When using images in two filters to report transformed magnitudes is it necessary that the images use identical comparison star sequences?
The quick answer is yes. Try it in VPhot? When you select two images (two different filters) and click Transform, your are asked to select a sequence. This can only be a single sequence so by definition the same comps are used to conduct the transform. (Assuming they are bright enough?)
Why do you want to use two different sequences? Did you not run your B image long enough compared to V?
Thinking about the equations, you're comparing instrumental b;v to known B;V and applying to the target so I would think you would normally want the same comps but I suspect I could think about the math some more and perhaps be convinced otherwise.
However, I have never thought about doing it with different comps, so I never tried. But again VPhot cannot do it that way in any case.
If you use TA for your transform there is no restriction that the same comp be used for different filters. The math does not require it. VPhot's constraint is a simplification to make the specification of the comps easier.
... very interesting. I haven't tried TA yet. Is there a simple way to do the photometry in VPhot and apply the transforms with TA? (I'm using comp star ensembles.)
First thing is that TA will not transform ensemble observations. This has been discussed in http://www.aavso.org/transforming-ensembles . The issue is that there is no information included in the AAVSO extended format record about the comps other than their average. As Arne said: "Ensemble photometry is very useful, as it not only averages out possible problems in your image, but also gives you a lot of diagnostic information. However, if I had to choose between transforming my data with a single comparison star, or not transforming but using an ensemble, I'd definitely choose to transform."
On the VPhot side your process will be pretty much unchanged. Set a sequence by star/filter and go through each image as usual: select image, apply sequence, view photometry, keep this. When done with images go to the analysis log, select the images and create the AAVSO report. This will be a reported in the format ready for WebObs and will be the input into TA.
TA is available at http://www.aavso.org/transformapplier . Download and sort out importing your transform coefficients. Only need to do this once.
Doing the transform is then is a matter of loading your report from VPhot and hitting the process button. TA will produce a transformed report ready for WebObs. It also produces a lot of detail information about the process that is worthwhile working through the first few times so you understand what is going on.
You're welcome to contact me directly with questions at SGEO@GASilvis.net.
Thanks, George, especially for that link to the the discussion on the Photometry forum staring with Arne's post on Mar.15, 2015. There's a lot more going on here than I had thought. Also, thanks for the offer of help with TA off line. I may have some more questions for you off line once I've had a chance to ponder all this.
You asked why I'd want to use a difference sequence for each filter.
I want to report transformed magnitudes for faint (below 17V) red stars. To get a half decent SNR's with exposures in B filters will take forever, so I'm going try using V minus Ic transforms to report these magnitudes and forget about B (at least when the stars are this faint).
Unfortunately the best AAVSO comp stars for V are sometimes not the best comp stars for I. I've had the situation where a good comp (good SNR, low uncertainty, near the target) in Ic has a 3 sigma deviation in V. Also, there are fewer AAVSO comps stars with Ic standard magnitudes, so if the same sequence is required for both filters the total number of useable comps stars is decreased.