Mid-Infrared Imaging of the Einstein Cross QSO

Eric Agol , Stuart Wyithe, Barbara Jones, Omer Blaes, Chris Fluke, PASA, 18 (2), in press.

Next Section: The IR Spectral Energy
Title/Abstract Page: Mid-Infrared Imaging of the
Previous Section: Observations
Contents Page: Volume 18, Number 2

The Infrared Source is Due to Dust

There are several lines of evidence that the infrared emission is extended on a scale greater than an Einstein radius. First, the mid-infrared flux ratios disagree with the (extinction-corrected) optical flux ratios at the 6$\sigma$ level, which indicates that while microlensing magnifies the optical emission region, the infrared emission region is too large to be affected by microlensing. Second, the infrared image ratios agree with the most complete lens models - the best is the $\lambda =1$ model of Schmidt, Webster, & Lewis (1995, SWL) which has a $\chi^2 = 8$ for the 1999 data and $\chi^2=7$ for the 2000. This can only be true if the infrared source is large enough that the magnification ratios of the images are not changed by microlensing. Finally, simulations suggest R > 0.2 RE at the 3 $\sigma$ level (Wyithe et al. 2001) based on the 1999 data alone.

Even stronger evidence for extended infrared emission comes from the Y2K data. Between the 1999 and 2000 observations, image C decreased by a factor of two (from 1.49 to 0.76 mJy) in V band (as monitored by OGLE, http://bulge.princeton.edu/~ ogle/ogle2/huchra.html, Wozniak et al. 2000). The other three optical images remained unchanged between the two observations; however, image A underwent a 30% amplification in between. The story is much different in the infrared: the infrared data taken in 2000 show no significant variation over a one-year timescale. Microlensing simulations convolved with sources of various sizes have been used to compute the probability that the infrared source could remain constant in flux within 10% while the optical flux decreased by a factor of 2 within 0.2 RE (the approximate distance traversed by the QSO over 1.2 years). This probability is nearly independent of the lens model, since it simply depends on the flux of a single image. Figure 1 shows the probability as a function of infrared source size assuming the SWL model 2. The largest source shown was limited by the size of the microlensing simulation (100 RE). This plot indicates that R < RE is ruled out at about 90% confidence, thus ruling out the synchrotron emission model.

Figure 1: The probability as a function of infrared source size that the IR flux would vary less than observed given the factor of two change in the optical flux of image C. The x-axis shows R in units of the Einstein radius (ER).
\begin{figure} \begin{center} \psfig{file=Im_c.ps,width=17cm,angle=-90}\end{center}\end{figure}

Based on these observations, we have strong evidence that synchrotron emission does not contribute to the infrared flux since a synchrotron source should undergo the same microlensing-induced fluctuations as the optical. The infrared emission cannot be due to a reddened starburst as there is no corresponding radio emission and the source is not extended. The alternative we are left with is that the infrared emission is due to QSO light absorbed by dust and then reradiated.


Next Section: The IR Spectral Energy
Title/Abstract Page: Mid-Infrared Imaging of the
Previous Section: Observations
Contents Page: Volume 18, Number 2

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