[Aavso-photometry] Differential transformation

[Arne Henden]

Chuck's answer is essentially complete, and Radu has an interesting
technique, but I thought I'd toss in a couple of more comments.
   First, even if you are using a single field, don't forget that
you cannot always neglect extinction.  At high airmass, there is
a significant airmass (and extinction) difference between the
top and bottom of even a small field.  What is "significant" depends
on the desired accuracy.
   Second, please don't use my normal field calibrations for
determining your coefficients.  Please use Landolt (and Cousins)
standards only.  While I am a careful observer, I will not guarantee
that my field calibrations are exactly on the Landolt system, and
if you use my calibrations, you basically remove yourself one more
step from being on the true standard system.  My field calibrations
are primarily for determining the magnitude and color zeropoints
for differential photometry.  I know I am pretty darn close, and
for rough values you will be fine.  For precise measures, a reviewer
will question the use of Henden photometry in this manner as systematics
have not been investigated.
Arne


Radu Corlan wrote:
> On Tue, 28 Dec 2004, Chuck Pullen wrote:
> 
> 
>>Geir - in an ideal world, your approach would be perfect.  However, for
>>determining transformation coefficients, you want a wide range of color as
>>well as stars of about the same magnitudes so they have about the same
>>signal to noise across the color range.  And, of course, the stars need to
>>be non variable, and calibrated against a known data set, such as Landolt
>>fields.  So, the odds of finding all of these requirements in your 15' by
>>15' image for a given part of the sky are pretty slim!
> 
> 
> Chuck - one can use Geir's approach and combine stars of different colors 
> from different fields, they don't need to be all on one field. If you have 
> enough stars with good photometry (for instance some of Arne's sequences, 
> which are spread across the sky) the errors reduce pretty well.
> 
> I use that approach (see http://astro.corlan.net/gcx/html/node9.html ) for 
> a description and some example graphs.
> 
> Radu
> 
> 
> 
>>The good news is that there are fields that do meet all these requirements,
>>such as M67 and NGC 7790.  No extinction calcs are necessary, and you can do
>>your BVRI images on the same pointing.  And yes, you can get better
>>statistics by calculating your coefficients over several nights, and
>>averaging them together.  This is because transformation coefficients don't
>>change all that much over time, unless you change something significant in
>>the optical path or in your CCD. Data for these fields is available from
>>Arne Henden's FTP site at
>>ftp://ftp.nofs.navy.mil/pub/outgoing/aah/sequence/.
>>
>>They can be crowded for amatuer focal lengths, so pick your stars carefully.
>>And note that some are variable, so look for ones on the edge of the fields,
>>with the most observations and least associated error.
>>
>>By the way, if you have an I filter, you might get better color data with
>>V-I rather than V-R.
>>
>>Chuck Pullen (PCH)
>>
>>----- Original Message -----
>>From: "Geir Klingenberg" <geir.klingenberg at gmail.com>
>>To: <aavso-photometry at mira.aavso.org>
>>Sent: Tuesday, December 28, 2004 2:23 AM
>>Subject: [Aavso-photometry] Differential transformation
>>
>>
>>
>>>Hi
>>>
>>>The methods I have seen for estimating the transformation coefficients
>>>involves using instrumental magnitudes in a LS solution. But - if one
>>>is only doing differential photometry and don't need to be concerned
>>>with extinction - is there anything wrong in using a differential
>>>approach? More specifically, if we ignoring extinction the relation
>>>between the stars magnitude, instrumental magnitude and color can be
>>>written in the usual form as
>>>
>>>V = v + a*(v - r) + b
>>>V - R = c*(v - r) + d
>>>
>>>This is for the V and R band - capital letters are magnitudes and
>>>small letters are instrumental magnitudes. So, subtracting magnitudes
>>>of star 1 and 2 and rearranging we get
>>>
>>>(V1 - V2) - (v1 - v2) =  a*( (V1 - R1) - (V2 - R2) )
>>>(V1 - R1) - (V2 - R2) = c*( (v1 - v2) - (r1 - r2) )
>>>
>>>which can be used to estimate a and c.
>>>
>>>Since all instrumental magnitudes in these equations are on
>>>differential form  between stars on the *same* image, won't the
>>>solution be more robust against sneaking cirrus clouds and changing
>>>atmospheric conditions? And can measurements from different nights be
>>>used to improve the coefficient estimates?
>>>
>>>Geir Klingenberg
>>>_______________________________________________
>>>Aavso-photometry mailing list
>>>Aavso-photometry at mira.aavso.org
>>>http://www.aavso.org/mailman/listinfo/aavso-photometry
>>>
>>>
>>
>>
>>_______________________________________________
>>Aavso-photometry mailing list
>>Aavso-photometry at mira.aavso.org
>>http://www.aavso.org/mailman/listinfo/aavso-photometry
>>
> 
>