I have an attraction to all-sky cameras ever since I developed one to detect aircraft moving in the vicinity of a laser guide star beacon we developed at the MMT twenty years ago. At that time, I saw the value of an all sky camera beyond what at the time was seen to be mearly a way of checking for clouds without having to go outside.
The MMTO has continued to run its all sky camera for many many years, and they've even developed a simple user interface to grab the data, which is stored in FITS format -- perfect for doing science.
In the past, I've poked around with satellite and meteor detection with some good results, but nothing conclusive because I never came up with an overall plan. There's still value in these projects -- I just need to come up with an actual project design.
Recently I've decided to go back to the MMTO all sky camera data -- once again without an actual plan -- and see if I could measure variable stars. The one that instantly came to mind was Algol. This is an eclipsing system in which star B passes in front of star A every 2.867 days, with the entire eclipse event taking about 10 hours.
So here's what I have to work with:
|A typical MMTO All Sky Camera view|
and here it is zoomed in on Algol:
|Zoomed in on Algol and surrounding stars|
So I've written some software to do simple aperture photometry on Algol and Gamma And and then show their relative photometry.
This data, as you can probably tell, is very noisy and pretty poor. Maybe this small project will motivate me to get ahold of the MMTO and see if I can work with them on improving their system.
In any case, the photometric results are also noisy, but show some promise. The following is a plot showing relative brightness on the vertical axis, and UT time (show in hours before and after 21 Nov 2014) on the horizontal axis.
|Relative Photometry for three consecutive evenings|
The photmetry is shown for three contiguous nights: 20, 21, and 22 November 2014 (red, green, blue, respectively). A value of '1' on the vertical axis says that Algol and the comparison star (Gamma And) were the same brightness. A value larger than 1 means that Algol is brighter, and a value less than one means that Algol is fainter. It was predicted that Algol would be at minimum on the 20th and 22nd. Looking at the plots, it seems as though I was able to catch those events. But the data is so noisy and otherwise crappy that I'm not too sure yet. I base this on the photometry for 21 Nov, which should be a pretty flat line and isn't. There is an auto-gain on this camera so maybe that's what's making the data appear so non-flat.
In any case, this is sorta fun and I'm going to work on this a bit more to see if I can improve the measurements and maybe see if I can make those "non-event" evenings look at a little more flat photometrically.