[vsnet-chat 4542] Re: re MisV0380

[Brian Skiff <bas@lowell.edu>]

     The errors in the current MSX catalogue are often optimistic.  I have
found numerous reliable cases where the positions are off  by 8"-10", even
though the nominal errors are 2"-3".  They are nevertheless a good indicator
of a 10-micron source like the IRAS 12-micron sources, but with much higher
sensitivity (goes fainter at that wavelength) and with better positions.
I expect that the version 1.2 of MSX will be better, as described in the
recent AJ paper on the project.  That's not available yet in convenient
form, however.
     The 2MASS positions, by the way, are reliable to 0".2 externally in
the absence of proper motion.  The epoch is recent (~1997), so most stars
will not show any motion.  Also there are zones where there are ~1" offsets
in Dec either north or south.  Presumably this will be fixed once they all the
strips-scans reduced and tied together.  My approach has been to compare
2
2MASS with GSC-ACT and A2.0 (using VizieR, not the very clunky GATOR), and
if the other two are more than ~0".5 from the 2MASS position, then the
other catalogues are probably right, and 2MASS is wrong.
     Another point in regard to 2MASS IDs (per vsnet-id 456) is to look at
the J-K color, not simply the magnitudes.  As Kato-san noted, the usual
red variables found by MISAO are very bright, even saturated in 2MASS, so
the JHK magnitudes are typically in the 5 to 8 range.  More importantly,
the J-K color (forget H) for red variables will be at least 1.2 or greater.
Thus many of the IDs with comparatively blue colors (J-K < 1.0 roughly)
appearing in vsnet-id 456 are wrong if they apply to red variables.  For a
rough correspondence, the J-K value is comparable to V-R in the visible.
Thus a K0III is J-K about 0.6, M0III is about 1.0, carbon stars much redder.

\Brian

VSNET Home Page

Return to Daisaku Nogami

---

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