[Aavso-photometry] Choosing Stars for F.O.E.

[Richard Miles]

Arne wrote:
----- Original Message ----- 
From: "arne" <arne at aavso.org>
To: "Greg Crawford" <gc at nelsonbay.com>
Cc: <aavso-photometry at aavso.org>
Sent: Thursday, November 10, 2005 2:15 PM
Subject: Re: [Aavso-photometry] Choosing Stars for F.O.E.


> Extinction is much simpler; the primary choice is a star that is
> constant. :-)
>
> There are two common techniques:  the Hardie method, which uses pairs
> of stars (one at high airmass, one at low airmass) observed nearly
> simultaneously.  The other is to follow a single star from near the
> meridian to high airmass.  This latter method requires photometric skies
> over a longer period of time.
>

There is a third as-yet unpublished technique which is very accurate for 
deriving 1st-order extinction (FOE) provided you are just wanting to work 
with V magnitudes.  It is based on the Hipparcos photometry, which is by far 
the largest and most accurate dataset for working in the V band.  Let me 
explain.

The idea is that you select a subset of 'constant' Hipparcos stars that span 
a narrow color range.  You move your scope around the sky from star to star 
measuring them singly.  The stars you pick should span the range of airmass 
that you want to test - for FOE the typical range will be 1.0<secZ<4.0.  You 
then plot v-V against secZ to get your extinction plot, where the slope is 
the FOE and the intercept at secZ=0 is your exo-atmosphere zeropoint, Zv, 
which is your holy grail.

Easier said than done you might think.  Well I am in the process of culling 
the requisite subset of data together with Excel spreadsheets that serve as 
an observing guide and exposure calculator.

I have finished one subset, which I used all of the 2004/5 observing season. 
This comprises 311 stars in the color range, 0.800 < B-V < 0.851.  They 
range across the entire sky.

Am two-thirds of the way through a second subset, which should end up with 
at least 600 stars in the color range, 0.010 < B-V < 0.041.  The idea of the 
two color ranges is to get a handle on any 2nd-order term as well as to be 
able to pick the better set depending on your targets - the 'red' subset are 
ideal for asteroids say.  Deriving an accurate 2nd-order term allows 
extrapolation of the FOE out to the typical colors of red stars.

When I carry out an extinction observing run, I usually take 7 images of 
each star to be sure of the stats before moving on to the next star.

Example of typical results:
On 2004 September 18, I carried out 3 short extinction runs using a total of 
14 stars with the following outcome in V:

20:12-20:37 UT    4 stars    kv=0.217    Zv=-18.768
21:30-22:06 UT    6 stars    kv=0.223    Zv=-18.768
22:53-23:09 UT    4 stars    kv=0.220    Zv=-18.782

Mean:                    kv=0.220    Zv=-18.773
St. Deviation:        kv=0.003    Zv=0.008

I also use the I-band Hipparcos data but this is not usually nearly as 
accurate.  However, for the record here is what the simultaneous I-band data 
came out as:

20:12-20:37 UT    4 stars    ki=0.146    Zi=-16.840
21:30-22:06 UT    6 stars    ki=0.138    Zi=-16.834
22:53-23:09 UT    4 stars    ki=0.146    Zi=-16.852

Mean:                    ki=0.143    Zi=-16.842
St. Deviation:        ki=0.005    Zi=0.009

As you can see there is definitely the third option, the Miles method for 
FOE.

Clear skies,
Richard Miles