Re: Analyzing Data

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

On Sun, 4 Aug 2002 17:29:22 -0400, Michael Richmond wrote:

>
>  Doug wrote:
>
>> I suspect that a very successful minor modification to the
>> WS statistic would be for it to calculate residuals from a
>> given night's mean magnitude instead of a global mean magnitude.
>> This would get rid of the effects of zero-point changes which
>> aren't currently calibrated out.
>
>  Doug, can you clarify that last statement?  The current pipeline
>gives each image a zero-point value based on the instrumental magnitudes
>of Tycho stars in the frame.  If two frames cover the same patch
>of sky on two nights, they should share the same Tycho stars.
>I don't see how large zero-point offsets could remain; perhaps I'm
>looking at it the wrong way.
>
>                                    Michael

A long time ago, before I quit on TASS in favour of
pruning grape vines, I was beating my head against just
this problem in DS20.

The basic problem seems to be that two frames on different
nights don't cover the same patch of sky, don't share the 
same Tycho stars and do manage to have unexpectedly large
zero-point offsets. What is worse is that these offsets
seem to manage to afflict individual stars without
affecting all nearby stars in the same way ...

I spent several months trying to get ensemble correction
to work using a selected area smaller than the whole
image and, as programmed be me, it did damn all good.

In order to look for tree branches and roofs in the data,
one needs to convert from celestial back to local
coordinates e.g. HA & elevation. Could somebody remind
me of the right way to do this?

Andrew Bennett, Avondale Vineyard