[Aavso-photometry] Color Photometry

[arne]

Stan Walker wrote:
> I think that Jeff Hopkins has largely covered what you mentioned so no need
> to say this again. The measures usually end up published somewhere but the
> AAVSO professes to collect these in a more usable format than having to
> refer to an old IBVS or some other publication. I just don't think that
> separating out U B and V makes the measures more usable.
> 
Making them computer-accessible makes them more usable.  Whether you
need U,B,V, or V, (B-V), (U-B) depends greatly on your project and
whether other observers have data and in what format.  It makes little
difference how the data is stored, as long as it can be retrieved
effectively and correctly.

> I was trying to get across the mesage that colour photometry gives
> information about physical changes in a star during a cycle. Noone would
> waste time doing time series photometry - which is largely what the AAVSO
> database is all about - in U. So I was hoping that one day true colour
> photometry could find a home in a database where it was instantly
> recognisable.
> 
Here I beg to differ.  I did many nights of U-band time series of cataclysmic
variables like WZ Sge, and the companion to Mira, while at USNO-FS.
Just because you don't take data in a certain way does not mean someone
else is at fault if they use a different technique for observations.
Again, the primary point is whether the data can be archived and then
retrieved.  Having one's data sit in a drawer (or hard drive) rather than
submitting it to a public archive makes little sense to me.  Color
information on many variables, such as cataclysmic variables, is
extremely difficult to do (they change brightness from frame to frame
as they flicker), and in other instances, is unimportant (the variation
is colorless in most CVs, for example).  Again, the answer is to store
the data in such a way as to be properly retrievable by many different
researchers.

I agree that about half of my personal observations are multifilter,
properly transformed, in the V-plus-color format, and I prefer this format
for much of my own research.  Those observations do help in the physical
insight for a particular set of problems.  The other half of my data
consists of single-filter time series, or even unfilted observations,
to answer other questions.  So however the AAVSO stored my data,
half of it would require modification to retrieve and use correctly.

> A point that Jeff makes is that some time and effort has gone into
> transforming UBV and JH into the international systems. Thus the measures
> are very reliable. In contrast, Elizabeth Waagen, in her recent post about
> LPV maxima and minima has commented about the disparities in CCD measures,
> presumably not reduced to any standard system. So mixing them together does
> a disservice to the people who provide very reliable measures.
> 
I'm not clear here - you can select almost any subset of data on a
particular star, so someone can retrieve your reliable measures quite
easily.  How else are we supposed to store data?  Yes, there are
discrepancies between visual observers and CCD/PEP observers, and between
individual CCD observers as well for a given star.  I intend to spend much
of this next year looking at these discrepancies and solving their offsets,
probably largely by creating new and better sequences.  Transformation
is not always the biggest error source.

> I was looking at stars using the plotting procedures to see which were
> interesting - no particular criteria for selection - and noticed some rather
> startling differences in magnitude of the type that Elizabeth mentioned. Now
> anyone submitting an R or I measure is not really contributing to the
> classical TSP database but to a different one which is more related to the
> physical nature, or changes in this, of the target star.
> 
> Taking this to its conclusion, the main database should include only visual
> or pg and the equivalent V measures, whether by CCD or pep. Anything else 
> should
> be in a separate area or areas. Hopefully this will come in time.
> 
I must disagree.  All measures of a star should be stored in a common
location, as long as tools exist to pick and choose whatever set of
observations you need for your research.  Having multiple databases
is a situation ripe for error, difficult to maintain and makes data
retrieval complex. Ic-band measures might be available because the
field is heavily reddened and the observer can't measure the star
at U or B, not necessarily because it gave more physical insight.

Please remember that not all variable stars and their comparison stars
are created equal.  You may have used a given comp star in your PEP days
that was a degree away from the variable, something that cannot be
done with a CCD.  A new faint CV might be discovered, where no
standardized stars exist in the field.

It takes time for quality reference stars to be selected and calibrated,
so data evolves with time. We do our best with our limited staff - if you
want to lend a helping hand, you are always more than welcome.
Arne