QSO Catalogue for Gaia

1

• QSO Catalogue for Gaia
• GWP-S-335-13000
• Alexandre H. Andrei (Observatório Nacional/MCT,
  and associated researcher to Observatório do
  Valongo/UFRJ and SYRTE/Observatoire de Paris),
  Anne-Marie Gontier (SYRTE/Observatoire de Paris),
  Christophe Barache (SYRTE/Observatoire de Paris),
  Dario N. da Silva Neto (UEZO), François Taris
  (SYRTE/Observatoire de Paris), Geraldine Bourda
  (Observatoire de Bordeaux), Jean-François Le
  Campion (Observatoire de Bordeaux), Jean Souchay
  (SYRTE/Observatoire de Paris), J.J. Pereira
  Osório (Observatório Astronômico da Universidade
  do Porto), Júlio I. Bueno de Camargo
  (Observatório Nacional/MCT), Marcelo Assafin
  (Observatório do Valongo/UFRJ), Roberto Vieira
  Martins (Observatório Nacional/MCT), Sébastien
  Bouquillon (SYRTE/Observatoire de Paris),
  Sébastien Lambert (SYRTE/Observatoire de Paris),
  Sonia Antom (Observatório Astronômico da
  Universidade do Porto), Patrick Charlot
  (Observatoire de Bordeaux)?

2

• CONTENTS
• 1) CU3QSO astrometric and photometric updates.
• ? 100,165 qsos present in at least one
  available optical image.
• ? compliant to the ICRF to 1.5mas, average
  zonal errors of 32mas, average precision at
  139mas.
• 2) Photometric redshift.
• ? neural network classifying results
• ? standard astrophysical packages HyperZ and
  LePHARE
• 3) Morphology and the signature of the host
  galaxy.
• ? 1343 fields trial on DSS2 and DR7 images
  leading to 30 PSF excess
• ? all sky DSS mapping on its way
• 4) Astrometric and photometric variability.
• ? 3 years monitoring of 14 long period qsos
  at the ESO2p2/WFI
• ? relationship between photometric and
  astrometric variability suggested

3

• 1) CU3QSO astrometric and photometric updates
• 100,165 qsos present in at least one available
  optical image, as recorded in the SDSS DR6,
  GSC2.3, and USNO B1.0.

• average zonal errors of 32mas,
• average precision at 139mas.
• directly tied and
• compliant to the ICRF to 1.5mas

4

• 1) CU3QSO astrometric and photometric updates
• The inclusion of the DR7 and
• the certifying of QSOs off the main optical
  catalogs
• add further 41,721 objects.
• The DR7 alone brings home
• 28,224 new objects.

• The magnitude and
• redshift distributions resemble
• those from the CU3QSO
• astrometric entries.
• But shifted towards dimmer magnitudes, as well as
  with
• larger proportion of undetermined magnitudes
  and redshifts.

5
2) Photometric redshift ? Neural network
classifying results Aiming to certify the
redshift of objects detected only once and/or
under not fully reliable conditions.

• Trial bench 53,152 qsos
• From the DR7
• BRI colors from the DSS2
• Results
• Standard NN, B,B-R,R-I seeds, 3 neuron levels
• sredshift 0.7 Pearson 0.44
• But no agreement if redshift is calculated from
  fake magnitudes (2mag at random)?

6

• 2) Photometric redshift
• Results
• 2nd degree complete polynomial
• on R,B-R,R-I
• 3 regimens of solutions
• zlt0.5
• 0.5ltzlt1.0
• zgt1.0

• Further work is needed, and on its way
• First results barely at 1s level of certainty.
• SYRTEs Neural Net, HyperZ, and LePHARE to be
  explored.

7

• 2) Photometric redshift
• Results (at the time of closing this edition)
• Another way of evaluating polynomial coefficients
  is by successive differences
• It happens of course to be quite alike to define
  color loci.
• Taking the three colors from B,R,I mags plus a
  linear combination of magnitudes, one gets s0.7z
• The adjustments hold well up to z1.5, enabling
  to estimate

• Further work is needed, and on its way
• SYRTEs Neural Net, HyperZ, and LePHARE to be
  explored.

8
3) Morphology and the signature of the host galaxy

• The distinction of QSOs from other AGNs (BLLac,
  CD, Seyfert) is just a matter of
• higher degree of brightness, variability,
  and the presence of the odd spectral line.
• Or, as the Unified Model goes just a matter
  of viewing angle.

Mullard Space Science Laboratory
Astrophysics Group webpage
9
3) Morphology and the signature of the host galaxy

• ? For AGNs in general, therefore also for QSOs,
  the host galaxy absolute magnitude should be
  brighter than -23.5.
• ? The host galaxy is thought of most of times be
  an elliptical or bulge dominated galaxy.
• ? The host galaxy luminosity seems to increase
  proportionally to the strength of the central
  source, i.e. QSOs host galaxies may expected to
  usually be brighter than those around less
  powerful AGNs.
• ? The size of the host galaxy also tends to
  follow the rule. Typical sizes for BLLac are
  13kpc.
• ? Host galaxies have regularly been resolved for
  AGNs to z lt 1.5 and 1arcsec resolution. Less
  regularly so for QSOs.
• ? The QSO space distribution peaks at
• z0.6 for B19, and at z1 for for B20.
  
  
• That is, the largest fraction of GAIA QSOs
• would be of nearby ones.

Number of quasars per deg2 as function of
redshift and magnitude (Crawford 1994)?
10
3) Morphology and the signature of the host galaxy

• ? One might expect a fair amount of contamination
  by alien AGNs among the GAIA extragalactic
  reference frame
• (because they would look alike by the GAIA QSO
  selection criteria,
• and because they still would look a lot
  pointlike).
• ? One might expect a fair amount of resolved host
  galaxies around the GAIA extragalactic reference
  frame QSOs
• (because the host galaxies do are large and
  bright enough,
• because of contamination by alien AGNs,
• and because the QSOs will be nearby ones).

? GSC2.3
? RORF
? CFHT
11

• 3) Morphology and the signature of the host
  galaxy
• ? All sky analysis using DSS2 images.
  1arcsec/px, B,V,R plates. 5arcmin2 images
  centered on each CU3QSO object. Database include
  the 3 colors (R completed) plus other images when
  available (e.g. SDSS).
• Trial bench on 1,343 R images for which also the
• SDSS DR7 images (0.396arcsec/px) were
  retrieved.
• Extreme magnitudes, colors and redshift
  stored,
• along with a representative DR7 sky
  distribution.
• ? Same IRAF pipeline run on both the DSS2 and
  DR7 images, to issue 3 PSF parameters SHARP
  (probing skewness), SROUND (probing roundness),
  and GROUND (probing normalness).
• ? Statistics from the in-frame comparison
  between the QSO and stellar PSF parameters, from
  the central quartiles average and standard
  deviation. Only retained frames with 20 or more
  stellar standards.
• ? Independent treatment for each parameter.
  Stellar sample defined in two distinct ways,
  leading to separate results
• from all objects above threshold 4s and
• from the objects assigned as stellar (class
  3) in the DR7.

12

• 3) Morphology and the signature of the host
  galaxy
• The PSF parameters are sound in comparison to the
  SDSS class separation estimators.
• SHARP is the worst behaved
• parameter, and the only one
• for which a linear term with
• magnitude was applied
• (for the DR7 fields).
• The comparison between the
• extremest cases of QSOs
• classified as non-stellar testifies
• both of the better quality of the
• SDSS images and pixelization, but at the
  same time of common reckoning on both images. The
  extreme non-stellar QSOs are 144 in the DSS2 and
  86 in the DR7 samples. Of these 50 are common.
• The objects not in common should be assigned to
  the characteristics of ther detectors and
  measures, as well as possibly to the difference
  of epochs (cf 3).

13

• 3) Morphology and the signature of the host
  galaxy
• The excess (rate of objects beyond 2s) of
  non-stellar quasars is significant as given by
  all the indicators, on both the DSS2 and DR7
  images, measured either against the field stars
  or the SDSS classified stars.
• Effect on the centroid error (assuming a host
  galaxy two magnitudes fainter than the
  quasar, and to a brightness distribution
  following on r2

14
4) Astrometric and photometric variability ?
Observations at the ESO Max Planck 2.2m
telescope, La Silla, Chile, f8, 0.238/pixel, WFI
4x2 CCD mosaic, on CCD 7 nearby the optical axis.
? Filters Rc/162 (peak 651.7nm, FWHM 162.2nm)
and BBB/123 (peak 451.1nm, FWHM 135.5nm). In
each filter 3 frames are taken, to a combined SNR
of 1000 (up to 2h total integration time).
Variability elements from Teerkopi (2000, AA
353,77). P 4y
15

• 4) Astrometric and photometric variability
• ? Data Treatment
• ? All images are treated by IRAF MSCRED for
  trimming, bias, flat, bad-pixel e split.
  Typically the image treatment enhances the SNR by
  a factor of 2.
• ? IRAF DAOFIND and PHOT are employed for the
  determination of centroids and (instrumental)
  magnitudes, with the entry parameters adjusted
  for each frame.
• ? Centroids and fluxes are obtained from the
  adjustment of bi-dimensional gaussians. The inner
  ring where the object counting is made and the
  outer ring where the sky background is counted
  are variable for each object and frame, but their
  ratio is kept constant.
• ? The plate scale and frame orientation are
  derived by IRAF IMCOORDS, from UCAC2 catalogue
  stars.
• ? The following tables bring the measures of
  precision (pixels).

Average precision (1512-0905 sample). Upper row,
best imaged objects. Lower row, all detected
objects (above threshold 4)?
Likewise for the R and B filters
16
4) Astrometric and photometric variability ?
Data Reduction ? (provisional) separated filter
solution
Object n of frame m
Average number of stars appearing at least in two
frames in brackets the minimum and maximum
17
4) Astrometric and photometric variability ?
Results Time line (415 days) and linear
correlations
18

•       Each player must accept the cards life
  deals him or her but once they are in hand, he
  or she alone must decide how to play the cards in
  order to win the game.
• Voltaire
• (and this WP)?
• Thank You