ASASSN data fails on VSX J201004.7+771114

American Association of Variable Star Observers (AAVSO)
Tue, 07/28/2020 - 22:02

Some time ago I used the ASASSN data to search for some interesting stars to expand my program somewhat. I looked for stars with very long periods and interesting light curves. One star I found was VSX J201004.7+771114. From the beginning on it was obvious, that this would be a very difficult object, a double star, both components are variable, especially the other component (SZ Cep, M, 328.21d, 8.7-16.0V).
After some time I compared my results with the ASASSN data and was very disappointed. Maybe I should say that I'm not the best observer, my curves tend thereto to be "edged", not always very smooth, sometimes I have outliers etc. Nevertheless, this result was disappointing. But I gave it a second try, and this time the result was finally a disaster - a fast rise, around one mag and I saw nothing! I pondered about the matter and very soon one became obvious: VSX J201004.7+771114 has always a maximum in the ASASSN data when SZ Cep has a maximum! The given period for VSX J201004.7+771114 at the ASASSN webpage is exactly the double of SZ Cep.
My question would be, can one verify this? And, should one tell this the people who run the ASASSN webpage? I'm not sure whether I should continue with this star or not. The AAVSO data shows an interesting CCD light curve around 2005 till 2009 (~ 11.0 - 12.0V; ~1000d).

N.B.: Unfortunately I cannot upload files to illustrate this. Please compare the ASASSN data for VSX J201004.7+771114 and SZ Cep with the AAVSO data.



American Association of Variable Star Observers (AAVSO)
ASAS-SN and surveys resolution

Hello Peter,
keep in mind that you can add links to the relevant data so there is no need to attach files.
I will share the two ASAS-SN light curves to illustrate what you found.

SZ Cep ASAS-SN phase plot.
Data folded with P= 322 d.

VSX J201004.7+771114 phase plot
Data folded with P= 650 d.

The first plot is that of the mira, no problem.
The second plot shows variations of the two stars superposed in a single light curve and folded with a period twice the value of the SX Cep period.
The problem is that the two stars can't be completely resolved in the ASAS-SN photometric aperture due to their small separation and bright magnitudes. They are 24" apart. If they were 15th or 16th mag. stars there would be no problem, but these objects are 10-11 mag. stars and vary a lot, especially SZ Cep, and thus their light curves are contaminated by each other. This is especially noticeable when SZ Cep gets brighter than VSX J201004.7+771114 and reaches its peak magnitude. The light from the mira gets poured over the fainter object and they get blended causing what you see in the second plot: the peaks of the mira and the variations of the second object when the mira gets fainter.

This is something typical with survey data. And ASAS-SN is nearly separating them. NSVS has them completely blended because its resolution is three times worse.

About telling the ASAS-SN team about this, actually it would be impossible to correct all variables that have contaminating companions. It's an intrinsic consequence of the survey resolution. Large pixel size, unability to resolve stars and close companions. You learn to identify these cases and interpret the data.

What we can do is correct any VSX entry that may include wrong data (periods, magnitudes, etc) due to light contamination from companions. Or even wrong identifications. We are about to correct more than 20,000 objects that were wrongly identified by surveys in an unpcoming VSX update based on the ZTF DR2 Variable Star Catalog. ZTF has a much better resolution than surveys like NSVS, SuperWASP, ASAS-3, ASAS-SN or CRTS (lots of variables from those surveys are currently wrongly identified!).

In the case of the fainter variable, the one you are observing, yes, you should continue observing it, because surveys like ASAS-SN can't do it properly due to the above issues. So our observers' data for objects like this are very useful!


American Association of Variable Star Observers (AAVSO)
Thanks Sebastian


thanks for your answer, Sebastian. I will continue observing the object, it was once a 104 comp. star for SZ Cep! Therefore a maximum of maybe 10.4 mag should be logic. At the moment I see the star around 12.5 mag, some observers made observations below 13 mag. In any way the amplitude seems to be quite large over all, but with very long cycles - in any case not an object for fast results, especially what I'm looking for!

Thanks again

Peter mpr