The Optical/Near-IR Colours of Red Quasars

Paul J Francis , Matthew T. Whiting , Rachel L. Webster, PASA, 17 (1), 56.
Next Section: OBSERVATIONS
Title/Abstract Page: The Optical/Near-IR Colours of
Previous Section: The Optical/Near-IR Colours of
Contents Page: Volume 17, Number 1

INTRODUCTION

It was long believed that quasars are blue. The optical/near-IR colours of optically selected QSOs are indeed uniformly very blue (eg. Neugebauer et al. 1987, Francis 1996). It was therefore a surprise when substantial numbers of extremely red quasars were identified in radio-selected samples (eg. Rieke, Lebofsky & Wisniewski 1982, Ledden & O'Dell 1983, Webster et al. 1995, Stickel, Rieke, Kühr 1996). The biggest sample of these objects is that of Webster et al, who were studying a sample of radio-loud quasars with flat radio spectra: the Parkes Half-Jansky Flat-Spectrum survey, a complete sample of 323 sources with fluxes at 2.7 GHz ($S_{\rm 2.7}$) of greater than 0.5 Jy, and radio spectral indices $\alpha $ (

$S_{\nu} \propto \nu^{\alpha}$) with $\alpha > -0.5$ as measured between 2.7 and 5.0 GHz (Drinkwater et al. 1997). While some of these Parkes sources had BJ-Kn colours as blue as any optically selected QSOs, most had redder BJ-Kn colours, and some were amongst the reddest objects on the sky.

Why should the Parkes sources be so red? A variety of theories were proposed:

  • The BJ magnitudes of the Parkes sample were measured many years before the Kn magnitudes. Quasars with flat radio spectra are known to be highly variable: this could thus introduce a scatter into the BJ-Kn colours, though it is hard to see why it should introduce a systematic reddening.

  • Elliptical galaxies with redshifts z>0.1 have very red BJ-Kn colours, due to the redshifted 400nm break. If the host galaxies make a significant contribution to the integrated light from the Parkes sources, this could produce the red colours. Masci, Webster & Francis (1998), however, used spectra to show that this effect was only significant for $\sim 10$% of the sample.

  • The BJ magnitudes were derived from COSMOS scans of UK Schmidt plates, and are subject to substantial systematic errors, which could introduce scatter into the BJ-Kn colours (O'Brian, Webster & Francis, in preparation), though this too should not introduce a systematic reddening.

  • Parkes quasars could have the same intrinsic colours as optically selected QSOs, but be reddened by dust somewhere along the line of sight (Webster et al. 1995).

  • Flat-radio-spectrum quasars are thought to have relativistic jets: if the synchrotron emission from these jets has a very red spectrum and extended into the near-IR, it could account for the red colours (Serjeant & Rawlings 1996).

In this paper, we test Webster et al's results by obtaining much better photometry of a large sub-set of the Parkes sources. To minimise the effects of variability, all our photometry for a given source was obtained within a period of at most six days. All the data were obtained from photometrically calibrated images, and rather than relying on only two bands (BJ and Kn), we obtained photometry in every band from B to Kn.

In principle, multi-colour photometry should enable us to discriminate between the dust and synchrotron models. If quasars have intrinsically blue power-law continua (eg.

$F_{\nu} \propto \nu^{-0.3}$, Francis 1996), reddened by a foreground dust screen with an extinction E(B-V) between the B and V bands (in magnitudes) and an optical depth inversely proportional to wavelength, then the observed continuum slope will be


\begin{displaymath} F_{\lambda} \propto e^{- 2E(B-V)/\lambda} \lambda^{-1.7}, \end{displaymath} (1)

where $\lambda$ is the wavelength in $\mu$m. This is plotted in Fig 1. Note the very characteristic `n' shape, as the dust absorption increases exponentially into the blue.

Figure 1: Continuum shapes of dust affected quasars. The extinction E(B-V) increases downwards: values are 0, 0.1, 0.2, 0.3, 0.4. Note the characteristic `n' shape.
\begin{figure} \psfig{file=dustcont.eps,height=100mm}\par\end{figure}

If, alternatively, the redness is caused by the addition of some red synchrotron emission component to the underlying blue continuum, continuum shapes will have a characteristic `u' shape, dominated by the underlying blue flux at short wavelengths but by the new synchrotron component at longer wavelengths (Fig 2).

Figure 2: Continuum shapes of quasars with an additional red emission component. To show some of the possibilities, two different arbitrary functional forms have been chosen for the red component: a power-law (left) and an exponential (right). The strength of this red component increases upwards. Note the characteristic `u' shape.The plausibility of synchrotron models is discussed in Whiting, Webster & Francis (2000).
\begin{figure} \psfig{file=redcont.eps,height=100mm}\par\end{figure}

If radio-quiet red quasars exist, they cannot be selected by conventional optical surveys. We show that by combining optical and near-IR data, it should be possible to select any radio-quiet sources with the colours of most of our radio-selected red quasars.

This paper describes the observations, presents the data, includes some simple phenomenological analyses of the results, and discusses the colour selection of red quasars in the optical and near-IR. We defer the detailed modelling of the data to another paper: Whiting, Webster & Francis (2000).

Next Section: OBSERVATIONS
Title/Abstract Page: The Optical/Near-IR Colours of
Previous Section: The Optical/Near-IR Colours of
Contents Page: Volume 17, Number 1

Welcome... About Electronic PASA... Instructions to Authors
ASA Home Page... CSIRO Publishing PASA
Browse Articles HOME Search Articles
© Copyright Astronomical Society of Australia 1997
ASKAP
Public