In addition to calculating ensemble (homogenised) parameters for the sample of CoRoT stars in the Gaia-ESO Survey this week (blog post to appear later), I’ve been working with a student of Thomas Masseron’s. Masseron wanted to know if we could identify spectroscopic binary systems from limited, noisy photometry alone, and infer the system properties (e.g., stellar parameters of both systems, mass ratios). It’s a cool problem for a lot of reasons.
Spectroscopists often just throw away the binary systems because they aren’t worth the effort to analyse. The fraction of data thrown away for this reason is of order a few percent. That’s a lot of stars for big surveys, which means being able to identify these objects from photometry is a big win. There are obvious scientific extensions too: the binary fraction itself, binary fraction distributions for multiple populations within globular clusters, mass ratio distributions, etc. Without any astrophysical priors on mass/radius/luminosity ratios, it turns out you can identify these systems very easily with modest photometric data. However as one might expect, the quality of inference is dependent on the properties of individual systems: stars of similar mass and evolutionary states are much harder to distinguish, because you’re essentially just seeing a not-quite-right blackbody curve. The student (L. Orfali) will investigate the inference quality for different binary system properties, and see what is the minimum photometric quality (and in which bands) are required to constrain these systems. Spectroscopic modelling will occur next week too, but that part is trivial and easier to intuit.
Laura Watkins (STSCI) gave an excellent talk this week on the possible existence of an intermediate mass black hole at the center of omega Cen. Lots of exquisite data (HST and ground based spectra), with very detailed modelling. It’s an awesome project!