On the determination of the light curve parameters of detached active binaries

Abstract

Photospheric cool spots are responsible for wave-like distortions observed in the wide-band light curves of close active binaries. They affect the determination of the photometric parameters in eclipsing systems making it necessary to accurately model the spot effects in order to estimate the true luminosity ratio, the fractionary radii and the inclination of the orbital plane in a consistent way. We address such a problem for the prototype active binary RS Canum Venaticorum, for which the availability of a long-term and accurate series of light curves allows us to minimize the starspot effects on the estimation of the quoted parameters. We have analysed the available light curve sequence by using a new parallel code we specifically designed for such an application by solving simultaneously for the system parameters and spot configuration. The uniqueness and stability problems related to the spot mapping are solved by means of a regularization approach, following the method described by Lanza et al. ([CITE]). We evaluate the best set of photometric parameters and their confidence levels by using a statistical method based on the total $\chi^{2}$ criterion applied to the solution of the entire light curve sequence. The advantages offered by our approach include simplicity and generality with respect to other commonly used methods based on the covariance matrix. We discuss the quantitative effects of the starspots on the estimation of the parameters and compare our method and results with those of other authors. As an application of the derived results, we study the position of the components of RS CVn on the H-R diagram and compare them with computed evolutionary tracks. We find that the position of the primary component is compatible only by assuming a trigonometric parallax of ~8.2 mas, i.e., at the lower boundary of the confidence interval given by Hipparcos. Conversely, the effective temperature of the cooler component turns out to be ~5% smaller than expected on the basis of its radius. The comparison with evolutionary tracks also yields an estimate of $\log \tau \simeq 9.41$ years for the age of the system. Moreover, we address the high precision with which the inclination of the orbital plane can be estimated in eclipsing systems, despite the presence of an intense starspot activity. This opens the interesting possibility of detecting precessional motions in young RS CVn binaries, the spin angular momenta of which are not yet aligned with that of the orbital motion.