[Aavso-photometry] Time seriess for GK Persei

[Jeff Hopkins]

Hi Gary,

Good advice.

To add to this, perhaps the biggest advantage to CCD photometry over 
single channel photometry is you can get a comparison star in the 
same image as the program star thus minimizing the effects of 
extinction (usually). If you are not doing differential photometry 
and just measuring a single star, extinction will be very significant 
even at the same air mass form night to night. Unless your 
observatory is on a very high mountain top, the nightly extinction 
changes significantly even at the same air mass. I do mainly single 
channel UBV work and try to get the stars as they cross the meridian 
(highest point in the sky for them and lowest air mass). Even on 
nights of near equal air mass and right around the meridian, the 
counts for the comparison star can vary significantly. These effects 
can easily exceed 0.03 magnitudes.

I think one BIG problem with CCD photometry is it seems too easy and 
can spoil you or make you lazy. If you are looking to find magnitude 
changes in the 0.01 to 0.05 magnitude range, you must be careful and 
MUST account for things that are not accounted for with most CCD 
software.

As I mentioned above, I primarily do single channel UBV photometry. 
My current star system of interest if fairly bright and I am easily 
tracking 0.001 magnitude changes. To do that I do differential 
photometry and determine nightly extinction coefficients and air mass 
and take them into account. It is interesting to watch the data. Over 
30 minutes when the comparison star is approaching the meridian you 
can watch the average counts go up. When at the meridian and going 
away, you can watch the average counts go down. These are not big 
changes, but consistent and must be compensated for for best 
precision. This is another area with CCD photometry that people miss. 
You do not see what is going on in detail until you start your image 
processing.

And Gary, suggesting working with non-varying stars first is an 
excellent idea. If the data produced shows variations, then one must 
try to figure out how to minimize them, or follow the system and see 
if it really is varying or is just your observations that are varying 
for a non-varying source. First try to get a set of measurements to 
be close during an evenings observations with the star near the 
meridian (ideally as close to the  zenith as possible). Get at least 
three sets of data over 30 minutes (say 10-20 second exposures 
combined every 10 minutes). Determine the magnitudes. Do a standard 
deviation. Refine your technique. Do this over several clear nights. 
See if you can improve the SD. When you have done all you can, then 
you will have learned much and be able to produce the best data you 
can.

Good luck.

Jeff

At 06:11 -0700 03/24/2008, Gary Walker wrote:
>Hello Gianlucaros;

>Since you are getting into error calculation, I would make one more 
>suggestion.  Take a star that is supposed to be constant in your 
>fields.  Measure the magnitude of it on each of the 5 frames.  Take 
>a star that is close to the target star in magnitude and color if 
>you can.  Then calculate the std dev of these 5 values.  This will 
>give you a better representation of your total errors, and is much 
>more representative than the 1/sn calculation which is usually a 
>minor contributor to the real errors.  
>
>You can repeat this on several different stars if you wish, taking 
>the std dev on the 5 measurements.  The idea is this:  If these 
>stars are constant, the repeatability that you get shows you just 
>how much scatter you should expect in your target star.  1/sn is 
>basically useless to see how you are doing.  
>
>BTW:  you can discover potential variable stars this way.  If you 
>take several constant stars, and find that the std dev is similar 
>for them, but one is much larger, then it is suspect.  Remember, 
>that the fainter stars will have slightly higher scatter.  Such a 
>star requires additional followup.
>
>I have seen constant stars using this method, that vary on some 
>nights, and are constant on others.  If you see a trend in the data, 
>one must remove the extinction to make sure that its not an air mass 
>effect--but this comes later.
>
>Thanks
>
>Clear Skies
>Gary Walker
>Maria Mitchell Observatory
>4 Vestal Street
>Nantucket, Mass  02554

-- 
Jeff Hopkins
HPO SOFT
Counting Photons
http://www.hposoft.com/Astro/astro.html
Hopkins Phoenix Observatory
7812 West Clayton Drive
Phoenix, Arizona 85033-2439 U.S.A.
(623)849-5889
(623) 247-1190 (Fax)
www.hposoft.com