Optical follow-up survey to XCS.

1
NOAO XCS Survey (NXS)

• Optical follow-up survey to XCS.
• Goal Identify clusters and measure cluster
  photometric redshifts via their red sequence.
• NOAO 4m Mayall telescopes at KPNO, Arizona and
  CTIO, Chile.
• 38 nights 6 observing runs. (3 -northern winter,
  3 -southern winter)
• Aim image 330 XCS fields 500 clusters.
• Wide field Mosaic CCD imaging. 36' x 36' f.o.v.
  encompassing 1 XMM-Newton field) generally
  containing multiple clusters.
• Image clusters to z1
• Acts as screen for high redshift clusters
  followed up at Keck.

Image credit NOAO/AURA/NSF
2
Observing strategy

• CCD mosaic imaging in the r'-band and z'-band
  filters straddling the 4000A break chosen to
  provide maximum separation at z0.5.
• Two 600s exposures in r', three 500s exposures in
  z' (three exposures to reduce high background
  levels).
• Depth r' 24.7, z'22.2 at 10s.
• Priority to XCS500 clusters (for L-T study).
  Lower priority assigned to fields with SDSS, INT
  coverage.
• XCS pipeline written for completeness use
  existing optical data as a preliminary screen for
  follow up targets.
• Dithering to eliminate chip gaps in final stacked
  image.
• Use fields from DR2 to fill in RA gaps.

Image credit NOAO/AURA/NSF
3
NXS Observed

• Open nights 50.
• Additional 10 nights scheduled this summer at
  KPNO to cover RA gap in Northern hemisphere (gt100
  cluster candidates).
• Requesting make-up time for last CTIO run.

4
Data reduction and object catalogs

• Reduction using IRAF's MSCRED package -attempts
  to treat handling of mosaic field as a single CCD
  image.
• Following the NOAO Deep Wide-Field Survey MOSAIC
  Reduction procedures.
• Problems pupil ghost (KPNO only) and z'-band
  fringing.
• Catalogs created with SExtractor.
• Star-galaxy separation using the concentration
  parameter C (Metcalfe et al.,1991)
• Apply atmospheric extinction and Schlegal dust
  map corrections.
• zeropoints determined using a combination of
  standard stars, sdss coverage (relative zp
  accurate to 2) and NXS standard star fields
  (lt1 accuracy).

5
NXS Red Sequence Algorithm courtesy of Matt
Hilton

• The model
• Empirical red sequence model calibrated to
  reproduce the red sequence of an average low
  redshift cluster z0.1-0.11 from the SDSS C4
  catalog (Miller et al. 2005).
• model evolved with redshift using Bruzual and
  Charlot 2003 population synthesis code with a
  0.1Gyr burst and Salpeter IMF at zf2.5
• Redshift estimation
• Select galaxies from within twice the x-ray
  extent of the cluster.
• Assign each galaxy a redshift using the model and
  weight by its colour error.
• Evaluate probability of each galaxy being a
  cluster galaxy or field galaxy in 0.05 redshift
  bins. Using those probabilities perform 100 monte
  carlo simulations to statistically select the
  cluster galaxies (similar to Pimbblet et al.
  2001). Then use the mean redshift of those
  iterations as an initial redshift estimate.
• Select galaxies within 0.5Mpc of the cluster at
  that redshift, calculate a new probability,
  repeat the 100 simulations and iterate that
  radius until the redshift converges.
• Deduce the overall error on each red sequence by
  comparing known spectroscopic redshifts to their
  measured red sequence redshifts (?z0.05 Gladders
  Yee 2005).

6
NXS Clusters
Graphs show the model red-sequence at the
estimated z the data points represent galaxies
within 0.5 Mpc at the cluster photo-z estimate,
blue symbols represent galaxies selected as
cluster members based on one iteration of the
field galaxy background subtraction, the red
symbols represent galaxies which weren't
selected. The errors shown are statistical
errors on the photometric redshift.. NXS Colour
Composite images courtesy of Chris Miller.
7
NXS Clusters
Graphs show the model red-sequence at the
estimated z the data points represent galaxies
within 0.5 Mpc at the cluster photo-z estimate,
blue symbols represent galaxies selected as
cluster members based on one iteration of the
field galaxy background subtraction, the red
symbols represent galaxies which weren't
selected. The errors shown are statistical
errors on the photometric redshift.. NXS Colour
Composite images with X-ray overlays courtesy of
Chris Miller.
8
XCS Redshifts

• Redshift sources Literature spec-z's from Keck,
  Gemini, NTT, WHT NXS CMR photo-z's SDSS CMR
  photo-z's SDSS LRG photo-z's and spec-z's
  X-ray 2Mass Neural Net photo-z's.
• 97 NXS cluster candidates with redshift estimates
  (as of 3rd run).
• 600 XCS cluster candidates with redshift
  estimates zlt1.45 (All need eyeballing but so far
  400 XCS clusters optically confirmed).
• 124 XCS500 clusters with redshifts.

9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
Data reduction

• IRAF MSCRED package -attempts to treat handling
  of mosaic field as a single CCD image.
• Following the NOAO Deep Wide-Field Survey MOSAIC
  Reduction procedures.
• Create a master flat field (sky flat) from images
  taken on similar nights. Used to remove pupil
  ghost (KPNO only) and z'-band fringing.
• WCS correction using USNO stars (correct to
  within 1). Remove cosmic rays with IRAF CRUTIL
  package. Satellite trails using Matthew Hunt's
  Sat-b-gone.pl script.
• Project images to each other, remove overall sky
  gradient and stack images in each filter.

13
Object catalogs

• Catalogs created with SExtractor.
• Star-galaxy separation using the concentration
  parameter C (Metcalfe et al.,1991)
• Apply atmospheric extinction and schlegal dust
  maps corrections
• zp determined using a combination of standard
  stars, sdss coverage (zp accurate to 2) and NXS
  standard star fields (lt1 accuracy).