Re: brief note on astrometry of Mark IV comatic images

[Andrew Bennett <andrew.bennett@ns.sympatico.ca>]

On Wed, 31 May 2000 23:13:36 -0500, you wrote:

>Uhh!  Sorry I don't understand what the 15% relates to.
rms due to confusion for weakest sources (i.e. most of them)
is 15%. Random error, 1 image, 20% (you say, further down).
But if you average 2 images, you are down to 14% random; 4 gives
you 10%. But the confusion error is still 15%.

>  The answer I want, 
>is how many of the 6000 stars have a neighbor within say 5 pixels.  Or put 
>another way, How many stars out of the 6000 have an error in measurement 
>due to a neighbor that exceeds the noise due to the sky.  I bet this number 
>is a lot smaller.
Sorry; you lose! As I say in another post, stars don't just stop at 
M=15. All the stars we can see are surrounded by a horde of stars we
can't see with the MK IV. If there are 6000 above 5 sigma, there
are something like 67000 above 1 sigma, 190000 above 0.5 sigma etc.
The odds against having some of these in just the wrong place
ain't good!

And that's without including clustering.

Andrew Bennett, Avondale Vineyard, Nova Scotia, Canada.