SED measurements of radio galaxies

1 < z < 5Bob Fosbury (ST-ECF) on behalf of the

SHzRGS team

Daniel Stern - PI, Peter Eisenhardt (JPL)Carlos De Breuck - Co-PI, Joel Vernet, Alessandro Rettura (ESO)

Bob Fosbury (ST-ECF)Andrew Zirm (STSCI)

Brigitte Rocca-Volmerange (IAP)Nick Seymour, Mark Lacy, Harry Teplitz (SSC)

Arjun Dey, Mark Dickinson (NOAO)Wil van Breugel, Adam Stanford (UCLLNL)George Miley, Huub Rottgering (Leiden)

Partick McCarthy (OCIW)Dave Alexander (IoA)

Objective

The hosts of powerful radio galaxies represent the most massive galaxies at their epoch

By observing the SED through the restframe H-band with Spitzer, we measure the stellar masses

The Spitzer images allow us to investigate the (proto)-cluster environment

The SMBH is characterised from X-ray and radio measurements

Evidence

gE and cD galaxies in the local universe

r^(1/4) light profiles in distant sources (NICMOS)

HzRG reside in (proto-)cluster environments

Large, luminous Ly-a halos

sub-mm detections => high star formation rates

NIR Hubble diagram (K-z; Rocca-Volmerange et al. 2004)

Correlation of stellar bulge and BH masses

Sample

The objective is to cover the radio luminosity - redshift plane as uniformly as possible in the range 1 < z < 4, covering two orders of magnitude in radio luminosity

Sample chosen to maximise number of supporting observation (but no bias towards cold dust properties - 850micron Obs not Det)

Solid symbols - HzRGs in our Spitzer sample with IRAC/IRS imagingLarge circles - MIPS observations as wellFilled circles - HzRGs with HST dataFilled squares - HzRGs with SCUBA dataPlusses - parent sample of 225 HzRGs from which our sample of 70 was drawn

Spitzer observations

3-camera programme: GO 3329 (Stern)

IRAC: 4x30s in all 4 bands => 5-sigma at m_AB=22 for ch 1 & 2 and at m_AB=20 for ch 3 & 4

IRS: 16 micron imaging (peak-up) mode; 2x60s => 5_sigma at 0.17mJy

MIPS: subset of 21 sources with low 24 micron background; all three bands [24 (30s), 70 (30s), 160 (50s)] but most detections at 24

Supporting observations

48/70 have HST imaging; total exposure of 1.5Ms

16 have Chandra/XMM-Newton data

43 have published sub-mm data

22 have deep optical polarimetric observations from Keck/VLT

Use to construct broadband SED and characterise AGN and dust re-radiation in addition to the evolved stellar population

Also use to study galaxy morphology and environment

WFPC2 - IRAC

HST/WFPC2 [top] and Spitzer/IRAC [bottom] images of 5 HzRGs

From left to right, galaxies shown are 3C266 (z = 1.275), MRC 0152-209 (z = 1.920), USS 1707+105 (z = 2.349), MRC 0406-244 (z = 2.427), and 4C+41.17 (z = 3792)

Images are 30 arcsec on a side, oriented with North up and East to the left. Note that a number of the fields show overdensities of infrared-bright, optically-faint galaxies

Supporting proposals

ESO and Palomar imaging (for 4000Å-break)

IRS mid-IR spectroscopy (Silicate, PAH and continuum)

MIPS - to extend subset

ESO imaging polarimetry (removal of AGN component)

MAMBO and SCUBA sub-mm

VLA 4.8 and 8.4 GHz to extend 3 GHz restframe depth: polarimetry to measure RM environment

HST archive

4C+23.56

This is our initial example object with very complete observational coverage.

Keck spectropolarimetry

Showing the dominance of the scattered quasar in the restframe ultraviolet(Cimatti et al. 1998;Vernet et al. 2001)

Composite SED

SED modelling

Initial modelling with stellar population (PEGASE.2), hot and cold dust (Siebenmorgen), and synchrotron components

Will add the scattered AGN

We need deeper sum-mm observations for most of this sample

Restframe H-band estimates

By Nick Seymour (elliptical templates +warm dust)

Full SED fitting tbd by Alessandro Rettura

Restframe H-band vs. z

Derived from best-fit models to the multi-band photometry

Sources with large error bars lack MIPS 24micron observations, making the decomposition of stellar and dust/AGN components of the SED less secure.

The error bars for these sources show that the range of reasonable stellar luminosities given (1) the total H-band luminosity and (2) that the stellar emission typically dominates at observed 3.6micron (as seen in the modelling of sources with MIPS data) and hence can predict a reasonable lower value to the H-band luminosity.

The red line shows the expected H-band luminosity of a passively evolving 10^12M_sun elliptical galaxy from the PEGASE.2 models (see Rocca-Volmerange et al., 2004)

First cut at M-z for Radio Galaxy hosts

BH mass vs. stellar mass

Local galaxies with measured kinematic and stellar masses (blue stars)

HzRG with measured X-ray luminosities (red squares)

Implications for BH growth at early epochs

Nick Seymour’s status pages