km/s velocity determination, (Calcium Triplet, cross-correlation) [Fe/H] measurement by EWs of Ca. T Fainter lines of other species …
Surprise! the “messy halo” of stars surrounding M 31 is actually a giant rotating disk! (not a train wreck halo) (Ibata et al. 2005)
DEIMOS spectra
Technique: Sort stars by kinematics Velocity distribution of all stars 1 hour Keck exposures Apply disk model (flat rotation curve to >70 kpc) A new ``extended disk'' galaxy component disk halo §Discovered that all structures participate in giant rotating disk
Separation of outer disk/halo in velocity disk Outer disk of stars rotates like the inner disk ~15% of light of inner disk >40% of angular momentum! Irregular morphology, lots of substructure … transitory? (Ibata+2005, Chapman+2006) Star velocities in giant disk 40 kpc Distance (Major Axis) 40 kpc
Separation of outer disk/halo in velocity disk halo What’s left? A primeval “Halo” of stars! -few heavy metals (formed early) -An relic of M 31’s beginning v = 125 km/s Monotonic decrease in v(R) No rotation. Detectable “spikey” substructure? (Chapman, et al. 2006; 2009 in prep)
Mass of M 31 from Halo stars First assume simple rotating isothermal halo: not rotating, and 125 km/s v Then ignore rotation and allow v to decrease linearly with projected R. Monotonic decline in v with radius … better fit.
Mass of M 31 from Halo stars Monotonic decline in sigma_v with radius - ignore stream spikes … (Trethewey+08) Fit to an NFW dark matter halo (assuming the stars are a reasonable tracer of the halo: And taking Klypin et al. 2002 limit for concentration (caveats) M_virial > 9 e 11 Msun, 99% confidence. Consistent with other estimates of M 31’s DM halo mass (satellites - Evans&Wilkinson, giant stream - Ibata, Chapman et al. 2004)
Compare HALO and DISK chemistry: R~70 kpc Extended Disk - Metallicities Average spectrum at each Keck position All fields have similar average metallicities [Fe/H] ~ -1. 0 [sigma=0. 4] - More metal rich than MW halo. Average “extended disk” star from 15 kpc - 70 kpc probing similar global star population!
R=10 -70 kpc Stellar Halo: Metallicities Average spectrum at each Keck position fields have average metallicities [Fe/H] = -1. 4 [ =0. 2] Stars selected like those in MW halo (non-rotating), have similar metallicity and velocity distribution. NFW model fit => 10^12 Msun Solves “puzzle” of metal-rich halo in M 31!
Metal Poor halo: no abundance gradient Halo radial Fe/H constant: detect the metal-poor halo component from 50 kpc right up to 17 kpc … as opposed to minor axis where velocities of all components overlap in the M 31 systemic velocity range. Giant stream consistent with Koch et al. 2007
Koch et al. (2007): M 31 minor axis from 10 -120 kpc combined M. Rich & S. Chapman Keck-DEIMOS data Abundance transition at 20 kpc: metal-rich to metal-poor … inconsistent with previous sparsely sampled minor axis study (Kalirai et al. 2006).
M 33 (1/10 the mass of M 31 and MW) kinematically selected halo (Mc. Connachie+06, Trethewey+09) Keck/DEIMOS M 31 halo study … on edge of disk/halo transition from Ibata et al. (2007) Keck spectra find: 1) Metal poor halo Fe/H = -1. 5 2) Extended disk 3) Unknown “stream”
“Mouhcine plot” VERY HARD to see metal-poor primordial halos in more distant galaxies without kinematics! L vs Fe/H correlated in spiral galaxy halos? (Mouhcine+05) Kinematically selected Halos in M 31 (Chapman+06) M 33 (Mc. Connachie+06) MW (Chiba&Beers+00, 01) … all Fe/H ~ -1. 5 Are we comparing apples with apples in distant (10 Mpc) spiral galaxies? Dots are Renda+05 model
Conclusions n n In halos of big (L*) Spiral galaxies (M 31), extended rotating components may be common => difficult to interpret more distant galaxies without kinematics Beginning to understand the primeval halo of M 31 (and the MW …), versus later accretions. n More work required to understand substructure and mass function of first accretions Halo stars in front of M 31, outer edge of the MW halo n Growing discoveries of d. Sph galaxies (and their characterizations) are an excellent testbed of galaxy evolution and cosmology. n
Halo stars in front of M 31
CDM Has a Missing Satellite Problem CDM predicts large numbers of subhalos (~100 -1000 for a Milky Way-sized galaxy) Milky Way only has 23 known satellites M 31 has 25 satellites V. Springel et al. 2001 What happened to the rest of them?
CDM Has a Missing Satellite Problem CDM predicts large numbers of subhalos (~100 -1000 for a Milky Way-sized galaxy) Many never form stars V. Springel et al. 2001
What is a dwarf Fossil*? *defined by Ricotti & Gnedin (2005) Survivors (M > 109 M ) * star formation started after reionization * mostly d. Irr, some d. E LMC M 32 Polluted fossils (M ~ 106 - 108/9 M ) * significant star formation after reionization * tidal effects from host cause additional bursts * d. Sph and d. E Pegasus True fossils (M ~ 106 - 108/9 M ) * < 30% of stars formed after reionization * never accreted gas from the IGM * d. Sph Cetus (Whiting et al. 1999)
R 02 a, b predictions. Known survivors Known polluted fossils Known true fossils New ultra-faint dwarfs Fossil Properties ~ Scatter in Z due to: - pollution from nearby halos - multiple bursts of star formation (ie. Stinson et al (2007))�� ~ Fossil properties at z = 0 are simply related to their properties at reionization. Ricotti & Gnedin (2005), Bovill et al. (2007)
Formation and Evolution of dwarf galaxies n Environment of dwarfs severely affects their properties. Most dwarfs have been orbiting around our Local environment for most of the age of the Universe (>10 Billion years) Mc. Connachie & Irwin 2005
CDM predicts late accreting DM halos n n n But we’ve never seen one … have any of them formed stars? Dwarf Galaxies still bringing in primeval material?
Late accretions: 4% today n n Objects that accrete late do so from larger average distances than those that accrete early. late accreting objects interesting both observationally and theoretically n n n Ludlow et al. 2009 spent the majority of their lives in different environments, far from the disruptive tidal forces of larger galaxies, a direct prediction of theoretical CDM model for structure formation.
Dwarf Galaxies still bringing in primeval material? DISCOVERY of And. XII, a faint Dwarf galaxy building up the Local Group environment: falling in for the first time! n. Direct observational evidence for the hierarchical formation of the Local Group. n. Insights into processes responsible for the dynamical evolution of dwarfs? (Chapman et al. 2007)
“Mateo” Plot n n n Simplest possible model: equilibrium, spherical, isotropic systems where M follows L (c. f. Strigari et al. 2007) Do all dwarfs live in similar halos? Is there a minimum mass for dwarfs? And 13 Kinematics in dwarfs: And 15 And 12 And 11 And 16 MW: Martin+07; Simon+07 M 31: Chapman+05, 07; Collins+08; Letarte+08 Globular star clusters, no DM And more to be studied/discovered … And 17, 18, 19, 20, 21 (Irwin+08, Mc. Connachie+08)
What Does This Problem Tell Us? n Two basic sets of possible solutions: n Modifications to CDM n n What modifications? Power spectrum, DM particle mass/decay/interaction crosssection? Astrophysics prevents stars from forming in most low-mass halos n Reionization, feedback, winds …
Angular Momentum (J) Catastrophe Sizes of galactic disks linked to J of parent DM halos (Fall & Efstathiou 1980) • distribution of halo spin parameters ( N-body simulations, e. g. Bullock+ 2001) • baryons and dark matter initially share the same distribution of specific angular momentum, j, within the halos (e. g. van den Bosch etal. 2002) • j is conserved as the baryons contract to form the disk (e. g. Mestel 1963). Disk sizes with these assumptions, roughly comparable to those observed. But, Hydrodynamics shows this process is invalid. => significant fraction of J of the baryons *is* transferred to DM, … disk sizes 10 x too small! (Navarro & Steinmetz 2000) SOLUTION? Feedback … remove incoming dwarf galaxy low-j baryons (Maller&Dekel 2003)
Forming the Edisk How to further increase angular moomentum by 50% ? ? ? Accretion origin to extended disk? (Penarrubia+06) BUT: Requires specialized conditions; large in-plane accretion(s); … would be consistent with observations
Evolving Fossils to z = 0 n Fossil properties at z = 0 are simply related to their properties at reionization. n Properties of the new Sloan and M 31 dwarfs agree well with predictions for primordial galaxies
- Slides: 41