Sat, 03/27/2021 - 08:05
I'm using a DSLR for my photometry work. The CCD photometry guide says ideally I should use 30-50 stars reference stars to calculate the coefficients but I have trouble getting anywhere near those figures. Using M67 I found 8 stars because the central area was too crowded.
So I wondered if there was a way of combining data collected over a number of nights using different standard star fields to get more accurate coefficients?
Cheers
Steve
Steve:
Yes, M67 is not an ideal standard field for DSLR lenses. It is too dense in your very wide field of view (FOV).
Take a look at the Standard Fields page: https://app.aavso.org/vsd/stdfields. It will allow you to search for other standard fields, which may offer many stars in a wider FOV.
The best answer to the question in your second paragraph is for you to experiment with obtaining transformation coefficient results from multiple fields over multiple nights. Prove to yourself whether you can obtain consistent results. Since the coeffs represent the characteristics of your filters (in this case the filters in your Bayer array), you would expect/hope that the repeated results would be the same unless you change them in between runs, which you obviously do not. If coeffs calculated from different fields or nights give significantly different results, I suspect you would not have much faith in the process of transformation!
It is likely that you will find through trial that the number and especially color range of the standard stars that you include in your list of stars will yield somewhat different coeffs. The question you will wrestle with is how significantly do the coeffs vary for your individual analyses. The basic discovery/observation from this experiment is that you want many stars and as wide a color range for these stars to get the "best" (precise and accurate) mathematical fit for the slope of each linear plot (transformation coefficient). I hope that you will determine that an average coeff from multiple fields and multiple nights will yield a reliably consistent result? However, keep in mind that if you only repeatedly use one standard field with very few usable stars, the slopes/coeffs may be consistent from night to night BUT very inaccurate because one or two of the few points has some systematic error (e.g., aperture overlap, inaccurate reported color) that you are not aware of? Note that it does take effort/practice to get reliable results.
Ken
Steve,
There is another way you could consider for finding stars for transformation coefficients.
The planetarium software Guide displays V and B-V values to 3 decimal places (and errors as well) for stars in the Hipparcos and Tycho databases. The values are, I believe, calculated from the original VT and BT magnitudes using a cubic spline function. B-V values are not calculated for very red stars.
I assessed how good the V and B-V values are in Guide by comparing them with the corresponding values for E Regions photometric standards (Menzies, J.W. et. al. 1989, S. Afr. Astron. Obs. Circ., 13, 1). These are southern standards, not visible from your location. For 10 stars selected from the catalogue to give a range of B-V values from 0.085 to 1.495, the difference between Guide and catalogue V mags was <0.01 (absolute value) for 7 stars, and 0.011, 0.016 and 0.061 for the other 3 stars. The differences in B-V values was <0.01 for 9 stars, and 0.04 for the other star.
In my opinion, this level of accuracy is good enough to use these stars not only for transformation coefficients, but also as comp stars, which I do for variables that have no comp stars in AAVSO charts.
The advantages of using values from Guide is that you could select relatively bright stars (since you are doing DSLR photometry), virtually at any time or date you wished, choosing stars at high altitude to avoid atmospheric extinction problems.
Roy
Thank you Roy, much appreciated. I'll look into it.
Steve
Steve,
Here's another option.
Load the M67 chart in VSP. (See Ken's earlier post for the link.) Then click on Plot Another Chart. In the new plot form change the size of the FOV to 90 arc minutes. You will see lots of relatively bright Henden standards well away from the crowded central part of M67. You could experiment with somewhat smaller or larger fields. I'm guessing the upper limit is 120 arc minutes. Even before this is reached the FOV may get too large and the plotter may time-out. In this case reset the minimum magnitude to be brighter.
Caveat: There will be lots of duplicate and triplicate star "labels" which may not apply to the same stars the next time you open the M67 chart in VSP, so indentify the stars you want to use by the AUID's shown in the table, not the labels.
Regarding the 30-50 stars recommendation which you mentioned, I would disagree. The important consideration is not the number of stars used, rather it is the range in color (B-V or V-I) of the stars used. There is nothing wrong with including lots of stars, and some people just "use 'em all". If you do that you're guaranteed to get the reddest and bluest stars in the cluster. This works for longer FL scopes which can clearly resolve the center of the cluster.
In your case, if you download the enlarged photometry table for the wide field M67 standards (select all, save, then paste into a spreadsheet) and sort by B-V you can pick some of the best stars from both ends, then add a just few more stars in the mid range. I think ~15, more or less, carefully selected standard stars should give you good results.
For years, new photometrists have been learning to calculate their transforms (with good results) using the M67 13 star sequence in Chapter 6 of the CCD Photometry Guide. That sequence may be too crowded for your wide field system, but you can do well with the wide range of star colors available in the M67 60 or 90 arc minute field.
Phil