Tutorial: Observing Variable Stars

[Dave Smith]

Observing Variable Stars
Why observe variable stars?   
It is fascinating to directly observe the variation in brightness of a star and ponder as to why it is varying. Some vary because the star is intrinsically varying e.g. due to it regularly expanding and contracting and some vary because there are actually two stars eclipsing one another. Some variations are highly predictable while others are not. Professional astronomers who study stars are highly dependent on the results of amateurs which mean any observations are very useful.

What equipment is needed?
Stars can be observed with just the naked eye but to be of scientific use then the minimum requirement would be just a pair of binoculars. Many variable star observers make valuable contributions to the databases using binoculars or telescope by visually comparing the brightness of the star with that of nearby standard stars over a period of time.  More precise estimates of brightness can be obtained with a telescope and camera (DLSR or CCD camera). Software (Free or paid for) is available to enable the measurements of variable stars compared with standard stars. Anyone who can take an astro-image and apply darks, bias and flats would be able to take useful variable star measurements.
When starting out it is not necessary to purchase any filters. Once you are capable of taking a reliable set of readings the first filter to consider is a Johnson V filter. Even then a filter is not vital to make a valuable contribution.

Which software?
I started off using MaximDL simply because I already had it. I would not recommend buying it (it is very expensive) for this purpose as better free software is available. There are two free astronomy applications available and I have used both. Firstly Muniwin. It does the analysis on images and produces a spreadsheet ready for submitting to the BAA VSS database and the AAVSO database. I have found, however, that it sometimes rejects some of the data for no apparent (to me) reason. More recently AstroImageJ has become available and seems far more consistent. Both of these two packages come with help files.

Collecting data
My experience has been entirely using a CCD camera for variable star observations and so the descriptions that follow are biased towards that. There are links at the end which will give guidance on visual observing or using a DSLR camera. I am restricting the software description to using AstroImageJ

Choosing a target star
Fortunately there is much help on this on the BAA variable star section database.  http://www.britastro.org/vss/
Follow the link to Observational Programmes – CCD
Eclipsing
Eclipse predictions.

I would recommend as a first choice an eclipsing binary star as the variability is continuous which makes detection of some variability highly likely on first attempt.

Finder charts
Some are available on the BAA site but I find the aavso site more useful.
https://www.aavso.org/
The charts that can be printed from there include comparison stars with their precise magnitudes. E.g. EG CEP

 

Locating the star.

Personally I think SGPro is superb for getting the star in the middle of the FOV especially if plate solving is used. There is no need to worry about orientation, just get the star centred in the FOV. Depending on your scope and mount guiding may be necessary to keep the star in position.

The task is then to obtain images over a period of time such that the target star and comparison stars are not at or near saturation.

Using AstroImageJ

Download it from https://www.astro.louisville.edu/software/astroimagej/

On first use it is important to upgrade to the latest version by clicking on “help” “update” and then going up to “latest build”.

To check that star image is not close to saturation

Open the file to be checked and then place the cursor over star and ALT-CLICK. A profile of the star will appear.

 

As long as the highest point on the graph is well below saturation point (around 65000 for a 16 bit camera) and the sky background level is reasonably below the signal level all is ok. It does not matter if there is a star in the field of view that is saturated as long as it isn’t being used as a comparison star.

Data collection, calibration and photometry

Once the data has been collected AstroImageJ can be used to do the calibration.

Click the DP button and two windows will appear. You need the Data Processor window. The other can be closed.

 

On the first line click the first blue folder symbol and navigate to where the light files are stored and click select. Then click on the second blue folder symbol and click on the first file in the series and click open.

Further down do similar selections for your master dark and master flat. If you want to use AstroImageJ to construct the masters then please read the full AIJ instructions (see links)

Once that is done click START at the bottom and the files will be calibrated and put into a sub-folder pipelineout. (You can change this name is you wish).

Platesolving

It is not absolutely necessary to plate solve the images but it can save a lot of trouble. If all the images are well aligned and there has not been a meridian flip then you may get away without the plate solving stage. I have found that plate solving the images leads to smooth running of AIJ in the photometry stage. It is also well worth installing a local version of astometry.net as explained on the AIJ forum (see references). It takes under 20 seconds per image with the local plate solver. If there are many images the computer can be left to itself while solving.

To plate solve with AIJ proceed as follows. Import the calibrated files to AIJ and select the astrometry icon.

 

Most of the selections are clear but it is important that you input the correct arcsec/pix for your scope camera and the R.A. and DEC of the target star (from the aavso finder chart). Then click START and if all is ok it will proceed with the plate solving.

Running AIJ photometry

 

Import the plate solved files into AIJ and click on the Multi-aperture-Measurements icon   The following window will open.

 

Tick boxes as shown and then click on the Place Apertures button. First left-click on the target star (the variable) and then the comparison stars. The variable will be labelled T1 and the comparison stars C2, C3 etc. When finished multiple windows will open including a plot and a measurements spreadsheet. The measurements spreadsheet needs to be saved as Measurements.txt

The plot can be configured to plot any aspect from the Measurements file including the star magnitudes (if the box on the penultimate line of the Multi-aperture measurements is checked.

Submitting results to the BAA and/or AAVSO  databases.

You do not need to be a member of the BAA or AAVSO but you do need to apply for a login. See the BAAVSS database or AAVSO website for details.

To generate files in a format to upload to either organisation, the BAA Photometry Spreadsheet is required and can be downloaded from the BAA database. There is also a full guide to AIJ and the Photometry Spreadsheet available.

I find it convenient to use Excel to plot a graph of magnitude against time as there is more flexibility with presentation. e.g.

If you get stuck at any point and cannot find any help in the references then do not hesitate to contact me by pm on SGL.                      Dave Smith  Feb 2019

References

BAAVSS website                 http://www.britastro.org/vss/

There are many links there to    VSS database

                                                            Eclipse predictions

                                                            Charts

                                                            CCD (Target lists) and many more

                                                                        Beginners

AAVSO website                   https://www.aavso.org/

AstroImageJ                         https://www.astro.louisville.edu/software/astroimagej/

AIJ user guide

https://www.astro.louisville.edu/software/astroimagej/guide/AstroImageJ_User_Guide.pdf

AstroImageJ Forum           http://astroimagej.1065399.n5.nabble.com/

Astrodennis guide to exoplanet imaging using AstroImageJ  http://astrodennis.com/

There is also much detail on visual observing on Gary Poyner's website http://variablestars.co.uk  Gary has impressivly submitted more than a quarter of a million visual observations from light polluted birmingham.

The Society of Popular Astronomy also have a guide to visual observing which is linked to on this page https://www.popastro.com/main_spa1/variablestar/

 

Edited February 4, 2019 by Dave Smith
Added more links

[Pig]

Very inspirational David, you almost talked me into giving it a try ? 

[Dave Smith]

5 minutes ago, Pig said:

Very inspirational David, you almost talked me into giving it a try ? 

only almost - that's a pity ?

Dave

[bish]

Great post Dave. I was just going to check my list of what to look for tonight as it is actually going to be clear. Living in a light polluted area means galaxy or nebula hunting is very limited. There's only so many times you can look at M13 etc!  Tracking the brightness of variable stars gets me outside to look at something that changes between sessions.  Submitting to AAVSO and BAA feels like I am contributing too.  I'm a basic visual only.

Edited February 2, 2019 by bish

[Pig]

I do think it is a fascinating subject and most certainly very rewarding ? factoring a camera in puts the mockers on it for me. Thus, by almost talking me into it is a very good result ?

[Helen]

Brilliant!  Many thanks for taking the time to put this together Dave   I've made it a sticky so it is easy to find ? 

Helen

[Dave Smith]

Thanks Helen

Dave

[DHEB]

Great post, Dave! Nothing better than a nice tutorial to inspire new observers to venture into variable star astronomy.

[Dave Smith]

On 02/02/2019 at 18:05, Pig said:

I do think it is a fascinating subject and most certainly very rewarding ? factoring a camera in puts the mockers on it for me. Thus, by almost talking me into it is a very good result ?

Some valuable variable star results are produced by visual observers. There is a guide to visual observation of variable ststs on the BAA web site. I will post a link in the morning.

Dave

 

Edited March 4, 2019 by Dave Smith
spelling

[Dave Smith]

I can't get a direct link but the page I had in mind can be found by following the Beginners link on the BAA website. http://www.britastro.org/vss/

There is also much detail on visual observing on Gary Poyner's website http://variablestars.co.uk  Gary has impressivly submitted more than a quarter of a million visual observations from light polluted birmingham.

The Society of Popular Astronomy also have a guide to visual observing which is linked to on this page https://www.popastro.com/main_spa1/variablestar/

Hope they are useful.

Dave

Edited February 3, 2019 by Dave Smith
missing link

[Pig]

Thank you David, I am off out for lunch soon but I will take a look later this evening ?

[Star101]

Thanks for the post David. Its going to be cloudy tonight, I will use my time to read through this then. In the 5 minutes I already spent looking into this I have learnt new things. Thanks. 

[JeremyS]

Thanks for posting this very helpful guide to VS observing, Dave!

 

I haven't used AstroimageJ yet, but when I do I will refer to your guidance.

[billyharris72]

Great post Dave - many thanks. I've been meaning to have another go at this with the ASI1600 but one thing that has been bugging me atmospheric correction. Do you have a view on how wide a field you can get away with in comparative photometry without doing corrections? I've seen some sites that suggest half a degree, but trying to find variables with  decent range of comparison stars in a half degree field has proven tricky.

My own attempted workings would seem to suggest that, for anything above 45 degrees, a field of two degrees would probably be okay, as the error would be only a few hundredths of a magnitude, but I could be wrong and have not seen this anywhere else.

Thoughts appreciated.

Billy.

Edited March 4, 2019 by billyharris72
Typo

[Dave Smith]

Hi Billy.  I have never had any concerns about atmospheric correction because my comparison stars have always been in the same field of view. If there are no "official" guide stars in the FOV I choose others using Guide software with the UCAC4 star database added. Mostly, however, I  use the comparison stars given on an AAVSO chart. My FOV is around 30'.  When I  used my 90mm refractor  the FOV was approx 1.5 degrees but had no problem with atmospheric correction.

https://www.projectpluto.com/

Hope that helps

Dave

[coatesg]

If I've been short of comparison stars, for a quick analysis I've looked at using the stars in the AAVSO APASS catalogue (you can access the data really easily by using Simbad). When I've requested an "official" sequence via the AAVSO process, they've used the same data.

Though, I did manage to request a set of comparison stars recently, where there was nothing usable above 14th magnitude in the FOV (I had a shade under 30' like Dave) - the errors in the brighter comparison stars were up to 20% of the range of the variable itself ? - exception made to use them as the data can be adjusted later if the comparison stars are improved by a subsequent survey. (It makes no difference for timing studies, but for getting absolute magnitudes it does somewhat!). 

 

Edited March 5, 2019 by coatesg

[han59]

Is it not time to use the Gaia data?  The accuracy is unprecedented:

https://gea.esac.esa...mentation/GDR2/

With the G, Grp (red), Grp (blue) you can calculate the Johnson-V magnitude as specified in the Gaia documentation. This is already done and available up to magnitude 17 in CDC, HNSKY and ASTAP programs.

-- Han

 

 

[coatesg]

Well that's an option, but the advantage of everyone using the same references is that they can all be corrected computationally at a later date. Hence I tend to use the official APASS mags from the AAVSO sequences.

Either way, I suspect with ensemble reduction, it won't necessarily result in much less noisy data - my data is worse than Gaia always!

[Guest]

This is an excellent tutorial Dave, I haven't done any variable star observation yet but I intend to. I'm just collecting information at the moment. The AAVSO web site is very good and I'm working my way through some of their documents.  I think I have all the necessary hardware to start. 

I was wondering if it would be useful initially to carry out some photometry work on non-variable stars in order to check I'm doing it right.

Cheers 

Steve

 

[Dave Smith]

12 minutes ago, woodblock said:

This is an excellent tutorial Dave, I haven't done any variable star observation yet but I intend to. I'm just collecting information at the moment. The AAVSO web site is very good and I'm working my way through some of their documents.  I think I have all the necessary hardware to start. 

I was wondering if it would be useful initially to carry out some photometry work on non-variable stars in order to check I'm doing it right.

Cheers 

Steve

 

Thanks Steve. You could do that but it would be just as easy and more rewarding to choose a known variable and see some variation that you know should be there.  What scope are you intending to use?

Dave

[Guest]

The reason I thought of it was that it would be easier to check my method. So I'd pick a fixed star of known magnitude. Then using the normal system of comparing with other nearby stars of fixed magnitude you work out the magnitude of the first star and it should match the know magnitude of the first star.  I know it seems a slightly pointless exercise but there is some method in the madness. If you started with a variable star you couldn't be sure of the result. Once you are happy with the method you can then go on to do proper variable star work.

I've got an 8inch SW Newtonian on an HEQ5 mount and a canon EOS 450D camera. The scope is too big for the mount really but it tracks ok so long as it's not too windy.  The AAVSO has a very good guide to using DSLR's for this work. I've done some astrophotography with ok results but I wanted to do something more sciency? The main problem I had was deciding which software to use, there seems to be quite a few about and this is where your tutorial is very useful.

Cheers

Steve

 

[AlexK]

Awesome tutorial. Very inspirational! Thank you Dave.
Though, why you are saying that eclipsing variables are the easiest to start from? Most are actually quite constant between eclipses for days. I would start from cepheids instead exactly for the reason they never stop changing (well, when they do that makes world news  ).

[Guest]

Hi Dave, I don't quite understand why you need to carry out Plate Solving on the images. 

Cheers

Steve

 

[Dave Smith]

3 hours ago, woodblock said:

Hi Dave, I don't quite understand why you need to carry out Plate Solving on the images. 

Cheers

Steve

 

You don't have to but it makes the analysis run very smoothly.  With my images there is a small drift and so I  have to make a small correction about every half hour to an hour and unless it is done manually the program will lose the star.

Dave

[Guest]

Hi Dave, sorry to be a bit dense.

The initial image files that you load into AIJ - have they been processed in anyway before hand? Looking at your light curve for V973Cep it looks like there is one point every 2.5 minutes so does each point represent one input image file? During your observing session you take a picture then 2.5 minutes later you take another picture and so on. What I mean is you don't take multiple images for each point and then stack them before inputting the stacked image into AIJ to represent one point on the graph.

Cheers

Steve