Re: re HD 155229

[Stupendous Man <richmond@a188-l009.rit.edu>]

  Tom wrote:

> Suppose I extracted all the constant stars.  I have spent a lot of time 
> just looking at star plots.  Most are quite constant.  I could now make a 
> catalog to replace the present Tycho catalog with the magnitudes determined 
> by the tass pipeline.  By picking stars with lots of data points and using 
> say mag 8-10 the errors would be less than the known errors in the present 
> catalog.  There would be a lot more stars for comparison.  This is 
> obviously a bootstrap operation.  Seems to me that this might be much 
> better than using the present catalog where the magnitudes are known to be 
> not so good???

  Ummm, while there is some merit to this idea, I don't think that
one can _yet_ replace Tycho with a Mark IV catalog.  There are several
different reasons.
 
  First, the Tycho catalog was compiled by a spacecraft, which rotated slowly
and observed simultaneously stars very far apart on the sky.
As a result, it is nearly immune from one class of systematic
error: variations in sensitivity from one region of the sky
to another.  The Mark IV cannot observe a star at RA=0 and
at RA=180 at the same time, or on the same night.  One can
imagine ALL KINDS of effects which would vary with the seasons:

              - ambient temperature (hence, camera temperature)
              - extinction in the atmosphere
              - dust on the lenses
              - focus variations with temperature of the lenses
              
It is quite possible for any of these to cause a small change
in overall sensitivity over long timescales; the result would
be, for example, stars in Aquila appearing 3% brighter than
they should relative to stars in Taurus.

  Second, I think we may not do an adequate job of measuring and
removing extinction during each night.  I haven't had a chance to
check for errors caused by extinction -- has anyone else?  There
could be small, but significant, changes across a single field;
there could also be color-dependent extinction which we are not
taking into account.

  There's an expression in the photometric business: "closing the
loop."  It refers to a hypothetical moment when one has been observing
a system of stars for an entire year, so that one sees again for the
first time the stars one observed at the start of the project.  If one
has been calibrating stars relative to each other, one has information
like this:

       star 1,  at RA =  0 hours    is mag  2.0
       star 2,  at RA =  3 hours    is mag  2.5  (relative to star 1)
       star 3,  at RA =  6 hours    is mag  1.8  (relative to star 2)
       star 4,  at RA =  9 hours    is mag  2.3  (relative to star 3)
          ...
       star 10, at RA = 22 hours    is mag  2.3  (relative to star 9)

and now

       star 0,  at RA =  0 hours    is mag  1.9  (relative to star 10)

  Whoops!  We started out with star 0 at mag 2.0, but when we observed
it again, it was 0.4 mag brighter than star 10 ... which made it mag 1.9.
Uh-oh.  It really should be 2.0, of course.  The relative measurements
must have drifted somewhere between the start and end of our survey.

  Until one has "closed the loop" at least once, one can't convince
anal-retentive photometrists that one's photometric system is even
internally consistent.  Most people would demand two years worth
of good, consistent measurements before they would consider a system
established on a sound footing.

  Tom, I think we've made a good start, but it's going to be some time
before we can think about replacing reference catalogs with our own work.
I know that Arne has thought about this a lot, since his plan is to
do exactly this sort of work.

                                             Michael