Stacking

[Tom Droege <tdroege@veriomail.com>]

Just for fun I used Image Scientist to stack 8 images from Data Set 
20.  The result was as expected.  A reduction of signal to noise of the 
square root of 8.  It seems to me that this is worth doing.  This gives the 
equivalent of an eight hundred second exposure.  There are more stars to be 
found at any one S/N setting.  It did not seem to take all that long.

This would give 7 data points over the 2 hour, 19 minutes that the sky is 
tracked for these runs.  This would seem to be enough to show any trend.

Taking longer real exposures really reduces the dynamic range since the sky 
brightness is so large.

I suppose that one could get the same accuracy result averaging the stars 
after detection, but then this results in fewer detections.  Any comments?

With the stack, the bright pixels sure show up.  Since there was about a 10 
pixel drift over the 1200 seconds of time covered by the eight 100 second 
exposures, the bright pixels show up as little single pixel streaks where 
the bright pixel is more or less constant.

Tom Droege