STRUCTURE FUNCTION ANALYSIS OF LONG-TERM QUASAR VARIABILITY Page: 5 of 18
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to Digital Library by the UNT Libraries Government Documents Department.
The following text was automatically extracted from the image on this page using optical character recognition software:
Long Term Quasar Variability. II
3 -,...r . ,. , ..-rr. . . _
2 -L - -
m - r
s , -
0 1 2 3 4 0 1 2 3 4
FIG. 3. Plot of the DR2 and 2dF quasars (left and right panel)
in the (u- g) and (g -r) color plane. This is one of the planes used
to select quasar candidates for spectroscopic follow-up with SDSS
(e.g., Richards et al. 2002). The stellar locus is indicated by the
thick solid white line, and illustrates the intrinsic color differences
between stars and quasars (hence the relative paucity of quasars in
that area). The SDSS sample has been divided into three redshift
bins (colored from dark to light): z < 0.4, 0.4 < z < 4, and z > 4.
Note the limited color range of the 2dF quasars compared to the
SDSS selection criteria.
trinsically by 10%, placing it inside a galaxy with the
same magnitude will lower the variability of the com-
bined system to 5%. The optical variability of an indi-
vidual quasar is not just a function of AGN luminosity
relative to its host galaxy luminosity, it also depends on
the redshift of the source. First, the contrast between
the AGN and its host galaxy increases dramatically to-
ward the restframe blue and UV wavelengths (as probed
by the passbands even at moderate redshifts). Second,
the (1 + z)4 cosmological surface brightness dimming
factor affects the extended galaxy more than the point-
source AGN contribution, again increasing the contrast
between the two components. Both these redshift depen-
dent trends diminish the unwanted variability-lowering
effect by the host galaxy, and based on Fig 1, it does not
appear to contribute beyond z e 0.4.
The bulk of our quasars (36 802) have redshifts be-
tween 0.4 and 4.0 (marked in Fig. 1), whereas 4424 (or
about 10% of the sample) are at redshifts below 0.4 and
might potentially be affected by their host galaxies. A
direct comparison between the results with and without
the low redshift data-set did not yield any significant dif-
ferences, except at the longest time-lags in the rSDSS-band
in particular (see 3.3.5).
The quasars added from the 2dF survey have a dif-
ferent brightness distribution (cf. Fig. 1, light-colored
points) than the SDSS quasars, mainly because of differ-
ent selection criteria. This difference remains, even if we
put the 2dF and DR2 overlap quasars back into the 2dF
sample, as is clearly illustrated in Fig. 2. Both the SDSS
and 2dF use multi-band photometric criteria to preselect
for quasar candidates. In addition, the SDSS sample has
been augmented by targeting FIRST and ROSAT coun-
terparts as well (Richards et al. 2002), and in general uses
a less restrictive color cut. The fraction of radio- and
X-ray loud sources is relatively low (2692 FIRST coun-
terparts within 2", and 479 ROSAT counterparts within
15"), so it does not significantly alter the distribution.
Figure 3 illustrates the differences between the SDSS
and 2dF selected quasars. The sources have been plot-
ted on the (u - g) and (g - r) plane, which is one of
the color-color diagrams used in selecting SDSS quasar
candidates (Richards et al. 2002). The 2dF quasar candi-
dates were selected from scanned UK Schmidt Telescope
(UKST) photographic plates, with magnitudes ranging
from 18.25 < bj < 20.85. In addition, the candidates
had to satisfy one of the following criteria (see Croom et
al. 2004): a - b < -0.36; a -b < 0.12 - 0.8(bj -r);
or b- r < 0.05. This results in a markedly differ-
ent color and (r-band) magnitude distribution from the
SDSS quasars. However, the important similarity is that
the redshift distribution between 0.4 < z < 4 is fairly
uniform as function of r-band magnitude (cf. Figs. 1
and 2). So, even though the quasars have been selected
differently, and actually populate the color-color diagram
of Fig. 3 differently, we feel that there is no a-priori bias
in either sample with respect to variability in general,
and variability on select time-scales in particular.
2.2. Photometric Calibration
Calibration of historic photographic plate material can
be achieved by virtue of using large numbers of random
field stars around the (quasar) position of interest. Plate-
to-plate variations in emulsion quality, and even varia-
tions within a single plate can contribute significantly to
measurement uncertainties. So, even though the POSS I
and GSC2 catalogs have been calibrated carefully (as a
whole), and brought up to CCD photometric standards,
there is still a lot of improvement to be made by recal-
ibrating the photometry. We basically follow the same
procedure as outlined in Paper I by using all the avail-
able photometry for the field stars within 5' of the quasar
position. This typically amounts to (depending on the
epoch) anywhere between 50 to 500 stars. We like to
stress that this "local" calibration is to be preferred over
complete plate corrections due to the potential inhomo-
geneities inherent to photographic plates (e.g., Lattanzi
& Bucciarelli 1991; Gal et al. 2003).
We will go over the calibration process step by step,
but we will refer to the calibration sections in Paper I
where appropriate. Most of the next discussion will high-
light the improvements we were able to make on the old
procedure, mainly due to the much larger data-set.
The first step is to calculate the best passband trans-
formations for each quasar individually, using the nearby
field stars. The transformations involved are, for the
POSS I: B to gSDSS, and R to rSDSS. Note that the B and
R magnitudes are already transformed to the Johnson
passbands from their photographic O and E emulsions
(see Reid et al. 1991, and Monet et al. 2003 for the B and
R transformations). For the GSC2 plates, the relevant
color transformations are J to gSDSS, and F to rSDSS. In
principle, our transformation will take care of the proper
passband corrections, possible plate / weather variations,
and the fact that SDSS uses AB magnitudes whereas the
catalogs are on the Vega system. However, an impor-
tant caveat we like to emphasize here is that the "best
transformation" is defined as the particular transforma-
tion that results in the smallest color rms for the stars.
As we have explained in Paper I, this does not neces-
sarily translate into the best calibration for the quasars,
which after all, is the transformation we are interested
in. There are two main contributors to this stellar-quasar
disparity: their optical spectrum is completely different
(cf. Fig. 3 of Paper I), and quasars typically have pow-
erful emission lines which depending on their redshift,
may, or may not, be present in the passband. These
Here’s what’s next.
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
de Vries, W; Becker, R; White, R & Loomis, C. STRUCTURE FUNCTION ANALYSIS OF LONG-TERM QUASAR VARIABILITY, article, November 15, 2004; Livermore, California. (https://digital.library.unt.edu/ark:/67531/metadc1410080/m1/5/: accessed May 21, 2019), University of North Texas Libraries, Digital Library, https://digital.library.unt.edu; crediting UNT Libraries Government Documents Department.