Skip to Navigation | Skip to Content

Astronomy 211. Planetary and Stellar Astronomy, Spring 2012

Variable Star Discovery Lab


In Astronomy 211 we study the life cycles of stars. Many stars exhibit time variable light intensity in certain phases of their life cycle. While the celestial coordinates of more than a billion stars have been cataloged, only a tiny fraction of these have had their brightness tracked over time sufficiently to determine whether and how they are variable. Therefore, as a multipart laboratory, student teams were assigned 1) to discover variability in a star not previously known to vary, and 2) to follow up on the discovery until the period is known and the nature of the variability is clear. Upon completion, all results were reported to the Variable Star Index, a refereed international clearinghouse of variable star information maintained by the American Association of Variable Star Observers. The table below summarizes the discoveries. A brief description of the laboratory procedure follows the table.


Star Name


Variable Type

Light curve



HAT 260-0014331

8.377516(5) hour

contact binary star

V0352+2135 light curve

Josue Diaz,

Rick McWhirter

VSX J043234.6+284518

8.28945(1) hour

contact binary star

V0432+2845 light curve

Nathan Harkema,

Sam Van Kooten

VSX J043413.2+244241

5.671719(2) hour

contact binary star

V0434+2442 light curve

Kathy Gibes,

Morgan Smith

VSX J060754.9+232146

9.731(3) hour

contact binary star

Hannah Pagel,

David Hauck

VSX J065254.5+210440

7.0686(44) hour

contact binary star

V0652+2104 light curve

Nathan Steensma,

Joe Garbini,

Mike Troupos

VSX J120504.8+350022

11.049689(5) hour

pulsating RR Lyrae star (RRAB)

V1205+3500 light curve Dan Van Noord

Notes on the table:

  • Click on any star name to see the full entry in the VSX catalog.
  • The number in parenthesis following the value of the period is the one-sigma uncertainty on the final digit of the period
  • The light curves are plots of system brightness (in units of astronomical magnitude) versus phase of the periodic cycle.
  • Click on any figure to see a full size version of the light curve.

Laboratory Procedure

  1. Discovery: Rather than spend observing time scanning the sky at random in hopes of finding a variable star, we recycled observations taken in 2007 by students studying asteroid rotation. Data spanning three to six hours on a given night aimed at the position of a target asteroid (essentially a random place in the ecliptic) were analyzed with software written by Calvin student Dan Van Noord. The locations of all stars in the image were mapped, light curves (plots of brightness versus time) were generated and sorted by the likelihood of showing intrinsic variation. Students learned to sift out a variety of false positives and identify genuinely variable stars. Each team analyzed as many data sets as needed until a genuine variable was found. The position of the variable was checked in the Variable Star Index to see if its variability had been previously noted.
  2. Follow up data: We needed additional data for each star covering multiple cycles in order to establish whether the variations were periodic and to determine the shape of the cycle and length of the period. In a number of cases, one or two additional nights of data were available in our data base (as the asteroid that was being tracked was moving slowly). In one case (the fifth entry in the table) this alone was enough to establish the period. For three other cases (the first three entries in the table) we were able to obtain additional data from the Catalina Sky Survey. Although sparse and noisy, these data span six years and so can be used to make very accurate period determinations. In the remaining two cases it was necessary to take additional data with our Rehoboth telescope in order to solve for the period. (Five nights were needed for the fourth entry in the table and three for the sixth entry.)
  3. Determining periods: We used the FALC algorithm (Fourier Analysis of Light Curves) implemented by Peranso to determine the period and its confidence interval. The first step of this analysis was determing whether the period was unambiguous or whether more data were needed to rule out possible alternative periods. All of the stars we found did turn out to have strictly periodic variations although there are classes of variable stars that vary aperiodically.
  4. Variability type: The first five entries show approximately sinusoidal variation. Their shapes are a signature of contact binary star systems: two stars orbiting each other so closely that their outer atmospheres actually overlap. The two dips in light intensity occur as one star is nearly in front of the other and then the reverse. Those systems viewed more nearly edge on have eclipses with flat bottoms (like the second entry). Those systems with slightly unequal eclipse depths have stars of slightly different temperatures (like the fourth entry). The final entry has a very different shape: a large and rapid rise followed by a gradual decay and then a time of no change. This is the signature of an RR Lyrae system: a single star pulsating due to an instability in its outer envelope.
  5. Submission to the VSX and discovery credit: With the conclusion of the semester, all six stars were submitted to the VSX along with the light curves to document their type and period. For all but the first star, new VSX names were assigned based on the celestial coordinates of the star and discovery credit was given to the students. The first star, however, was independently discovered to be a variable leading to a catalog entry being made in the time between the initial student discovery and our submission of the data. However, this entry had both an incorrect period and variable type. Our submission was used by the VSX to correct this essential information about the star, although the students did not receive discovery credit.

posted 8/29/2012 by L. Molnar


Lorem ipsum dolor sit amet, consectetuer adipiscing elit. Nulla ut nibh.


Lorem ipsum dolor sitamet, consectetuer adipiscing elit. Nulla ut nibh.