Over the past several years, there have been a number of forum threads on the topics of Best Equipment, Best Practices, and Best Algorithms for imaging photometry. To my knowledge, however, there have been no (reported) attempts to try these Bests against standard stars. I would like to challenge our observers to try their equipment, practices, and algorithms on standard targets, in particular, some of Arlo Landolt's northern standards*. There are several pairs of not-too-dim stars that will fit in a typical camera's FOV. One can be used as a comparison, the other as the "program" star. I think it would be very interesting to see how well people can match Arlo's results. There are two pairs well-placed for early evening work right now (coordinates J2000).
Pair One------------------------------
Comparison 09:42:58.48 +43:50:04.9; B=10.139, V=9.590
Program: 09:42:49.28 +43:49:35.8; B and V roughly 1mag dimmer than comp
Separation= 1.7 arcmin; delta(B-V)<0.250
Pair Two-------------------------------
Comparison: 09:47:04.54 +44:22:31.2; B=12.936, V=12.133
Program: 09:46:53.61 +44:25:05.7; B and V very roughly 0.7mag brighter than comp
Separation= 3.2 arcmin; 0.250 < delta(B-V) < 0.400
I have deliberately left the exact magnitudes of the program stars unspecified. Of course, you could cheat and look them up, but I encourge you not to do so. You can post your BV measurements (including uncertainties) on this thread, and after a while I will post Arlo's magnitudes. If there is interest, I have more pairs extending to RA=0h.
Tom
* I am less familiar with standards in the southern hemisphere - perhaps southern observers could suggest some?
Hi Tom,
The Cousins E-Region standard stars are suitable for a similar excercise for southern hemisphere observers. There are 9 regions spaced evenly around the sky at -45 degrees declination, each region containing approximately 80 to 100 well measured stars mostly between V mag 6 and 10. The following reference has the details:
MENZIES J.W., COUSINS A.W.J., BANFIELD R.M., and LAING J.D.
South African Astron. Obs. Circ., 13, 1-13 (1989)
UBV(RI)c standard stars in the E- and F-regions and in the Magellanic Clouds - a revised catalogue.
Cheers,
Mark
For Pair One, use X24414P. Comp is between 9 and 10 o'clock, 1.7' from program star.
For Pair Two, use X24414O. Comp is at 7 o'clock, 3.2' from program star.
Both charts to 14th magnitude, CCD orientation, 18.5' FOV.
Tom
Hello Tom
To use Maxim, we will need a check star for each of these fields--I suggest that everyone use the same. To submit to WebvOBs we will also need a Program/Target designation that it recognizes. That way, results can be shared with the community and plotted on the LCG.
Gary
About the best check star I can come up with for Pair Two is TYC 2999-181-1 at 09:47:19.94 +44:24:28.5. It is two full magnitudes brighter than the comp star in B&V.
According to SIMBAD, there is another star practically on top of the Pair One program star - I need to look at this further.
In order order to make AID/WebObs accept data for the Landolt standards, we would need to add those stars to VSX. I don't know how the VSX team would feel about that. Personally, I think it would be a great idea for the AID to support standard stars so people can check their photometry. But perhaps we should first experiment with these pairs to see how well they work. If folks submit their results here, I can make a graphic of some sort.
Tom
The close companion appears in the TYCHO catalog as TYC 2998-959-02. It does not appear in Gaia DR2, which shows no stars brighter than G=19.8 within 1 arcmin. The TYCHO entry looks spurious, but perhaps someone can take an image.
Check stars are, again, hard to come by. Try TYC 2998-1018-1 at 09:42:22.79 +43:54:55.5 B=12.21 V=12.15
I think this is a great idea. Arne did a similar test some years ago but, as I recall, with a variable star (XZ Cet?). I was very disappointed not to be able to participate as I happened to be sans observatory for a couple of years after moving house.
Maxim (at least 5.07) doesn't require a check star. In any case, TYC 2999-181-1 is too bright for me - I would be down to a few seconds exposure to avoid saturation. Plus it would result in increased noise on the target/comp stars. If we feel we need a check star then I would probably choose GSC 2999-403 at 09:47:06.82 +44:224:34.6 (though I can't seem to find any information about the star so I don't know what its colour may be.)
If it actually clears off this evening as forecast I will shoot some B & V images.
Rick
APASS provides this:
V: 13.755 Verr: 0.111 B-V: 0.956 B-Verr: 0.113
Also, if you want an exellent portal into astronomical data and images in any wavelength, then Aladin is your best friend. It's a JAVA app available here: https://aladin.u-strasbg.fr/aladin.gml
And have a look at: http://cds.u-strasbg.fr/
Landolt standards are available every night. I encourage my CCD1 students to use them (with due regard to uncertainties) to access how they are doing once they determine their transforms (a skill taught in the CCD2 course). Use some standards in the check and ensemble and others to measure. Finding one's measures are within uncertainty brackets of standard star measures builds confidence. I would think that any combination of equipment and processing that consistently results in measures within uncertainties would be just peachy.
Ed
I finally was able to image pair two tonight. I used VPhot to get the photometry.
For pair two, I get for the program star:
Vmag 11.335 err: 0.006
Bmag 12.433 err: 0.019
Airmass was 1.27.
Based on 4 x 100s B and 75s V images of pair two I get B=12.419+/-0.008, V=11.321+/-0.004 from Maxim photometry, transformed.
Airmass 1.04.
After seeing what Richard did, I am resubmitting my photometry. I had intially obtained photometry on each of 4 images in each of B and V. I then averaged the results. This time I stacked the 4 images in each of B and V (using VPhot average stacking method) and obtained slightly different results, and with lower err and better SNR (naturally). So here we go:
V: 11.336 +/- 0.005
B: 12.446 +/- 0.012
BTW, these like the previous submission, are NOT transformed.
B=12.421, V=11.304
program V uncertainty= 0.0005, B-V uncertainty= 0.0009
comp V uncertainty= 0.0007, B-V uncertainty= 0.0009
Richard is spot-on in B, not quite as good in V.
Michael: try pair one, where the color contrast is very small (and get your transforms!).
There is a nice selection of new targets beginning at approximately 13 hours. I will post soon.
Tom
I'll try the new targets. I'm wondering if the airmass at the time of observation was a significant factor. Richard just a wee bit over 1.02 and I was getting close to 1.3. I'll do that experiment to see the difference in results over that airmass range. And oh yes, the transforms. I need to kick myself into that!
Richard:
For the benefit of this PEP observer, can you provide more details about what was done with Maxim? Are the four exposures (per band) being combined and evaluated as one frame, so that the reported uncertatainty is Poisson, or are they being processed separately and averaged, so that the uncertainty is Guassian? Is second-order extinction being applied to B band?
Tom
A wide field might get pairs 3 and 4 (which share a comp) in one frame.
Having entered all this as tabular data for the posting, it now appears that tables do not work :(
Hi Tom,
I was able to fix your table with my Admin power. Is that OK now?
Thanks,
Sara
It is very strange that the table displays fine while it is being composed, but then falls apart upon posting. Looks good now.
The latter: 4 B images, 4 V images, Maxim photometry done on all 8, transformation applied, results averaged and std deviation reported as the error estimate. Which makes me suddenly realize that I should have divided those estimates by sqrt(3) to get the error on the mean. So the corrected values should be B=12.4189+/-0.0045, V=11.3212+/-0.0024, random errors only.
No secondary extinction performed.
Richard:
Assuming that your second order extinction coefficient is about -0.035, your actual B magnitude is about 11mmag dimmer than the reduction indicates. Is Maxim able to apply a second-order correction if the coefficient is known?
Maxim doesn't deal with extinction at all. Or transforms. I use Maxim just to do the basic photometry - identify the target, comp, and check stars in the first image and input the magnitude of the comp star. Maxim then measures all the images and outputs a CSV file. I do this twice - once for all the B images and once for all the V images. Then I put all the data into spreadsheets where I calculate b-v, B-b.... to get zero points, apply transforms, do any required averaging etc. I don't have extinction coefficients yet - just getting good transform coefficients has been a problem that I haven't yet worked out to my satisfaction.
Tom et al
I have been promoting this challenge at the joint RASC/AAVSO conference held in Toronto last week. We should see some new observers come on board to submit their estimates. Maybe we can get some new sets of pairs to try starting around 15-18 hours RA range. The later twilight times of Summer are now problematic.
13H-19H in RA