I just saw a news clip today about the TESS data on 12 Boo. I can't say I understood all of it, but that;'s why I'm learning. I was going to see if AAVSO has data on this star but see that Sebastian Otero has already plotted the TESS data. Has anyone overlaid (overlayed, overlain?) AAVSO data with the TESS data? Might be a good learning exercise for me to try? Then again, everything I dp is a learning exercise!
This also may be relevant to the discussion of the future relevance of AAVSO I've seen discussed elsewhere.
Thanks for your tolerance,
Thanks for pointing my attention to the fact that someone else had published something on 12 Boo (d Boo).
I had independently discovered it to be an ellipsoidal system last week and published it in August 11 but it turned out that a team of researchers had analyzed the same data and published their results 8 days before me!!
I had no idea the system had been in the news.
So I have now changed the discovery credit to properly reflect their finding.
About combining TESS and AAVSO data on it, you have to keep in mind that TESS detects very small amplitude variability. Here, we are talking about 0.002 mag. variations. So it makes no sense trying to get photometry from the ground because we could never be even close to what TESS observes.
An eclipsing system with 0.002mag (2mmag) amplitude and 9.6d period is basically impossible to do from the ground. On occasion, I've seen light curves of transiting exoplanet systems with 2mmag precision, but that means you can barely detect the change, not show any of the structure of the light curve. Such a transiting exoplanet light curve also takes place over a few hours on one night, much easier than trying to piece together segments of a light curve over more than a week. You should leave these kinds of systems to the space missions and concentrate on the many larger-amplitude, equally important, systems!
To do much better than 50mmag (0.05mag) accuracy (where multiple observers are contributing to the light curve) on an arbitrary target requires transformation and care. Ulisse Munari's amateur team studying novae can do ~10mmag light curves combined. Joe Patterson's CBA group can have this level of performance on cataclysmic variables, but with zeropoint adjustments and overlap between observers.
I think these are basic limits with the current state of amateur photometry. As Tom Calderwood has shown, the AAVSO does not reach these limits, and much work remains in learning how to do high-precision photometry in the amateur world. I think Dr. Munari's success should be a goal for our organization.
What Arne suggests is correct. Back in the day, using the very stable single-channel photometer on the Lowell 21-inch telescope, I got rotational lightcurves for 10 and 11 Leonis Minoris (a G8 giant and a G8 dwarf, resp.) as part of a large project. Because the two comparison stars were especially stable, the nightly rms scatter was only ~3 millimag across a whole observing season. This enabled us to see both the rotational variability and the tiny change in color-index over two seasons. See:
...specifically Table 1 (bottom few lines) and Figures 1 and 2. The residuals on the phased curve in the Stromgren b-y color make this only a 1.5-sigma detection. If the data hadn't phased beautifully with the y lightcurve, I doubt we would have believed it.
Using CCDs on our 31-inch and 42-inch telescopes on similar sorts of stars, about the best I've been able to do is somewhere in the range of 5 millimag (rms per data-point over weeks/months timescales), roughly 2x the single-channel threshold. This even though the nominal S/N can be like 1000. You'd think, given that the target and comps _and_ the sky background are all measured at the same time (versus all done sequentially with a photometer), that things ought to be at least as good. The usual explanations are the limitations of fiat-fielding, stability (or lack) of the camera electronics, scattered light in the system etc.
If you are "learning", and I have always believed that we should always be learning every day of our lives, then you surely can repeat the analysis yourself and overlay any other available data you want. Much of science today is automated. It doesn't hurt to have a human in the loop checking things. It is also good to keep your software and brain-supplied analysis tools sharpened with exercise.
Jim DeYoung (DEY)