[vsnet-chat 1412] USNO-A2.0 magnitudes

[bas@lowell.Lowell.Edu (Brian Skiff)]

     Berto Monard mused a couple of days ago about whether USNO-A2.0 magnitudes
would be better than A1.0 for getting rough sequences for southern variables.
The answer is certainly "no", at least not without some correction for field-
to-field variations in zero-point.  As a further example to the ones I posted
from many fields a few weeks ago, here is a comparison of USNO-A2.0 magnitudes
against a photoelectric sequence around the faint Cepheid AA Normae.  All the
sequence stars are relatively uncrowded.
     The columns show the star ID followed by the observed V and B-V.  Next are
estimated R [from V-R = 0.508(B-V) + 0.04] and calculated B [simply V + (B-V)].
The columns mr and mb are the USNO-A2.0 red and blue magnitudes.  Finally, the
differences mr-R and mb-B are shown.  As can be seen, in this field the A2.0
magnitudes are systematically much too faint, especially 'mr'.  (Star F has
'possibly erroneous' magnitudes in A2.0, but the offsets are no different than
for the fainter stars.)  Applying Kato-san's A1.0 conversion to mv would help
little, since the implied A2.0 colors are close to zero.  The mean offsets are
shown at the bottom of the list with their per-star standard deviations on the
last line.  The scatter of 0.3 mag. is about as expected for photometry from 
digital scans of a single small-scale plate.


AA Nor sequence from van den Bergh et al. 1985 ApJS 57,743

ID    V    B-V      R     B      mr    mb    del-r del-b
F   12.98  0.41   12.7  13.39   14.6e 14.2e  (1.9  0.8)
G   13.60  0.40   13.4  14.00   15.1  14.6    1.7  0.6
H   14.55  0.48   14.3  15.03   15.8  15.3    1.5  0.3
I   14.21  1.36   13.5  15.57   15.8  16.0    2.3  0.4
J   13.41  0.42   13.2  13.83   14.8  14.8    1.6  1.0

                                mean offset:  1.8  0.6
                              std deviation:  0.3  0.3


The sequence stars span less than two magnitudes in brightness, so no scale
error is obvious if present.  The red offset of the single moderately red star
suggests there may be a color term in the A2.0 red system here, but many more
stars would be need to be sure of this.  In sum, simple zero-point offsets
seem to be all that's required to bring the catalogue stars onto the standard
system in this particular region.
     Similar comparisons with numerous sequences all over the sky sugget the
A2.0 magnitudes are usually a bit too bright compared to standard B and R,
but clearly there is a large variation in these offsets from zone to zone.
Nevertheless, for small regions of the star catalogue, it is probably possible
to use sequences such as this to adjust for both zero-point and a color term.
This would put the A2.0 magnitudes systematically on the standard system, so
that photometry done from a large number of such stars would have fairly small
errors.  Per-star errors would not be improved; only the systematic errors can
be corrected in this way.  If several sequences appeared on one survey field,
all the stars could be used to improve the transformation for larger regions.
Anyway, this technique might be useful as a stopgap in the absence of properly
observed comparison stars for transient objects.  But the USNO-A series data
are not yet ready to be used as they appear in the source catalogue.

\Brian


P.S.:  As I finish writing this, it appears Berto has done just what I've
suggested here, applied to the case of BL2005-489, posted to vsnet-alert.  The
nearest sequence is the GSPC group about 1.5 degrees away.  I'll be curious to
know how it works!

VSNET Home Page

Return to Daisaku Nogami

---

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