[Aavso-photometry] Filters to minimize differential extinctioneffects

[RICHARD MILES]

Hi Gary,

The observatory is at 75 metres altitude and located in southern England 
(IAU code = J77).

Richard

----- Original Message ----- 
From: Gary Walker
To: rmiles.btee at btinternet.com ; tom_krajci at tularosa.net ; 
aavso-photometry at mira.aavso.org
Sent: Friday, June 27, 2008 12:44 AM
Subject: Re: [Aavso-photometry] Filters to minimize differential 
extinctioneffects


Hello Richard;

I would be curious as to the approximate location and the elevation of these 
measurements.



Thanks

Clear Skies
Gary Walker
Maria Mitchell Observatory
4 Vestal Street
Nantucket, Mass 02554




-----Original Message-----
From: RICHARD MILES <rmiles.btee at btinternet.com>
To: Tom Krajci <tom_krajci at tularosa.net>; aavso-photometry at mira.aavso.org
Sent: Thu, 26 Jun 2008 6:44 pm
Subject: Re: [Aavso-photometry] Filters to minimize differential 
extinctioneffects


Tom,

Further to Arne's comments, I would add that the type of CCD camera may
matter when working unfiltered.

CCD response will be significantly different if you use a Sony chip (more
blue-sensitive) compared to Kodak, Marconi, etc.  My feeling is that
unfiltered photometry is best if the variable star has a colour index close
to the 'mid-response' wavelength of your camera.  You can find this value in
say B-V terms by imaging a range of red-blue pairs.  You will find working
unfiltered that if you plot the transformation coefficient against the mean
colour of the red-blue pair, its value changes sign around this point.

Here's an interesting study I did fairly recently:
Using three cameras each equipped with a Sony chip (SXV-H9 camera), I
measured extinction using sets of Hipparcos stars selected to be in two
different colour ranges (differing in colour on average by 0.80 mags).  For
each extinction coefficient I measured between 4 and 10 Hipparcos stars at
different altitudes up to X=4.  This was done simultaneously using a V
filter, Ic filter and Unfiltered (cl) on each of six different nights.  This
is what I obtained.

Red stars (0.80<(B-V)<0.85)
Night          k(i) k(cl) k(v)
Night 1:  0.075 0.183 0.204
Night 2:  0.108 0.221 0.253
Night 3:  0.117 0.254 0.276
Night 4:  0.121 0.243 0.298
Night 5:  0.151 0.251 0.269
Night 6:  0.158 0.303 0.339
Mean:    0.122 0.243 0.273

Blue stars (0.01<(B-V)<0.04)
k(i) k(cl) k(v)
Night 1:  0.081 0.220 0.205
Night 2:  0.121 0.265 0.251
Night 3:  0.139 0.291 0.268
Night 4:  0.146 0.316 0.320
Night 5:  0.139 0.302 0.274
Night 6:  0.162 0.372 0.369
Mean:    0.131 0.294 0.281

So from this you can see that stars having a B-V colour of close to +0.3
will appear to dim at about the same rate when measured unfiltered as do
stars measured with a V filter!  This is close to the mid-response of the
CCD chip which is about +0.5 in B-V terms derived using red-blue pairs.

If stars have a 1.0 mag difference in B-V colour, then you can expect them
to change by an enormous 0.06 mags for each airmass when measured
unfiltered.  So you can see how critical it is matching the colour of the
comparisons star to that of the variable.

I am not sure when working unfiltered whether a CCD camera fitted with a
Sony chip has any advantage for measuring CVs (which are relatively blue in
colour) when compared to a more red-sensitive chip.  If you follow the CV
over a wide range of airmass will the Sony camera be less prone to
differential extinction effects?  I suspect not as this second-order effect
is >90% contributed by the properties of the atmosphere and not the sensor.

Hope you get on O.K. with the filters.

Cheers,
Richard Miles