Extreme data mining for Extremely Variable Objects

[bigdenru]

Hello all!

I would like to point your attention towards the new results we have
just published at astro-ph with my friend and colleague Kirill
Sokolovsky. This is an example of "extreme data mining", as Kirill has
put it. We have discovered eight cataclysmic variables and two
optically violent variable quasars using just the photomeric data of
USNO-B1.0 catalogue alone! Quoting our article, "the USNO-B1.0
catalogue contains the 'light curves' in two passbands, although
consisting of two points only, plus one point in infrared band". We
used the fact that POSS-II Red and Blue plates of the same areas of
sky were taken on different nights, and POSS-I plates - during the
same night immediately one after another.

Since the USNO-B photometry is nominally good to about 0.3m, we were
searching for the high-amplitude objects only. Actually for those with
range of variability more than 2m, yet just the blue ones to exclude
mirae and semiregulars. And also we have limited our search to within
30" of X-ray sources from ROSAT catalogue (both faint and bright) to
the North of -10 declination.

We started our work well before Wils et al. published their results on
new CVs from SDSS catalogue. Our discoveries are all outside the SDSS
coverage. This is a good news since 8 out of 10 objects can be
observed right now (two are actually circumpolar for the northern
hemisphere folks). By coincidence, 4 of 10 new variable objects reside
in Draco constellation! Bad news is that all our objects are quite
faint, with the brightest at 15m in outburst. So they are definitely
the targets for CCD observers. But the good news, again, is that
virtually nothing is known about these objects! No type of
variability, no orbital periods, no outburst interval, not even the
exact varaibility range! So if you start following them up, you can
discover a lot of new things.

The list of objects with the 10'x10' finder charts are available at my
page:

http://hea.iki.rssi.ru/~denis/CV-USNO.html

The article is accepted to Russian Astronomy Letters (should go to
11/2010 or 12/2010 issue). Meanwhile you can download the pdf or
PostScript file from the archive of electronic preprints at this link:

http://arxiv.org/abs/1007.1798

Best regards, and Happy Data Mining!

Den in Moscow

[John Rachlin]

Den has kindly contributed a synopsis of his work to the Data-Mining
for Astronomy blog for the lay reader.
Thanks Den!

http://astrodatamining.net/2010/07/14/finding-cataclysmic-variables-with-a-click-of-your-mouse/



On Jul 12, 8:11 pm, bigdenru <de...@hea.iki.rssi.ru> wrote:
> Hello all!
> .
> .
> .
>
> Den in Moscow