I was messing around with some radial velocity data from NASA's Exoplanet Archive and the Lomb-Scargle periodogram for Pegasi 51 yielded two very sharp peaks. One corresponds to about 4.23 days which is the reported value for Pegasi 51b while the other is about 1.3 days. Existing archives and paper only account for one orbiting planet. Would anyone happen to know why or what caused this extra signal to be picked up? Thank you!
$\begingroup$
$\endgroup$
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
2
Ground-based radial velocity surveys can be affected by strong aliasing because the cadence of the observations is 1 per day. This can produce alias frequencies in the periodogram at $1$ day$^{-1}$ $\pm$ the true frequency. You can think of these as beat frequencies between the true frequency and the observation frequency.
If you have a true frequency of $(1/4.23) = 0.236$, then you could get aliases at frequencies of $\sim 1.236$ day$^{-1}$ or $0.764$ day$^{-1}$. The latter corresponds to a period of 1.31 days and is I suspect what you have found.
-
$\begingroup$ Thank you, I came across a paper that offers a similar explanation: arxiv.org/abs/1502.01344. However I'm still not entirely satisfied with this answer since the raw data is far from evenly sampled: exoplanetarchive.ipac.caltech.edu/data/ExoData/0113/0113357/… Please let me know what your thoughts are. Thank you once again. $\endgroup$ Dec 13, 2022 at 3:36
-
$\begingroup$ @chuffyduffy why not generate some fake data at the same observation times, but with a different real period and see if the second peak shifts accordingly? $\endgroup$– ProfRobDec 13, 2022 at 6:55