[vsnet-chat 5534] V magnitudes and visual observation

["Sebastian Otero" <varsao@fullzero.com.ar>]

Dear colleagues,

There's been, there is and there will be a lot of discussion concerning the
use of V magnitudes for comparison stars in visual observation. Or the
conversion of every "visual magnitude" to the V system. Or a lot of things
like these.

Does the fact that an observation is made with the eye mean it is in the "v
scale" ?
If this observation is made with V comparison stars, the observation will be
made with the eye but the result will be different than the result we would
obtain if the comparison stars were transformed to "v" according to the
known formula.

The knowledge that the old AAVSO charts without V magnitudes were based on
this visual scale is wrong.
The L2 Puppis AAVSO chart is an example of this.
I don't have it here right now, but once I made a table and red stars were
given
as brighter than their V magnitude and blue stars as fainter. That is
totally the opposite to the v scale but it's consistent with cone vision for
naked eye stars.
So, as you see, things are not simple.

The different response of rod and cones is a key to this problem. Different
cells with different responses. All in the
same detector. A slight change in the WAY we look and different cells get
into play, meaning different results in the magnitudes we estimate. That's
why my opinion is that all these parameters must be controlled when making
an estimate.
The situation is that there are actually no "visual charts". This leads to a
mix of magnitudes. This is funny:  magnitudes in the "eye system" are used
with V system comparison stars and taken as if they were in the v system so
they are then converted to the V system based on that assumption.
If we use V as standard for visual observations (that's the case now) it is
not surprise that the use of stars redder than 1.0 is banned... Comparison
stars values would be totally wrong for the visual  observer if this wasn't
the policy...
Actually a difference between a -0.2 and a 0.9 comparison stars would be too
great for a sequence like this since, for instance, a "66" (V= 6.64)
comparison star with  B-V 0.9 would be a little dimmer (v= 6.83) than a "69"
(V= 6.85) comparison star with B-V -0.2 (v= 6.81). And this can happen in
any "official" chart right now.
Let's just suppose we have two different coloured stars like these:

A)  V= 5.99;  B-V=  2.00;  v= 6.41
B)  V= 5.99;  B-V= -0.23;  v= 5.94

Half a magnitude different. But it all depends on how you look at it. And in
your personal eye response. And in a lot of other things.
I can see both 6.00 mag. stars as equally bright. How can then a color term
be applied to my observations if they have already taken into account these
color difference?
If I have to estimate a variable of V= 6.10 and B-V -0.23 (v= 6.05) it is
supposed that I could use (yes I know that the use of red stars as
comparisons is banned, but this is just an extreme example) the "59" and
"64"
comparison stars for my estimate. But I see that the V= 6.10 star is clearly
dimmer than both comparison stars. I can not make an estimate.
The eye response depends on which cells of the eye are used. So the use of a
"v scale" is not accurate if no other parameters are given. More rules need
to be stated in order to get consistency in any system. If we want to do
photometry we can not do it that way. Not in a way in which the results are
strongly variable depending on how the procedure is done and the process is
never explained.

Furthermore, one thing is to look at a bright red star and a different thing
is to look at a faint red star. Cones need a stimullus strong enough to get
into action. With the naked eye, (that is under the brightest sky possible
for a given place and time), red stars get brighter. Through a telescope
(and the more you magnify and thus darken the sky), red stars look faint.
This can be easily checked.
This week there was a discussion in the AAVSO discussion list just about how
a red star can get more than a full magnitude brighter if you use a bigger
scope. That's true, more (coloured) light will enter the eye and stimullate
the cones.
BUT, be careful, if you increase power in the same small telescope, and the
star is faint, you will make the background darker and the red star will
lose brightness and contrast, even for the same amount (even one full
magnitude). That's why the lowest power is always recommended.
So... Is it a "v" vs "V" convertion possible? It depends on a lot of things
never stated to the observers. It is not only a question of a different
coefficient for every person. This coefficient is strongly variable for the
same person depending on the instrument, power, background sky, brightness
of the object and -of course- visual technique he/she is using.

The first stage of every estimate I make every night consists in the
checking
of the magnitude difference between the comparison stars I am going to use.
This "callibration" gives me the chance of detecting any color problem if
it exists. I only use the cells of my eyes that allow me to have a V
response and this is easily achieved depending on the color of the star you
are observing.
Extremely red stars (B-V >3) will give some trouble, since the Purkinje
effect is difficult to eliminate because you need to stare at them for cone
detection. But these are the only cases when things get difficult and they
get difficult even for electronic detectors... It IS difficult to convert
the magnitudes of such stars to the standard system.

I think that corrections to visual observations are not possible the way
things are right now.
Let me give another example, this is very clear.
Let's suppose we have Mira (omi Cet, B-V= 1.60) tonight and we will observe
it through our 7x50's with the following comparison stars from a new AAVSO
or RASNZ or xxx "official" chart (V magnitude sequence, colors <1.0):

A) "60" in the chart = V= 6.04; B-V=  0.87 (you don't care about these
numbers because they are never given) ---> v= 6.22

B) "64" in the chart = V= 6.35; B-V= -0.16 (remember, you ignore these data,
you just estimate "visually") ---> v= 6.32

The magnitude difference between A and B for the "average observer" is 0.1
mag.
Tonight, Mira is v= 6.23, the observer estimates it between A and B but very
similar (almost equal) to A.
According to the formula, Mira is then V= 5.90.

Now that we know how bright Mira actually is, let's see what the visual
observer will do.
He will use the current official standard charts (in V) and he will record
the folowing observation:

Mira= 60 (1) V (9) 64
Mira= 6.0

He will send this in and forever and ever it will be considered a visual
observation that will need to be transformed to the standard V if we want to
make useful comparisons.
So people will take the formula and this 6.0 will be turned to V= 5.66,
considering Mira's redness.
However, we know that Mira was actually V= 5.90. The reported value was a
whole lot better than the transformation applied to the "visual scale".
And the observer WAS AN AVERAGE OBSERVER, so imagine what an observer with a
better red response could cause !!!!!
The comparison stars used and their colors will never be traced properly, we
know that, and their colors are considered "ok" if <1.0....

In short: NO visual SCALE in use, NO V SCALE in use.
You can't transform unless you make an investigation that will take a
century on how the observation was exactly made!!

Even knowing every person's color coefficient would be a tremendous waste of
time for people at the VS organizations. A lot of work that, as I have
pointed out above, doesn't guarantee anything if these charts are used or if
the conditions under the observation was made are not controlled.
And at present they are not.
Scatter will always be a friend of us this way.

Self-callibration would eliminate all these problems and the need of
correction that would be different depending on who was the one who
observed.
But remember that the problem is not a color-coefficient, but the different
conditions under all the observation are made.

As I have mentioned above, I checked the magnitude difference between
comparison stars and I look at the stars of different colors in different
ways to get the V magnitudes. With the eye working this way and actually
observing a 0.2 mag. difference as a 0.2 mag. difference and not as a 0.3
(according to the tranformation), then any correction to this will be
non-sense.
If I really perceived the v magntiudes instead of V, this transform would be
possible, as long as the comparison stars were previously corrected but my
approach is to stick to V from the beginning so quoting my first example,
both V= 6.00 mag. stars will look 6.00 once I am callibrated.
In my opinion, cone and rod vision differences totally determined the way
you observe.

In my opinion, a real transformation from a non-existing "v" to V is
impossible.
But a personal callibration beforehand will be a great way (maybe the only)
to improve every database (specially the data on thousands of neglected
small amplitude variables).

Always waiting for comments on this very important issue,

Sebastian Otero,
Buenos Aires.
LIADA; CEA




---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://vsnet.grisoft.com).
Version: 6.0.386 / Virus Database: 218 - Release Date: 10/09/02

VSNET Home Page

Return to Daisaku Nogami

---

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