|
|
Extragalactic Scenario for AVO Demo 2004
Workflow
Step by step instructions of extragalactic demo
Obscured Quasars
Science Drivers
The unified model for active galactic nuclei (AGN) is widely accepted. The physics of black hole, accretion disk, jet, and obscuring torus is convolved with the geometry of the viewing angle and can explain most of the apparent disparate properties (and nomenclature) of active galaxies (the use of the word "torus" here is generic for the obscuring region). Type 1 sources are those in which we have an unimpeded view of the central regions and therefore exhibit the straight physics of AGN with no absorption. Type 2 objects arise when the view is obscured by the torus. While many examples of local, and therefore relatively low-power, type 2 AGN are known (Seyfert 2s), it has been debated if their high-power counterparts, type 2 QSO, exist. If so, they are expected to make a significant fraction of the X-ray background. These sources are heavily reddened and therefore fall through the "standard" (optical) methods of quasar selection. The hard X-rays, however, are thought to be able to penetrate the torus. Type 2 QSO, therefore, should have narrow, if any, permitted lines, powerful hard X-ray emission, and a high equivalent width Fe K line.
Some recent papers: A Classic Type 2 QSO, The HELLAS2XMM survey. IV. Optical identifications and the evolution of the accretion luminosity in the Universe
These science goals fit perfectly with the emphasis on spectral data which will characterize the AVO demo. The two GOODS fields, already used for the first AVO demo, are the most suitable areas of the sky.
|  |
Related pages
Data
The science case revolves around spectra (VLT/FORS2) and the heavy use of multiwavelength data to discover new objects.
Reduced FORS1/2 spectra are publicly available or were made available to us by the CDFS team.
We will also use X-ray catalogs and ACS images. The 5-epoch, stacked images are available, for example, from the ESO GOODS page
Demo Workflow
- 1. use X-ray (Chandra) catalogs, available for both the North (2 Ms) and the South (1 Ms). These have fluxes in various X-ray sub-bands.
- 2. compute the hardness ratio, defined as HR = (H-S)/(H+S), where H is the "hard" X-ray count rate (2.0 - 8.0 keV) and S is the "soft" X-ray count rate (0.5 - 2.0 keV). This will require manipulation of two columns in the X-ray catalogs. When H is an upper limit set HR equal to -1, when S is an upper limit set HR equal to 1. Following Szokoly et al. (2003) we define absorbed sources as having HR >= -0.2. XMM spectra in ASCII format might also be available for a few representative sources.
- 3. turn to optical data: these will include
- 3.1 if optical (FORS2) spectrum is available:
- 3.1.1 pop-up the spectrum by clicking on the optical image, overlaid with the X-ray catalog, using SpecView (what about XMM spectrum? do we want an SED?)
- 3.1.2 determine the type (1 or 2) and redshift in some automated fashion (measure line width/strength/ID); tools to do this have been developed but none is available (at least to us) for on-the-fly classification in a VO environment. The classification will then be left to the astronomer here; the redshift will come from two catalogs: one (still proprietary) provided by the CDFS team for the South, and another one, public, for the North (Barger et al. 2003)
- 3.1.3 derive X-ray power; i.e., take the X-ray flux in a given band, take the redshift, and compute the power; anything above 10^42 - 10^43 erg/s has to be AGN related. Need de-absorbed flux for type 2 sources. This can be done, for example, by assuming an X-ray spectrum typical of type 1 sources (alpha_x = 1) and by then deriving the absorbing column from the hardness ratio. Correction is around 15% in the 2 - 10 keV band for a typical HR ~ 0.5 and z = 1.
- 3.2 if no optical spectrum is available:
- 3.2.1 determine optical/IR fluxes by firing off a request to the Astronomical Catalogue Extractor (ACE); we need the ACS I-band fluxes (see below), which can be derived from the ACS I magnitudes by converting first to Vega magnitudes (I_Vega = I_ACS - 0.42) and then using the following relationship (from Zombeck) to derive the I-band flux: I_flux = 10^(-0.4*I_Vega-9.080))*2400.
- 3.2.2 Perform cross-matching between optical and X-ray positions. The optical positions are determined in step 3.2.1.
- 3.2.3 derive X-ray-to-infrared flux ratios from previous step (X-ray fluxes are available from the Chandra catalogues) and determine the type (1 or 2) according to the following prescription: if log(f_x/f_i) > -1.4 (equivalent to log(f_x/f_r) > -1 for (R - i) ~ 1: see below) AND HR is >= -0.2 then the object is a type 2; if log(f_x/f_i) > -1.4 AND HR is < -0.2 then the object is a type 1; the cut in f_x/f_i gets rid of the galaxies.
- 3.2.4 determine redshift; this can be done in two ways:
- 3.2.4.1 get photometric redshift from optical colors. Computation of redshift on-the-fly (photo-z) is difficult; images in different bands and/or taken with different instruments need to be registered on the same astrometric solution; plus, the different PSF have to be matched, i.e. aperture corrections have to be made, to obtain meaningful colours. No tools are available to do this on-the-fly yet. One solution is to degrade the ACS data to match them with the ISAAC data. Work on a tool which utilizes Hyper-z is almost completed by ASTRO-GRID, using at the moment the four ACS images. This will work only for relatively bright sources (R < 24 - 25).
- 3.2.4.2 alternatively, for type 2 sources (only) one can estimate the redshift from the X-ray power, derived from the X-ray to optical flux ratio, as in this paper, namely log Lx = log(f_x/f_r) + 43.05 (Fig. 5). As we will be using ACS i-band fluxes, we will use typical (R-i) colors to convert from R-band to i-band. Since these sources have typically <(R - i)> ~ 1, log f_r = log f_i - 0.4*<(R - i)> = log f_i - 0.4. Therefore log Lx = log(f_x/f_i) + 43.45. (All logs are log_10.)
- 3.2.5 derive X-ray power; i.e., take the X-ray flux in a given band, take the redshift, and compute the power; anything above 10^42 - 10^43 erg/s has to be AGN related
- 3.3 it would be good to display image cutouts, both optical and X-ray (in various bands); that will give a very good visual impression, especially for absorbed sources
- 4. final output: table with X-ray power and type; type 2 sources with Lx > 10^44 erg/s are the type 2 QSOs!
Here you can find all the manipulations we'll need to be able to do for the demo.
Could also use radio detections of Szokoly QSO 1 and 2's -Aladin screenshot of VLA steep-spectrum QSO-2 v. flatter spectrum AGN-1 in CDFS- more sources and details in RadioData.
Flow diagram
Fig. 1: Flow diagram; enumeration refers to bullet list above this figure.
Flow diagram - step 3.2.4.1
Fig. 2: Detailed flow diagram for step 3.2.4.1
|
|
Edit menu
