[vsnet-chat 2315] (fwd) Hipparcos/Tycho photometry

[Taichi Kato <tkato>]

(fwd) Hipparcos/Tycho photometry

Date: Sun, 26 Sep 1999 13:21:14 -0700 (MST)
From: Brian Skiff <bas@lowell.edu>

     Maciej Reszelski asked about the differences between the Hipparcos/Tycho
photometry and the standard Johnson V and B systems.  There _are_ differences,
and in some instances confusion can arise as it pertains to variable-star
observing.
     The Hipparcos "Hp" magnitudes result from essentially "unfiltered" data
taken with the detectors on the spacecraft.  The sensitivity curve has a very
wide passband across the visible part of the spectrum, and the magnitudes
have been adjusted in an approximate way to conform to V.  However, there is
a significant color term in these data, such that they should not be used for
any other than internal comparisons for the _same_star_ (i.e. not even
differentially), and color changes in, for instance, Miras will yield
additional problems.  The 5th/95th percentile ranges given for the Hipparcos
stars merely represent the span of magnitudes within which 90 percent of the
measurements fall.  This gives some indication of variability when combined
with knowledge of the scatter for certifiably constant stars of similar
brightness, and allows for statistical tests, etc.  Even for bright stars, this
range is typically a few percent, and for stars fainter than about mag. 8, the
per-observation scatter becomes rapidly larger (the errors are about 0.4 mag.
by mag. 10!).
     The Tycho part of the instrument observed in two filter passbands that
were similar to B and V, but again there is a substantial color term, and in
the published Tycho catalogues these raw "instrumental" data have been
transformed to the standard system and are shown explicitly.  The "raw" data
are coded as "VT" and "BT" for the yellow- and blue-light observations---these
are the ones to avoid!  The corrected magnitudes are specified as Johnson
B and V.  I don't understand why the consortium bothered to show the "VT" and
"BT" magnitudes.  The idea was, I gather, to preserve whatever internal
precision there is in the data, which any transformation will inevitably
reduce.  This seems in my opinion to be a misplaced emphasis, since the
internal errors are already 0.1 mag. at V mag. 8.0 (i.e. lousy photometry),
and 90 percent of the Tycho stars are fainter than this.  Another result has
been that those unfamiliar with photometric systems (including professional
astronomers) have confused the two sets of data, leading to all sorts of
problems.  
     Although many star-chart software packages have the H/T data incorporated,
I am unsure about which sets of magnitudes are used.  It is easy to look up
data for individual stars using the Strasbourg 'VizieR' utility on the Web:

http://vizier.u-strasbg.fr/cgi-bin/VizieR

There are mirror copies in the US (which I find slower than the French one) and
in Japan.  Ask for catalogue "i/239" in item 1 of the form this URL brings up.
It's best to have accurate (~0'.1) coordinates handy for this.

\Brian

VSNET Home Page

Return to Daisaku Nogami

---

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