Structure in the sub-mm background towards the HDF

1
Structure in the sub-mm background towards the HDF

• John Peacock
• the UK sub-mm survey consortium

Novartis Institute, November 1999
2
Outline

• Motivation for resolving the sub-mm background
• The SCUBA 850-mm map of the HDF
• Optical identifications of the S850 gt 2 mJy
  sources
• Statistical detections of starbursts down to 0.1
  mJy
• Emission from z gt 5 galaxies
• Clustering of the 850-mm background

3

• Edinburgh James Dunlop, Dave Hughes,
  Andy Lawrence, John Peacock
• Imperial College Andreas Efstathiou, Bob Mann,
  Seb Oliver, Michael Rowan-Robinson, Steve
  Serjeant
• Cambridge Andrew Blain, Malcolm Longair
• UCL Rob Ivison
• JACH Tim Jenness

4
The SCUBA 850-mm HDF map
50 hours integration 100 arcsec radius Beam
14.5 FWHM - 30 and 45 chop Noise 8
FWHM - rms 0.45 mJy - doubles at 90 (divided
out)
5
Number counts
5 sources gt 2 mJy Confusion analysis Euclidean
counts to S lt 0.3 mJy 7000 deg-2 gt 1 mJy
6
Optical counterparts
50K b 1.5 photometric zs HDF 850.1 must have
z gt 3
7
HDF 850.1 continued
Downes et al. astro-ph/9907139
IRAM 1.3-mm position coincident with VLA source.
Between z1 and z2 galaxies, but inconsistent
with either.
8
Structure in the background
Subtract 5 brightest point sources Residual
structure from source superposition correlation
with known HDF starbursts?
9
Predicting sub-mm emission
Photometric redshifts UV flux
L1500
Wide z range expected
Dust-free starburst SFR L1500 / 1021 W
Hz-1 S850 / mJy SFR / 20 h-2 Msun yr-1
10
Predicted sub-mm maps
Significant cross-correlation over 1 lt z lt 6
zlt2.5
zgt2.5
11
Emission at extreme z
NICMOS GTO data found 8 z gt 5 candidates
(including 4-473.0 at z5.6) 2 close pairs
match well with 1-mJy level peaks
12
Mean starburst emission
Detect non-zero flux at 0.1 mJy level

cf. detection limits of 1-2 mJy for single
pointings cf. ratios gt 100 for ULIRGs like 850.1
13
Meaning of hidden SFR
All star formation is hidden never expect
dust-free burst

hidden visible ratio depends on lifetime of
high-mass stars and dust burnoff time
Jimenez et al. models (astro-ph/9910279) predict
61 for Salpeter IMF
Similar boost factor for SFR inferred by Pettini
et al. from Balmer lines, but foreground screen
is not the right model
14
(No Transcript)
15
Clustering of the background
Measure clustering via dimensionless power
spectrum
assume w (q / q0)-0.8 find q0 lt 2 arcsec
shot noise
reasonable if sub-mm galaxies have 1 lt z lt 6
Limber factor 3-4 reduction in w wrt
Lyman-break galaxies
true clustering
16
Summary

• HDF starbursts emit at 850-mm
• statistical detections down to 0.1 mJy
• mean hidden visible SFR is 4.5 1
• The 850-mm background is clustered
• consistent with shot noise
• limits background clustering to q0 lt 2 arcsec
• consistent with very broad redshift range
  contributing to background
• Future issues
• what is the upper redshift limit?
• nature of HDF 850.1?
• comparison with NICMOS