Following Dr Riess's suggestion in Saturday's webinar that a valuable contribution amateurs can make to the improvement of the Hubble 'constant' and the expansion rate/change of the universe would be to observe long period Cepheids I did a search in the fabulous VSX for DCEP with periods > 5d (and located north of -20d as I live in southern Canada, and south of +55 for the trees along my northern 'horizon') and I find that most of the stars are quite bright. Only 27 of 158 stars have a maximum fainter than 12 and none are >13. And a significant number of them don't even have AUIDs. Seems hard to believe that supergiant variables should show so few fainter members.
Does anybody have any thoughts? I suppose one could search the GAIA DR3 for (many many?) more potential targets.
Compliments and thanks to the staff who nabbed Dr Riess for a talk - it was really interesting!
That is interesting, and I must admit that I have mostly worked on type II variables, rather than classical Cepheids. I don't immediately have anything to add regarding the best sources for targets, but will see what I come up with after the holiday in the US. ASAS-SN will have good coverage on some Cepheids, but one has to be careful of blending because of large pixels and the bandpasses will be limited. Maybe someone else will chime in with good ideas. -- Horace
On the SPP section page, go to the link “AAVSO SPP Program”. On that page, there is a table of the stars on the AAVSO legacy observing program. The periods of these stars range from 1 day to 68 days. Because these stars were well observed by the AAVSO a half century ago, we have the opportunity to check for changes in their periods.
I'm curious about what is proposed, and confess I did not see the Riess seminar.
My understanding of the contribution of studies of galactic Cepheids to the cosmic distance ladder is that they callibrate the period/luminosity relationship for these stars. This is possible because the distance to galactic Cepheids can be determined directly and thus absolute magnitudes can be calculated. This has been done.
But to go further and collect data to determine the Hubble constant requires studies of extragalactic sources. Such studies in optical wavelengths have used professional ground-based and space telescopes.
I know that amateur spectroscopists can determine redshifts. But where do amateur variable star astronomers studying Cepheids sit in the endeavour to refine the Hubble constant?
I gather there is still need for ongoing and improved observations of relatively nearby Cepheids by the professionals to improve accuracy/statistics etc. So the amateur can help everybody keep up-to-date on the phase of pulsation of these stars (in view of inaccuracies in periods over long periods of time or changing periods) so when the professionals do their observations they have an accurate phase for their scheduling or their analyses.
Perhaps there is something else as well. After your last post, I looked for literature on the issue.
For a few years professional astronomers have been looking at what effects if any metallicity has on the Cepheid period/luminosity relationship.
It turns out that there does appear to be an effect, and there is interest in quantifying it as precisely as possible.
Another sources of classical Cepheids is the list at the David Dunlap Observatory of the University of Toronto.
There are some into the 14th magnitude range in V on this list. I'm not sure that the periods can be refined much more than has been done on these objects. The distances from GAIA are going to have the biggest impact of refining the P-L relation. Although reddening is going to have an impact as well.
The periods of Cepheids change over time, so that even a good ephemeris for the time of maximum light may gradually become not-so-good. Thus, new observations may be necessary to tweak predictions of phase for, say, spectroscopic observations. Period changes are also useful for testing stellar evolution theory. ASAS-SN data can help with period updates, but are not always ideal for particular variables. That is one reason for the Legacy program SAM mentioned.
Getting your own photometry of cepheids, rather than only depending on professional provided survey data, requires a dedicated program of observing to get enough coverage of the light curve to allow good period determination. The legacy program objects are a good place for new observers to start. Period determinations are probably the most amateur astronomers are likely to contribute to the "cepheid problem". As a cepheid ages it moves through the instability strip in the H-R diagram and they are expected to have period changes. Light curve period changes most likely would help with the understanding of the pulsation physics.as the star evolves rather than anything related to the P-L relations. Btw, my first cepheids were observed visually for the AAVSO cepheid program (Tom Cragg) over 40 years ago.
Jim DeYoung (DEY)
DEY, It sounds as though you've been in it for the long haul on Cepheid variables! After all, while 40 years is a small time in the life of a Cepheid, it's a long time in the observational history of most of them. Even Delta Cephei itself has only been known to be variable for less than two and a half centuries. I suppose that the brighter Cepheids began to be picked up in the Harvard patrol photographs 120 or so years ago, so that 40 years would be about a third of the observational history even for them. -- Horace
This is something that interests me as well, not so much (in fact, not at all) from the cosmology angle, but the Cepheids-qua-variables. One of the problems in planning an effective observing scheme is that there is no really convenient place to look to see when the stars were last observed, in what filters, and whether more (good) data are required. Part of this is that much of the existing historical data and even recent data are not readily available to allow someone to make these assessments. And what about all the wide-field survey data? Does one need to do any of this observing at all? TESS data, covering only short intervals and unfiltered, is of almost no interest. On the other hand, ASAS-SN runs in V or Sloan g and seems to be fairly good in certain magnitude ranges. This sort of thing needs to be assessed star-by-star. Recently a Hungarian researcher has published a pretty comprehensive collection of archive photometry:
...that might help with this. I have collected/keyed-in large amounts of historical data that needs processing (bookkeeping work), some of which Geza Csornyei said they had not identified for the above dataset. Essentially none of it is in the AID. Then there are things like Scott Engle's PhD thesis, which contains a huge amount of robotic telescope data, only quasi-published and _not_ in machine-readable form:
Another issue is just what one should observe. It seems to me from what I've read that having well-calibrated data (and how 'bout them comp stars?) in something like B,V, and I (or even U, V, and Sloan z to maximize the wavelength baseline) is what's desired for the various purposes to which the data might be put. I would say the R filter is not useful since one doesn't learn anything about the stars you don't get from V and I, and one wants to maximize the number of stars observed/data acquired, thus you want to get only the minimum number of filters.
Actual expert opinion on these subjects would be welcome.
a list of stars to be observed in BVI filters would be very mch appreciated.
As I described in my previous note, indeed "a list of stars is needed"! A few years ago I wrote to Leonid Berdnikov, who has done the majority of published UBVRI Cepheid photometry in the last 30 years of so. He responded that the number of simply unobserved Cepheids was small, but was not able to supply a list, nor say which well-observed stars needed continuing observation. He has veered off into radial-velocities for the Cepheids, so seems to have stopped the photometry. (He must also be pushing into retirement as well.)
A long star-list plus the source for _some_ Cepheid photometry (and rv's) is the database built by Doug Welch:
However, not much has been done on this in the last 20+ years(!), and it contains only a fraction of even the published data prior to that time. A few years ago I did point Doug to Berdnikov's summary files in Moscow, which he has parsed out into his set-up (Berdnikov's files are arranged peculiarly, probably from having started assembling them in the early 1990s, when his access to computer hard/software was surely limited. Ed Schmidt did a lot of observing starting in the 1980s (through the early-aughts), and on-line files have been copied (I have more unpublished stuff from Ed directly). I think most/all of Arne Henden's stuff is in there, also. But I have many thousands of lines of data keyed-in that are missing from the database. An egregious example is Polaris, where Welch has only Arne's 22 UBV observations, and all of _three_ rv's (count 'em!). My file has 3900 lines of photometry + rv's collected from the literature (some of it impossible to use, however). All the data ought to be in the AID, the CDS-VizieR, and/or other places so it can be both discoverable and accessible. Doug Welch is a senior academic (ahem), so what happens to the database when he's gone fishin' for good?
But besides the simple collection of the data, the real problem, for which expert help would be needed, is to assess all this star-by-star, and decide what is worth observing, in what filters, and whether previous and on-going wide-field surveys (ASAS-3, ASAS-SN, Evryscope, ZTF etc) mean there is no need for new data (only analysis of them). Changing periods, binary companions, and other effects need to be assessed. I also found in a year of exploration of trying to do this with our robotic telescope that the situation in re comp stars is often poor. To be worthwhile, any new data _must_ be adjusted to the standard system within a couple percent. Otherwise there is no way to compare with historical observations to look at long-term trends without making ad hoc arbitrary zero-point/color-term adjustments (including re-measuring the comp stars used by previous observers). Berdnikov has emphasized this in his papers. So part of the observing involves doing the calibration of field stars on multiple nights with multiple filters against high-weight standards. I have a start on some of that. (APASS is often inadequate.)
One masterful assessment of (northern) Cepheids is Laszlo Szbados's series of Konkoly Observatory monographs. Start here:
...and look at numbers 76, 77, 94, and 96. Fans of RR Lyrae will find a treasure-trove here, too. This carries the literature summary and analysis up to the mid-1980s, but now what? Since then our understanding of stellar physics has vastly improved, much photometry and spectroscopy has been done, methods of analysis transformed, and so on. What does that mean for observers? I simply do not know the answer to that. "Just take more data" is not a good answer, in my opinion, since it may be being done by others better. How does one make start on finding out?
I finally looked at the video of Dr. Reiss's talk... yesterday. What he seemed to say is that AAVSO observers could contribute by observing cepheids with periods greater than 5 days so that professionals would know the (pulsation) phase of the variables when they observe with the big/space scopes. Many of the observing requests produced by the AAVSO alert/observing request system are already of this type--to know the state of a variable or its pulsation phase for correlation with other observations by the big or space telescopes. I have only been observing a few CEP with periods less than 10d during the last few years. What I would recommend is that new observers pick cepheids in the 1 to 9.99d range and that do not have any recent photometry and give them good long-term coverage over at least one full observing season.
If observations aren't readily made available they are as good as not having been made. The old AAVSO Cepheid Program (Tom Cragg) concentrated on cepheids with periods of 10d or longer. Since I retired I have been following about 55 long period cepheids most with periods longer than 10d. The data are a mixed bag of DSLR and CCD but all are transformed and most now have data from 2 years or more.. The data are good enough to get times of maximum and be smoothed to get mean curves. Most have BVRI but some only BVR. At least one cepheid may be an RVa but the light curve only won't help with classification so I continue to observe it... I'm biased that it really is a cepheid. The 55 range from very bright & easy to very faint with very long exposures using my small 100 and 127mm refractors. Some of them are very reddened so B magnitudes especially can be low SNR measurements.
My data hasn't been uploaded yet and won't be until the finished paper is in hand... it should provide a good starting point for others to jump off from however.
Jim DeYoung (DEY)