Skip to main content

I think I've found an 'uncharted' variable. What should I do?

1ponders's picture
1ponders
Offline
Joined: 2014-07-05

Hi folks, totally new here so here goes. I've done a fair bit of background reading on variable stars and photometry but have never really given it a go til now. A few nights ago I was collecting a series of images of Pluto for an animation and while watching the images come in I noticed a faint star next to Pluto gradually brighten. I took the images and ran them through MaximDL photometry function and AIP4WIN as well and came out with the same thing. A light curve that brightens by around 2 magnitudes in an hour before slowly tappering off. I've caught this same pattern on two nights now when I have had around 9 hours of data for each night. The star is initially much fainter than Pluto and then brightens to around the same. 

I've looked for an identification for the star but I haven't been able to find it in the databases here or elsewhere. 

Double Check
WGR's picture
WGR
Online
Joined: 2010-08-17

Hello 1ponders;

You did not mention the check star.  What did the check star do while the variable was changing 2 magnitudes?  What time (JD is fine) did the variable peak on these two nights?  Can you post a light curve?

 

Gary

Uncharted variable
PKV's picture
PKV
Offline
Joined: 2010-07-26

1ponders,

Have you used the VSP by entering coordinates of your field and displaying all GCVS stars?  Or utilized the VizieR program of the CDC?  If so, you may have a new discovery!  

Kevin

Follow up
uis01's picture
uis01
Offline
Joined: 2010-07-25

Once you are really sure you have got something variable (follow Gary's and Kevin's advice) it would be useful to have a spectrum to tell what kind of variable it might be.  You could ask for help with that on the AAVSO spectroscopy forum.

Thank you for the quick
1ponders's picture
1ponders
Offline
Joined: 2014-07-05

Thank you for the quick replies. I've attached copies of three Maxim graphs. The first two were done with the same data but using different check stars (binned colour camera SXVR-M25C). The third was with a mono camera, SXVF-H9. They should contain most of the info. 

 

Unfortunately the chart I downloaded from AAVSO doesn't have listed check stars for that FOV.

I have checked the databases
1ponders's picture
1ponders
Offline
Joined: 2014-07-05

I have checked the databases here, but this star is fainter than Pluto. That was how I found it. Pluto happened to be almost touching it. I was taking continuous 3 min images with the aim to produce an animation of Pluto. As the night progressed I noticed that this little star was getting brighter. So I ran a check through Maxim. I used multiple checks with different stars (hard to get clean fields around stars in that area) and the results all came out similar. 

Attached Light curves
WGR's picture
WGR
Online
Joined: 2010-08-17

Hello 1ponders

I did not see the attached light curves.  I don't think they came thru.  Did you use "File Attachements" at the bottom of the Forum Page?  You do not need an AAVSO check star, you can use several in the field, and the probablility is that two or three of them would not be variable.  You can check your Suspected Variable against several background/foreground stars.

 

Gary

Another Resource to Check
uis01's picture
uis01
Offline
Joined: 2010-07-25

You are right that there is nothing nearby that would explain your find in VSX or the GCVS.

To try to find out what might be known about the star you think might be varying see check out Aladin:

http://aladin.u-strasbg.fr

Based on the information you shared, but without having seen your images, I think your suspected variable is very red star.  You can matchup Aladin with your image to identify the star and see what might exist in various catalogs about your suspect.

I would offer to try to get a low res spectra for you, but the declination (-20) is pretty low for my site.  A spectrum would tell us a lot about your suspect target.

There are a few things you can do to double check the variablity.  First, re-run the photometry using a comp and check star that are similar in brightness to the suspect.  From what you posted the magnitude difference between the comp and the suspect appeared to be more than one magnitude.  Depending on the size of your field, there might nor be a comp and check available at a similar brightness, but this is a good reassuring double check if it is an option.  

Second, try your best to eliminate other factors, like changing calibration with temperature of the detector, airmass, sky brightness, etc.  Inspect your darks and flats in the region of your CCD where the suspect falls.  I once went on a wild goose chase for a phantom variable caused by a star draging over an intermitent hot pixel in the CCD.  

I have attached the three
1ponders's picture
1ponders
Offline
Joined: 2014-07-05

I have attached the three attachments to my last post and they show as links to me. I will try uploading again. 

 

My apologies for not introducing my self. My name is Paul Russell and I'm observing on the Sunshine Coast in Australia. I've been involved in digital astrophotography since the Canon 300D and Orion ED80 came out a few years ago. I've read quite a number of texts on photometry (Good, Berry, Burnell, Henden and Kaitchuck, and Budding and Demircan) but have not actually dabble as yet. This could change all that. 

files attached
1ponders's picture
1ponders
Offline
Joined: 2014-07-05

files attached

GSC 6289:3173 is the closes
1ponders's picture
1ponders
Offline
Joined: 2014-07-05

GSC 6289:3173 is the closes star to my suspected variable. The variable is 1.74" south from the GSC star and as is listed as mag 15.24 in the GSC catalog.  This distance is according to coordinates from The Sky but that seems a bit close looking at the image, probable nearer to 10 or 11" measuring directly from the image.

Survey data and your magnitude scale
Sebastian Otero's picture
Sebastian Otero
Offline
Joined: 2010-09-19

Hello,
The first place to check would be VSX.

If you want people to check data on your star, you should always give its position (J2000.0) and identifiers if you have them.
Make a positional search in VizieR to find information from all available catalogues.
Colors (e.g. J-K from 2MASS, B-V from APASS) will help identifying the variability type along with the light curve.

You may also try a search by position in VSX and use the link-outs drop-down menu to check if your star has publicly available data from surveys like CRTS, NSVS or ASAS-3.
If the answer is yes, combining those observations with yours you will be able to determine the star's elements.
And indeed the answer is yes, you have ASAS-3 data:
http://www.astrouw.edu.pl/cgi-asas/asas_variable/185141-2017.2,asas3,0,0,500,0,0

You can then prepare a phase plot (if you have two nights a first approximation of the period can already be obtained) and make a submission to VSX, providing all the information required (read the guidelines and ask questions here if you have some doubt).

Your light curve suggest an RRC variable but the ASAS plot indicates a very small amplitude. Your magnitude axis is not expressing magnitudes, maybe what you were referring to as magnitudes are 0.01 mags??

I haven't analyzed ASAS-3 data, do you want to give it a try?

Cheers,
Sebastian

Thank you for the feedback
1ponders's picture
1ponders
Offline
Joined: 2014-07-05

Thank you for the feedback Sebastian.I have searched VSX via a simple search, but I'll try your suggestion and see how I go. I have tried VizieR but no luck there either. 

The magnitudes on the y axis are instrument mag and have no bearing on the real magnitude but it is certainly larger than 0.01 intervals. I watched the star go from much fainter than Pluto to as bright as Pluto in about an hour. This is what alerted me that it might be a variable. From what I watched, a 2 mag jump wouldn't be unreasonable. 

Also tried the coordinate
1ponders's picture
1ponders
Offline
Joined: 2014-07-05

Also tried the coordinate search in VSX as you suggestion and nothing at that position. There are two other variables close by but I know of both of those from previous seaches. I checked ASAS and the DSS image and that's it. How does one go about analyzing the data and incorporating my own data?

 

I've added the latest chart from Maxim combining 3 nights this week, 2nd, 5th & 6th. 

I'm don't think that is the
1ponders's picture
1ponders
Offline
Joined: 2014-07-05

I'm don't think that is the star that I'm measuring. I checked the DSS image and it seems to indicates the star next to the one I'm measuring. The green lines are from ASAS, the red are centered on my star. 

 

And here's the latest graph including last nights data.

Your star
Sebastian Otero's picture
Sebastian Otero
Offline
Joined: 2010-09-19

Hi Paul,

The best way to identify a star is to look at an Aladin image and compare it with your image and then obtaining the coordinates from Aladin to search for the star in VizieR and getting accurate coordinates from astrometric catalogues like UCAC4, PPMXL or 2MASS.
In VizieR, once you get the coordinates from UCAC4 you can use them and repeat the search to obtain its separation from nearby stars.

The star is 2MASS J18514252-2017091 = UCAC4 349-158409 with V= 14.86 (SPM 4.0) and J-K= 0.33 (2MASS)

It is at 18 51 42.53 -20 17 09.2 (18" away from the brighter star GSC 06289-03173 (V= 12.79).

A separation of 18" means ASAS-3 is measuring both stars together so the magnitudes are for the blend and the amplitude of the variations from the 15th mag. star are strongly reduced. Actually there is nothing showing up above the scatter.

So you're on your own. No survey data for this star.
You need to get more observations to be able to solve it.

I am still not comfortable with your magnitude labels. An amplitude of 2 mag. would even show up in the ASAS-3 blended results.

The other nights don't show something similar to the first night and this suggests changing amplitude. This would be okay for a multiperiodic GDOR variable. The colors of this star support it, but then again the amplitude should be a few hundredths of a magnitude and not 2 magnitudes.
A re-reduction of the data using proper magnitudes would be a good thing to do, and also checking what is exactly being displayed in the brightness axis of your plot (maybe the program settings are not displaying standard magnitudes).

Choose some APASS stars in the field (APASS mean magnitudes are available from the UCAC4 entry in VizieR) to use as comparison stars.
I assume all of the observations are unfiltered, is that correct? If so, use the V value from APASS and your resulting magnitudes will be flagged as CV.

Cheers,
Sebastian

Thanks you very much for your
1ponders's picture
1ponders
Offline
Joined: 2014-07-05

Thanks you very much for your help Sebasian. This is the sort of info I'm looking for. I'm pretty good on the basics and ok on much of the theory, its the where to look thats the problem. I'm fairly confident on all the readings, for all the right reasons :) but its the 'where to go from here'. I'm well on the way to collecting more data as we're having a blessed run of crystal clear cool nights and an 8 hour window to collect data. Fingers crossed its going to stay like this for then next couple of months, and it usually does this time of the year. So with a new 12" RC coming this Friday and with mag 5.8 skies I'll be in astronomy heaven, once the moon disappears. :D 

A technical question. I'm imaging at 2.14 arcsec/pixel for 180 sec ( camera 2x2 bin). My f/9 scope has been focal reduced to f/6.2 and my seeing is pretty ordinary most nights 2.5-3.5 arcsec at best, often worse. Would I be better off doing away with the binning and staying with just the f/6.2 focal reduction and image for a longer period at 1.07 "/pix? Coming from a 'picture' background I'm aware of the importance of the correct sampling. In this case my thinking would be to go with the 1x1 and oversample for the benefit of the occassional good seeing. Would the same apply in this situation? I'm also conscious of the fact that its a very faint star and I want as much sensitivity as possible as well. 

Could you point me in the
1ponders's picture
1ponders
Offline
Joined: 2014-07-05

Could you point me in the direction of some more suitable software for keeping track of the star. I'm currently using MaximDL which I'm not sure it totally suitable.  I haven't been able to work out how to use the magnitude of a comparison star to get an approximate magnitude reading. The only other software I have that does variable curves is AIP4WIN and again I'm not sure if I can do comparisons. 

Software
WGR's picture
WGR
Online
Joined: 2010-08-17

Hello 1Ponders

Maxim is a fine software to keep track of this variable.  Have you read the 'help" file for the photometry tool?  It should take you thru an example of how to use it.

I have over 30,000 observations, all reduced with Maxim.  It is a class software.  After reading the help file, let me know if you have any questions, I would be glad to answer. 

Write me the steps that you are using, and I can see where you are getting hung up. 

 

Gary

Thanks for the offer Gary.
1ponders's picture
1ponders
Offline
Joined: 2014-07-05

Thanks for the offer Gary. I'll look at it in much more detail now with that recommendation.

Thanks Gary for the
1ponders's picture
1ponders
Offline
Joined: 2014-07-05

Thanks Gary for the suggestion of a bit more reading in Maxim. I've attached a chart with the observations over a 9 day period including a reference star magnitude of 15.12 (from UCAC A2.0 is the best I have atm). The reference star shows as the grey line through the middle of the variable graph. Its still early days but a quick use of Maxim's folding period function seemed to give a fairly smooth (except for that anomaly) at 13.65 hours. Am I on the right track here? Or am I jumping the gun a bit in the period assumption?

Redid the phase period to
1ponders's picture
1ponders
Offline
Joined: 2014-07-05

Redid the phase period to 1.1575 days. 

RRAB with Blazhko effect
Sebastian Otero's picture
Sebastian Otero
Offline
Joined: 2010-09-19

Hi Paul,

Now that you got more data and added those plots, we can say almost for sure, that this is an RRAB with strong Blazhko effect.
The amplitude of the first cycle looks extremely large and there is a lot of scatter at maximum, maybe that night wasn't photometric? Anyway the changing amplitude is apparent. I don't remember having seen such a strong change.
The period is 1.13/2 since pulsating stars show only 1 maximum and 1 minimum per cycle.
You are close to a VSX submission.
You need to determine a time of maximum, use that maximum as phase 0 in your phase plot (times of minima are used for eclipsing binaries, RV Tauri and dark-spotted stars) and improve the period as much as possible.
Also add the 2MASS name of the star to the plot.

It would be great if you can keep collecting data to confirm the changing amplitude behaviour.
And then you should submit the star to VSX, read the VSX guidelines to see the requirements and if you have any questions, let us know.

Cheers,
Sebastian

New variable
TCB168's picture
TCB168
Offline
Joined: 2010-07-26

1ponders wrote:

Redid the phase period to 1.1575 days. 

Very interesting Paul

Now you should be able to get another maximum with a reasonable prediction of when it will occur.

 

Terry

AAVSO 49 Bay State Rd. Cambridge, MA 02138 aavso@aavso.org 617-354-0484