AA and the Evolution of the Galaxy Population

1
AA? and the Evolution of the Galaxy Population

• Matthew Colless, UCL, 28 June 2004

2
AA? Key Performance Features

• Throughput peak is 15-20, i.e. 2-3x 2dFs
  throughput
• Spectral resolution 1-8000 MOS (13000 IFU)
  gt2x 2dF
• Wavelength coverage 370-950nm, 300nm sim. at
  low-res.
• PSF spatially-invariant PSF for all setups
  wavelengths
• Stability better ?-calibration, better
  sky-subtraction than 2dF
• IFU 11"x22" FoV sampled at 0.7"
• Sampling better spatial sampling, wider fibre
  separation
• Nod Shuffle can NS all 392 fibres (also
  mini-shuffle)

3
AA? Survey Performance

• Instrument AA? 2dF 6dF Sloan FLAMES VIMOS WFMO
  S
• Telescope AAT AAT UKST
  VLT VLT Gemini
• Mirror diameter (m) 3.9 3.9 1.2 2.5 8 8 8
• Mirror area (m2) 11.9 11.9 1.1 4.9 50.3 50.3 50.3
• System efficiency 0.4 0.15 0.3 0.3 0.3 0.4 0.
  3
• FoV diameter (deg) 2 2 6 3 0.42 0.50 1.5
• FoV area (deg2) 3.1 3.1 28 7.1 0.14 0.20 1.75
• objects 392 400 150 640 132 560 4000
• photons/unit time 1900 700 50 900 2000 11000 600
  00
• Survey duration 0.07 0.18 0.11 0.10 0.47 0.25 0.04
  
• (time taken to cover
• fixed area of sky)

4
AA? S/N Estimates

• S/N per Angstrom in MOS mode, in 1(4) hours dark
  time
• 
  Seeing
• Magnitude 1.0" 1.5" 2.0"
• V18 64 (129) 55 (110) 45 (90)
• V20 20 (40) 16 (33) 12 (25)
• V22 4.1 (8.5) 3.1 (6.5) 2.3 (4.7)
• B22 6.5 (14) 5.0 (11) 3.7 (8.0)
• I 20 7.0 (14) 5.3 (11) 3.9 (7.8)
• (n.b. expected median seeing lt1.3" with dome
  air-conditioning)

5
Galaxy Evolution Programs

• Stellar Populations in Low-Redshift Galaxies
  Characterizing the stellar populations and
  star-formation histories of the current galaxy
  population as a function of global properties and
  local environment at low redshift (follow-up to
  2dFGRS)
• Resolved Stellar Populations and Dynamics
  Using AA? IFU capability to obtain
  spatially-resolved dynamics, ages and abundances
  in individual nearby galaxies
• Old Stellar Populations at Early Epochs
  Ages and metallicities of stellar
  populations in luminous red galaxies as a
  function of mass (velocity dispersion) at early
  epochs, from 0.45ltzlt0.7 (follow-up to 2dF-SDSS
  LRG survey)

6
1a. Stellar Populations in Galaxies

• Goal survey to characterize the stellar
  populations of galaxies (age, metallicity,
  star-formation history) as a function of global
  properties (luminosity, morphology, mass) and
  environment (local density/structures)
• Sample use 2dFGRS/SDSS to pick 100 galaxies in
  each bin of a grid evenly sampling luminosity (20
  bins over M-3 to M5), local density (10 bins
  in log?, voids to clusters) and morphology (5
  bins from E/S0 to Irr) ? 105 galaxies
• Method obtain spectra at 1Å resolution and
  S/Ngt50 to measure spectral indices (ages,
  metallicities, SFR histories) and, where
  possible, velocity dispersions (i.e. masses)

7
1b. Survey Observations

• Observations 2hr exposures at 1.4Å resolution
  (R3500, 75 km/s) give S/N45 to bJ19.5 and
  S/N90 at bJ18
• Spectral coverage chosen appropriately by
  type/colour
• Late types red/blue arms cover 3700-4500Å and H?
  resp.
• Early types red/blue arms cover 4600-5400Å and
  CaT resp.
• Old stellar populations determine age
  (distribution?) and metallicity (distribution?)
  from multiple spectral indices
• Star-formation rates obtain both current and
  recent star-formation rates from Balmer lines and
  other SFR indicators
• Masses from velocity dispersions for E/S0
  galaxies and bulge-dominated spirals

8
1c. Survey Duration Problems

• Time required with 300 galaxies/field, 4
  fields/clear night and 60 clear fraction, a
  sample of 105 galaxies will take 140 nights (say
  2 years at 35 nights/semester)
• Problem 1 - aperture effects depend on galaxy
  size, redshift and seeing, so can be ameliorated
  by choosing sample to minimize effects,
  correcting results via models
• Problem 2 - rotation velocities how best to get
  rotation velocities for galaxies with substantial
  disks?
• Partial solution IFU observations for a (small)
  subset of galaxies would allow the calibration of
  aperture effects and provide a sample of disk
  galaxies with rotation curves

9
1d. Value and Interpretation?

• Would this survey provide added value over
  SDSS/6dFGS?
• Yes - iff stellar population models of can make
  use of high S/N spectra to unravel star-formation
  histories and extract age and metallicity
  distribution of stellar populations
• This needs to be clearly demonstrated, and it is
  probable that substantial further work is
  required to improve stellar population models and
  star formation rate indicators

10
2a. Resolved Populations Dynamics

• Goal spatially-resolved dynamics, ages and
  abundances in individual nearby galaxies using
  IFU capability of AA?
• Observations want both the blue and red setups
  of pops survey to study bulgedisk/oldyoung
  stellar populations
• Dynamics map velocity field (1st to 4th moments)
  at many locations, likewise spectral indices, to
  build detailed picture of the dynamics and the
  stellar populations
• Sample One galaxy in each of 8 bins in
  luminosity covering (M-3...M5), 5 bins in log?
  (voids to clusters) and 5 bins in morphology
  (E/S0 to Irr) ? 200 galaxies

11
2b. IFU Observations

• Exposure times each IFU fibre is only
  0.5 arcsec2, and the desired S/N is high (50) so
  exposure times critically depend on extent to
  which IFU pixels can be co-added
• S/N and SB e.g. 4 hour exposures only reach a SB
  limit of 19.25 mag/arcsec2 and (if doing 2
  setups) this means the survey duration is gt300
  nights!
• Adaptive sampling clearly need to co-add many
  IFU pixels to reach interestingly faint surface
  brightness levels
• Competition should this be done with AA? on AAT
  or with SAURON on WHT or WIFES on SSO 2.3m?

12
3a. Old-Population at Early Epochs

• Goal study the formation and evolution of the
  oldest stellar populations by observing luminous
  red galaxies (LRGs) over 0.45ltzlt0.7 (follow-up of
  2dF-SDSS LRG survey)
• Sample select galaxies from LRG survey of 104
  luminous (MgtMz0) galaxies with rest-frame
  colours of old stellar populations (age gt 5Gyr)
  there are 100 LRGs/2dF field
• Method obtain spectra at 1Å resolution and
  S/Ngt30 to measure precise spectral indices (ages,
  metallicities, SFR histories) and velocity
  dispersions (masses) long time baseline resolves
  age-metallicity degeneracy in spectra

13
3b. Old-Pop Observations

• Setup need additional dichroic with split at
  8000Å and another 1000R VPHG to allow blue arm
  to cover 6900-8100Å and red arm to cover
  8000-9200Å at 0.6Å/pixel
• Measurements measure spectral indices H?, Mgb
  and various Fe lines in rest-frame range
  4800-5400Å, plus velocity dispersions with
  resolution ? 50 km/s
• Observations 8 hr exposures expected to give
  S/N30 to i19.5-19.8 (LRG survey limit) superb
  sky-subtraction is needed, but as surface
  density of targets is 100/field can use
  beam-switching and NS to reach Poisson limit
• Duration 50nt x 60 clear x 1 fld/nt x 100/fld
  3000 gal

14
Conclusions

• Key features field of view, multiplex,
  throughput, spectral resolution, stability, sky
  subtraction, IFU (multi d-IFUs?)
• Science programs high velocity precision or high
  S/N spectra of many faint objects over wide
  fields ? Stellar populations of galaxies
• Critical issues
• do models have interpretative power to add value?
• is fibre survey useful? is IFU survey
  competitive?
• Key advantages
• fastest survey spectrograph in the world
• opportunity for ambitious large programs