[vsnet-chat 181] V and v - comparisons

[W S G Walker <astroman@voyager.co.nz>]

Berto Monard wrote:
>
> Dear readers,
>
> I would like to touch on the possible discrepancy in the reporting
> of visually observed magnitudes.
> In order to do the correct thing we must use v sequences, not V.
> How many charts have v sequences?
>
> To avoid any misunderstanding on my side: I assume V is meant
> to be photometric V ( = photopic response) and v is visual ( =
> 'mostly(?)' scotopic  response). B magnitudes originate from
> measurements in the blue B-band as defined...
> In general it can be stated that v = 0,6 V + 0,4 B. This means that in
> the case of red objects v -V often > 0,5.
>
> Question is, how should we go about it? If the sequences are
> given in V and B we can convert them to v.  Is that practical?
> Guidelines please!

This topic was discussed at Auckland in the 1970s when a group was
preparing V magnitudes for RASNZ VSS charts. The co-efficient quoted by
Brian Skiff varies widely with different people which is only the first
problem. A comparison of pe and visual measures over a couple of decades
an empirical value for Auckland visual observers as compared with the pe
measures in V was probably around the 0.25 mark.

The effects could be reduced by matching the colurs of sequence stars to
the variables. When this was suggested the idea was dismissed as too
complicated. In most cases there are not enough stars of the correct
colour to provide an effective colour-matched sequence. Also, in the case
of the red variables, there are very few stable stars with B-V 1.6+. The
same might be said for dwarf novae, as most O and B stars seem slightly
variable. K and M giants seem to have variations of 0.03 magnitudes or
more on a time scale of a decade or so. Reddening alters the B-V value so
for real accuracy the U-B must be considered as well.

Sequences with all F and G type stars were suggested to make the effect
smooth. But even this seemed difficult and in practice we have A and K
stars mixed with F and G. Thus the values as the eye sees them vary by as
much as 1.5 * 0.25 or whatever.

The question comes down to what you're trying to prove with the visual
results. If a Mira star has an amplitude of five magnitudes then the
usual +/- 0.1 to 0.5 of visual measures doesn't matter much. Epochs of
maximum (or minimum) can be determined quite well from a dozen or so
measures. The same with outbursts of dwarf novae, although here the
subjectivity of visual observers comes in. DN seem to be measured too
bright when compared with pe, RCrBs too faint. Wishful thinking or some
real colour effect?

As far as visual measures of superhumps or similar - say eclipses of Z
Cha or OY Carinae - on the occasions we have compared pe measures with
visual we've wondered if we were looking at the same star.

One star where we have used both pe and visual measures extensively and
successfully is BH Crucis. B-V of this star in 1970 used to vary from
+2.5 (max) to +3.5 (min). The range is now +2.8 to about +4.5. Whether
this is real - the period has increased from 420 to 520 days so the
colour should become redder - or is simply a calibration problem we're
unsure and can't really check out there! But the visual measures are 0.2
fainter at maximum and about 0.8 at minimum.

Another star with extensive pe/visual measures is L2 Puppis. This has a
B-V of about +1.5 and a negative U-B. It gets bluer at minimum. Here the
visual and pe measures are quite close and show no systematic trends.

About the only other things you can do with visual measures is try to
determine amplitudes of variation and changes in these. Visual measures
are adequate to show real changes in the individual maxima of DN and
Miras, but are not sufficiently accurate to show any long term trends -
the pe measures of BH Crucis show that maxima have been increasing in
brightness as the period lengthened, but this was not evident from visual
measures.

Another factor which bears on this is the so-called personal equation.
Can it be removed from the measures? From personal experience this seemed
affected by how much wine, or how little sleep, the observer had. But
quite obviously there is something there. In analysing the visual
measures of BH Crucis we obtained the best results from using one
homogeneous data set from one observer. Removing all personal eqs to
match up datasets seems a horrendous exercise. With Mira stars this
probably doesn't matter - but the pers eq makes visual measures of
short-period, low-amplitude Cepheids by a mixture of people virtually
useless.

The point I am making at some length is that in the areas where visual
measures are most valuable - measuring the long-term behaviour of the
large magnitude variations in stars the discrepancies between v and V
don't matter. Any variations between the two systems are submerged in the
noise of the visual measures or can be scaled out when comparing light
curves.

A concluding thought - atmospheric extinction is probably equal to or
greater than the colour term discussed by Berto. A mid-B comparison - and
there are some on the VSS charts - will fade by 0.3 to 0.5 magnitudes
when compared with a mid-K star over a range of about 65 degrees,
depending on how sensitive the observer's eye is to the near UV.
Unfiltered photography finds lots of "variables" because of this.

Regards,
Stan Walker

VSNET Home Page

Return to Daisaku Nogami

---

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