Project #2: RR Lyrae binaries
Doug Welch
Data-mining Project #2
----------------------
Goal: Searching for Galactic RR Lyrae Binaries
Number currently known in the Milky Way: 0
(Number currently known in the Universe: 0)
Why: It is extremely puzzling that no RR Lyrae
have been found to be in *any type* of binary. Their
red giant brethren have about a 25% binary frequency.
Does binarity prevent from stars evolving to become
RR Lyrae?
Why Look in the Milky Way: Because the RR Lyrae
in the field and in the bulge of the MW are MUCH
brighter and less crowded than those in the
nearest galaxies, the Large Magellanic Cloud and
Small Magellanic Cloud. Follow-up photometry and
radial velocities can therefore be more precise and
physical characteristics such as mass function, mass,
etc better-determined.
Why hasn't it been done yet: The half- to whole-day
pulsation periods are a difficulty for establishing
binarity through radial velocities - partly due to the
large velocity amplitude of the photosphere and partly
due to how quickly it changes. A single full night's
worth of velocities gives a single point for looking
for orbital motion. The orbital periods would be weeks to
months for the systems most likely to be found.
Any detached eclipsing binary would have eclipses
lasting a very short fraction of the orbital period.
Furthermore, if there are eclipsing binaries among
the RR Lyrae, their occasional effects would be
subtle on the lightcurve and easily overlooked or
dismissed without searching specifically for them.
The Blazhko effect also serves to make the search
more ... interesting.
How would one go about it: Get the 3525 MACHO lightcurves
for Milky Way Bulge RR0 stars identified by Kunder et al (2008)
from me (or download them yourself). Extract the VOTable
lightcurves to your modelling program. Model and remove
the fundamental mode (and, if present, Blazhko) variation
and look for evidence of eclipses at regular intervals
in the residuals.
Another way to establish binarity is with the orbital
phase variation produced in the lightcurve. This is possible,
but tougher due to a number of factors such as:
-the small light-time orbital period modulation (5-10 minutes
compared to a pulsation period of about 800 minutes),
-the uncorrelated "period noise" that RR Lyrae have (for
unknown reasons), and
-the distribution of orbital plane orientations in the sky
would result in only the most favorable orientations producing
detectable phase modulation.
Chance of success: Very high since there are thousands
of RR0 in this set. Even a null result would be significant
since it would establish that the dearth of RR Lyrae
binaries is not just an observational selection bias.
Possibly related fact: The population of Type II Cepheids
in the Large Magellanic Cloud has been found to have an
unexpectedly large eclipsing binary fraction. Type II
Cepheids are a population of former horizontal branch
stars closely related to RR Lyrae.
References:
"The Optical Gravitational Lensing Experiment. Catalog of
RR Lyr Stars in the Large Magellanic Cloud"
Soszynski, I., et al. Acta Astronomica, v.53, pp.93-116, (2003)
http://arxiv.org/abs/astro-ph/0306041
Three RR Lyrae showing both pulsation and eclipses in their
lightcurves. (All were blends)
"In Search of RR Lyrae type stars in eclipsing binary systems.
OGLE052218.07-692827.4: an optical blend"
Prsa, A. et al.
Astronomy and Astrophysics, Volume 489, Issue 3, 2008, pp.1209-121
http://arxiv.org/abs/0808.3560
"Low-amplitude Variables: Distinguishing RR Lyrae Stars
from Eclipsing Binaries"
Kinman, T.D. and Brown, W.R.
The Astronomical Journal, Volume 139, Issue 5, pp. 2014-2025 (2010).
http://arxiv.org/abs/1003.3656
"The Extinction Toward the Galactic Bulge from RR Lyrae Stars"
Kunder, A. et al.
The Astronomical Journal, Volume 135, Issue 2, pp. 631-636 (2008).
http://arxiv.org/abs/0712.0389
Contains MACHO identifications for 3525 fundamental mode (RR0)
RR Lyrae stars in the galactic bulge obtained by the MACHO Project.
I have extracted VOTable lightcurves for all of these
objects and they are available to study.