Tales of Cataloguing VII -- QSOs in pokes

In the April 2013 release of the Million Quasars catalogue v3.3, I
removed 144 "quasars" which I had sourced from the classic VCV13
quasar catalogue(2010,A&A,518,10). These objects were originally from
Iovino et al, 1996 A&AS,119,265 which presented 1581 quasar spectra,
but marked 144 of them as "?". I became aware of these outliers
because many of them were anomalously bright, and investigation showed
that none of the144 had any radio or X-ray association, whereas 66 of
the other 1437 did have. Thus, if the "144" are all true quasars, the
chance of none having radio/X-ray is (1371/1437)^144 = 1/10th of 1%.
Therefore it is 99.9% likely that they are not all true quasars.
Further numerical analysis reveals that no more than 14 are likely to
be true quasars. So I removed them all.

In this way it became clear that the Verons had been a little too, er,
inclusive in their collection. So recently I have investigated more
doubtful objects from VCV13 by consulting the original discovery
papers, and have found about 700 more to remove, as follows:

(1) Schneider/Schmidt/Gunn 1994 AJ 107,1245 published 928 objects of
which, they stated, 305 were quasars of z0.68 and 623 were
emission-line galaxies (ELGs) of z0.45. However, all (not pre-empted
by other surveys) were taken up as quasars by VCV13. I have now
removed 484 of these ELGs.

(2) Similarly, Schneider/Schmidt/Gunn 1999 AJ 117,40 published 96
objects of which 55 were ELGs of z0.45, all taken up by VCV13 as
quasars. Those 55 ELGs are now removed.

(3) Papovich et al, 2006 AJ 132,231 published a large list of many
object types. VCV13 selected the quasars among these, but also took
up 13 ELGs / type-2s as being type-1 QSOs, which I have now removed.

(4) the 2dF-GRS galaxy survey (Colless M. et al, astro-ph/0306581) did
not publish quasars, although some authors, notably Madgwick D. et al,
2002 MNRAS 334,209 did mine it for quasars. However VCV13 evidently
did some mining of their own, promoting 63 2dF objects with nominally
high redshift into quasars. I've inspected many of these on DSS, and
confusion with neighbours is common. However, 5 of the objects are
quasar-like with radio/X-ray association. The remaining 58 have been
removed.

(5) La Franca F. et al, 1999 A&AS 140,351 published a large list of
quasars but excepted some objects as galaxies. VCV13 took up the lot.
I have removed 41 galaxies, thanks to NED for differentiating them as
the paper is not on ADS for some reason.

I selected these papers as those showing large numbers of anomalous
objects, so these are all that I have done. More work will have
diminishing returns. I'll do more when someone funds me. :-)

By the way, NED too has been naughty in places. Their "quasars" with
names starting with FPT and [LPK2009] (about 40 in total) were from
papers showing galaxies in the fields of quasars -- those galaxies
promoted to quasars in NED -- and demoted back out of the Million
Quasars catalogue.

Eric Flesch
Wellington, New Zealand

Hi Eric

In his book "The static Universe", (1) Hilton Ratcliffe publishes a
table of 40 quasars with verified proper motions (page 107).
He provides an online reference:

http://www.site.uottawa.ca/~tcl/fact......5..521V.pdf

You get to that PDF from

http://www.laserstars.org/V1982/index.html

Just click in the
"Also available in Adobe Acrobat PDF format."
link at the top of the page

and look at table 2.

Could you please have a look Eric?

As an expert in these matters, you can maybe confirm/infirm
this:

Are some "quasars" MOVING?

You yourself found that some quasars had "wrong coordinates",
and I remember proposing you that maybe those quasars are moving...

Now, investigating this highly heretical subject, I have found
these references, that are surely worth some of your time...

Could it be that those quasars with "wrong coordinates" were
just moving ones?

Of course (if they move) I do not think that your observations
have the necessary time depth to observe actual movements but MAYBE
if you look again at one of the quasars where you "fixed" the
coordinates, and see a small discrepancy it could be... well
a movement?

Thanks in advance, and when you receive the Nobel for this
discovery please do not forget to invite me to a beer :-)

jacob
-----
(1) The static Universe. Exploding the myth of Cosmic Expansion.
Hilton Ratcliffe. Edited by Apeiron, Montreal.
ISBN 978-0-9864926-2-4

Very heretical astronomy book available at Amazon. Of course I
do not endorse all theories proposed there, I lack the
necessary background to really assert what is going on. But a
fascinating reading for people that like the science of astronomy

In article , jacob navia
writes:

vvvvvv
In his book "The static Universe", (1) Hilton Ratcliffe publishes a
table of 40 quasars with verified proper motions (page 107).
^^^^^^^
There is a tad of irony here. (OK, I haven't read the book. Probably
by "static" he means "not expanding" and uses large(?) proper motion of
QSOs to argue that they are not at cosmological distances, thus
"challenging the paradigm".)

Of course, if quasars were really ejected from nearby galaxies at high
speed, we WOULD see significant proper motion.

At some level, almost all quasars should have some proper motion. If
your resolution is good enough to measure it, then probably the quasar
will no longer be a point source, but has a definite shape, which
changes with time. So, it is difficult to define a fiducial point whose
motion one can detect. That doesn't mean that it is impossible, but
realistic proper motions will probably be swamped out by such effects,
at least on the timescales we have access to now.

On Fri, 18 Oct 2013, jacob navia wrote:
In his book "The static Universe", Hilton Ratcliffe publishes a
table of 40 quasars with verified proper motions (page 107).
http://www.site.uottawa.ca/~tcl/fact......5..521V.pdf

That article dates from 1980 and it is important to note that
astrometric precision was far less then than today. So, as with the
canals of Mars, it was easier to fool oneself.

table 2. Could you please have a look Eric?

Sure, and I will preface my remarks by saying that checking on proper
motion is a trivially simple thing to do, because we have all-sky
images available from different epochs. Specifically, the Digitized
Sky Survey (http://archive.stsci.edu/cgi-bin/dss_form) gives us sky
cut-outs from the POSS-I survey (epoch 1950s-1960s) and the POSS-II
survey (epoch 1980s). We also have SDSS finding charts from epoch
late-2000s, such as the SDSS-DR9 finding chart tool
(http://skyserver.sdss3.org/dr9/en/to...art/chart.asp). Therefore
we can select any object and see its position at each of those epochs.

But let's look at a few of these, and rather than trawling through
table 2 for objects, suppose I select out the three mentioned in the
abstract as being of most note, these a

PHL 1033 located at (J2000) 01 33 43.2 +03 57 36
LB 8956 at 08 57 26.7 +18 55 24
LB 8991 at 08 58 30.1 +18 37 07

For these three, I've taken an SDSS-DR9 finding chart for the 2005
epoch, and a DSS POSS-I cut-out for the 1960 epoch, and you can see
each epoch side by side at these 3 addresses:

http://quasars.org/goodies/PHL 1033.jpg
http://quasars.org/goodies/LB 8956.jpg
http://quasars.org/goodies/LB 8991.jpg

You will see how much better the SDSS (left side) resolution is
compared with the POSS-I (right side) resolution, but I've scaled them
to the same size, 3.5 arcminutes on a side. I trust you will not take
long to satisfy yourself that these 3 quasars have not moved at all,
not even a little bit.

Since I've shown you where to get these finding charts, and how to do
it, you can therefore check out the individual objects of table 2 for
yourself, and test for movement. I daresay your results will not vary.

Still, the news isn't all bad. Some quasars do indeed move. But it
depends on what you call a "quasar". How about this for a definition
that even the most learned amongst us could agree on: A "quasar" is
an object classified as a quasar by the SDSS-DR10 release at
http://www.sdss3.org/dr10 . Good, eh? So let's look at a couple of
their quasars:

(1) http://quasars.org/goodies/SDSS J090514.88+090424.2.jpg

These 3 images are POSS-I 1950s epoch on the left, POSS-II 1980s epoch
in the centre, and SDSS-DR10 2008 epoch on the right. Observe the
bright quasar near the centre of each photo, see how gayly it moves
across the stellar background. Then there's this:

(2) http://quasars.org/goodies/SDSS%20J0...8+231752.6.jpg

First 2 images on left are red & blue from POSS-I (1950s), the next 2
images are red & blue from POSS-II (1980s), and the right-hand image
is from SDSS (2008). Observe how the quasar and an M-star (red)
companion are moving in tandem, a stately procession from right to
left across the stellar background.

So yes, quasars do move. But, err, that means they weren't quasars at
all. Well golly gosh darn, those are white dwarf stars. But their
spectra looked quasar-like enough to fool the SDSS-DR10. This is,
firstly, because the DR10 is an automated survey which makes mistakes,
and, secondly, because sometimes the spectral quality isn't the best
and we can't be sure which it is, a quasar or a star. Then if we see
that it moves, as these two examples do, then we know know that it is
indeed a star after all.

So the bottom line is, if we find that a quasar is moving, then we
know that we were wrong to classify it as a quasar, and that it is
instead a star. You probably wouldn't find that to be a very
satisfactory outcome to your query, and indeed it isn't very
satisfactory, but it's the best we can do when working with faint
objects and faint spectra.

However, in the 1980 paper that you cited, note the reviewer's
appended remarks at the end where he makes the excellent point that if
quasars were stars, then we should expect to see most of them in the
Galactic disk -- but instead, we see them out the Galactic axes where
extragalactic visibility is best.

Hope this has helped,

Eric Flesch

In article , Eric Flesch
writes:

First, great response, Eric. I'm sure that clears everything up!

So yes, quasars do move. But, err, that means they weren't quasars at
all. Well golly gosh darn, those are white dwarf stars. But their
spectra looked quasar-like enough to fool the SDSS-DR10. This is,
firstly, because the DR10 is an automated survey which makes mistakes,
and, secondly, because sometimes the spectral quality isn't the best
and we can't be sure which it is, a quasar or a star. Then if we see
that it moves, as these two examples do, then we know know that it is
indeed a star after all.

So the bottom line is, if we find that a quasar is moving, then we
know that we were wrong to classify it as a quasar, and that it is
instead a star. You probably wouldn't find that to be a very
satisfactory outcome to your query, and indeed it isn't very
satisfactory, but it's the best we can do when working with faint
objects and faint spectra.

Just to be clear, this is not a tautology. Many white dwarfs are
actually found as mis-classified quasars, and stellar astrophysicists
follow them up with high-resolution spectropscopy etc. So, the fact
that it moves does not ipso facto make it a non-quasar, but rather is a
strong hint that it is not a quasar, and this is confirmed by followup
observations.

Excellent answer Eric, thanks a lot.

I will write to Mr Ratcliffe to
see what he has to say about this,
but the argument of quasars not being
in the galactic plane is a very strong one.

Thanks again.

In article , jacob navia
writes:
vvvvvv
In his book "The static Universe", (1) Hilton Ratcliffe publishes a
table of 40 quasars with verified proper motions (page 107).
^^^^^^^

Phillip Helbig---undress to reply replied:
Of course, if quasars were really ejected from nearby galaxies at high
speed, we WOULD see significant proper motion.

At some level, almost all quasars should have some proper motion. If
your resolution is good enough to measure it, then probably the quasar
will no longer be a point source, but has a definite shape, which
changes with time. So, it is difficult to define a fiducial point whose
motion one can detect.

Indeed, there are lots of VLBI observations of time-variable substructure
in quasars. A few examples plucked at random from an ADS search:
3C 263 http://adsabs.harvard.edu/abs/1996ApJ...464..715H
NRAO 150 http://adsabs.harvard.edu/abs/2007A%26A...476L..17A
I'll say more about this below.


But there's a larger issue: high-precision absolute astrometry
[measurement of the ABSOLUTE position of an object
with respect to some fundamental reference frame,
i.e., the sort of measurement you need in order to
measure proper motions]
is *very* difficult and horribly prone to systematic errors -- it
involves looking for *tiny* shifts in position over a period of many
years, and there are all too many instrumental effects that can mimic
these position shifts.

I think it's fair to say that almost all astrometry experts think
Ratcliffe's claims are NOT valid.


While I (and almost all astrophysicists) have a strong Bayesian prior
that quasars are very distant and hence have only very tiny proper
motions, it's still a reasonable question to ask how one could test
this.

First, a cautionary tale: In 1935 Van Maanen claimed to have measured
the proper motion of stars in nearby spiral galaxies:
http://en.wikipedia.org/wiki/Adriaan_van_Maanen
http://adsabs.harvard.edu/abs/1972QJRAS..13...25H
These observations used optical photographic plates. In hindsight the
plate measurements were wrong (systematic errors claimed proper motions).

More recently, ESA's Hipparcos mission did high-precision absolute
astrometry on ~120K stars from space, avoiding many of the systematic
errors of ground-based measurements. Alas, quasars are relatively
faint optical sources, while Hipparcos could only detect relatively
bright sources. Apart from a marginal detection of 3C 273, Hipparcos
didn't directly observe any quasar. Radio stars were used to tie the
Hipparcos reference frame to quasars
http://adsabs.harvard.edu/abs/1998ASPC..144..381L
but I don't know if this explicitly tested whether the quasar system
was internally a "rigid body".

With today's technology, radio VLBI is the best ground-based way to test
whether quasars have nonzero proper motions. Some interesting papers on
this include:
* MacMillan
http://adsabs.harvard.edu/abs/2005ASPC..340..477M
explicitly looked for proper motion in a large VLBI data set, but
concluded
"The problem, however, is to how to determine how much [of] the
observed apparent motion is due to unmodelled source structure
effects."
* Titov et al
http://adsabs.harvard.edu/abs/2011A%26A...529A..91T
detected the drift in quasar positions due to our Sun's orbital motion
about the center of our galaxy
* Bartel et al
http://adsabs.harvard.edu/abs/2012ApJS..201....3B)
describe observations done specifically to test whether certain
quasars are moving with respect to each other. The answer is
basically no to within ~ 20-50 microarcseconds/year, limited by
time-variable substructure.

ESA's Gaia mission is due to launch very soon (the current plan is for
late November). This should do high-precision absolute astrometry on ~1e9
objects, including ~500K quasars. Gaia should be able to set limits on
proper motion down to the ~20-200 microarcsecond/year range, depending on
the object's brightness.

--
-- "Jonathan Thornburg [remove -animal to reply]"
Dept of Astronomy & IUCSS, Indiana University, Bloomington, Indiana, USA
"There was of course no way of knowing whether you were being watched
at any given moment. How often, or on what system, the Thought Police
plugged in on any individual wire was guesswork. It was even conceivable
that they watched everybody all the time." -- George Orwell, "1984"

In article , "Jonathan
Thornburg [remove -animal to reply]"
writes:

But there's a larger issue: high-precision absolute astrometry
is *very* difficult and horribly prone to systematic errors -- it
involves looking for *tiny* shifts in position over a period of many
years, and there are all too many instrumental effects that can mimic
these position shifts.

Agreed.

I think it's fair to say that almost all astrometry experts think
Ratcliffe's claims are NOT valid.

While I (and almost all astrophysicists) have a strong Bayesian prior
that quasars are very distant and hence have only very tiny proper
motions, it's still a reasonable question to ask how one could test
this.

Note that Arp claims that certain quasars have been ejected from certain
nearby galaxies. In such cases, a relative proper motion between the
galaxy and the quasar would prove his claim. (Of course the galaxy has
its own proper motion, but if Arp's claim is true, the relative proper
motion between quasar and galaxy a) would be quite large, much larger
than the average proper motion of a nearby galaxy and b) would be in the
direction of the vector connecting the two.)

More recently, ESA's Hipparcos mission did high-precision absolute
astrometry on ~120K stars from space, avoiding many of the systematic
errors of ground-based measurements. Alas, quasars are relatively
faint optical sources, while Hipparcos could only detect relatively
bright sources. Apart from a marginal detection of 3C 273, Hipparcos
didn't directly observe any quasar. Radio stars were used to tie the
Hipparcos reference frame to quasars
http://adsabs.harvard.edu/abs/1998ASPC..144..381L
but I don't know if this explicitly tested whether the quasar system
was internally a "rigid body".

There was also some work, by Christian de Vegt and colleagues, using
optical observations to tie the Hipparcos frame to the quasar frame.
With photographic plates, they could just get astrometry-quality
exposures of faint hipparcos stars and bright quasars on the same plate.
This was done in Hamburg while I was there, probably one of the last
scientific observing programs at a sea-level optical observatory.

In article ,
"Jonathan Thornburg [remove -animal to reply]" writes:
With today's technology, radio VLBI is the best ground-based way to test
whether quasars have nonzero proper motions.

A network of quasars defines the International Celestial Reference
Frame for Earth's rotation. See http://iers.org and in particular
http://www.iers.org/nn_10404/IERS/EN...tml?__nnn=true
If any of the 600 reference quasars had proper motion, that would be
obvious in the data.

--
Help keep our newsgroup healthy; please don't feed the trolls.
Steve Willner Phone 617-495-7123
Cambridge, MA 02138 USA