Skip to main content

Maximum timing of long perid Cepheid

5 posts / 0 new
Last post
Maximum timing of long perid Cepheid

Hi, all

Some Cepheids have somewhat long period of varition, from several days to 100 days

For very long periods, for example 100 days, I might use same methods as Mira variables to determine maximum timing.

What methods are better to use for Cepheids that has periods from several days to weeks?

Usually I can get only sparse photometric data (1/day or less). How can I determine maximum timing from such sperse observations?


                                 Seiichiro Kiyota, Japan



Cepheid maxima

Hi Seiichiro,

Because a Cepheid light curve is stable, I see two options: one is to group observations from a number of cycles and calculate a time of maximum from the phase plot.  A second method is to make a light curve model from a large number of observations and use that model to fit the data of a single cycle.


wel's picture
Estimating epoch of maximum brightness

Hi Seiichiro,

Patrick is correct in his recommendations. A couple of addition things to consider:

- If one is interested in maximum precision for period changes, the time of maximum is far from ideal. While the signal-to-noise might be highest there for measurements, the signal is not changing rapidly with time and so the uncertainty in the time of maximum will be greater than, say, the uncertainty in the time of the mid-point of the ascending branch of the lightcurve for the same set of observations.

- Grant Foster did a video presentation about estimating the uncertainty in the time of maximum, which is an important quantity that essentially everyone has ignored! Perhaps someone can provide a link to that presentation.

- Some Type 2 Cepheids have very flat maxima and therefore the time of maximum brightness is very poorly defined.

- Cepheids with periods greater than 70 days are very, very rare.



Grant Foster's talk

Grant Foster's talk on the uncertainty in finding maxima can be found on this page  It is mostly geared towards Mira stars which may have a changing light curve from cycle to cycle.

Doug is right that the mid-point of the ascending branch can often be determined more accurately.  When fitting a model curve however it doesn't really matter which phase (maximum, minimum, ascending branch, ...) is chosen as the time to report, as in fact the whole cycle is fitted (only for fixed light curve shapes, of course).  Also the points close to the most rapid change will have the largest weight in determining the epoch.


Thank you, Patrick and

Thank you, Patrick and Doug

I will try phase plot.

I have already watched video of Grant Foster's talk and understood (??) how difficult to determine maximum timing from lightcurve.

Following link are example of phase plot lightcurves of DCEPs.

They are by-products in my nova patrol with camera lens+CCD (Ic filter).


Seiichiro Kiyota

Log in to post comments
AAVSO 49 Bay State Rd. Cambridge, MA 02138 617-354-0484