[vsnet-chat 1364] Magnitude estimates from unfiltered CCD images

[aah@nofs.navy.mil]

Taichi Kato made a point that I glossed over in my discussion:  it is difficult to
find non-variable red comparison stars for red variables like Miras and carbon stars.
I specifically avoided the problem by only giving as an example a yellow variable like
a cepheid.  Miras, because of their extreme color and the change in that color during
a pulsation cycle, make magnitude comparisons from different observers very difficult.
This is true even if you transform to a standard system, since the reddest standards
have (B-V) less than 3 (and Miras are often redder than this), so that you have to
extrapolate.  Second-order extinction becomes important.  Small errors in your transformation
coefficients make tenths-of-magnitude errors in the final results.  Molecular bands
make wide-band photometry an interesting challenge.  I'm currently following a Mira
in a highly reddened region, so that (V-I) = 7.5, and have to use a single CCD+filter
in order to get internally consistent results.  I usually avoid red variables!

Kato asked:
>    Regarding this point, several observers have shown interest what system
>names can be most appropriate in reporting several types of non-standard
>filtered CCD observations (e.g. IR-blocking filter).  We'd appreciate
>suggestions in these cases, and like to know what variations of filters
>are actually used in CCD observations.

  I don't know of any standardization of names when referring to unfiltered, or IR-blocked
only, CCD observations.  You can certainly invent your own nomenclature, and it would
likely be adopted.  This might be an interesting topic in itself for this maillist,
as VSNET is probably doing more unfiltered observations than any other group.
  As for the current filter systems that are used for CCD observing
in the professional world, there are numerous systems that are used for specific
problems.  Here are a few that I am familiar with, but don't consider it an exhaustive
list.

  Johnson/Cousins  UBVRI or UBVRcIc, depending on the publication.  Since the Johnson
      R and I filters are rarely used anymore, I prefer UBVRI.  In general use.
      
  SDSS (Gunn) u'g'r'i'z'  A wideband system like Johnson/Cousins, but better suited
      for separating galaxies.  z' is of special interest since it is a nice wideband
      filter redward of Cousins I.  These filters are expensive.  While not in current
      general use, SDSS will measure millions of stars in the near future and this
      system may find much more general acceptance.
      
  Vilnius intermediate bandpass system.  Ideally suited for classifying stars, determining
      metallicity and interstellar reddening.  Filters are UPXYZVS, going from 300nm
      to 700nm.  Primarily used by the Europeans, but a very interesting alternative
      to spectroscopy.

  Washington CMT1T2 wide-band system (this is the one that has a 'C' filter).  Specifically
      developed for providing effective temperatures and metallicities for G and K
      giant stars, but used for many other applications since.  C is like U+B; M, T1, T2
      are roughly equivalent to V,R,I.
      
  HST filters.  Common ground-based filters are F555W, F675W, F814W (roughly equivalent
      to V,R,I.  Used for comparison to HST observations, they are named by their central
      wavelength in nm, plus a 'W' designation for Wide.
      
  Stromgren uvby intermediate bandpass filters.  Not as heavily used for CCD work as they
      were for PEP observing as they tend to be too blue for front-illuminated CCDs.
      Nice for metallicity determination.
      
  Wing near-IR filters, specifically chosen for measuring red stars, with continuum and
      molecular band filters.  Optec sells a set of these.
      
I'm sure this list can be extended ad infinitum.  For example, several observers are
now using R,G,B filters (like from Edmund Scientific) for tricolor imaging, and I'm sure
they will use them for photometry as well.  This will cause confusion with Johnson R,B.
The main point I guess is to make sure that some additional specification, like the
name of the filter system, but included with any observations to uniquely identify
which one you are using.

Arne

VSNET Home Page

Return to Daisaku Nogami

---

vsnet-adm@kusastro.kyoto-u.ac.jp