The Effect of Star Spots on Accurate Radius Determination of the Low Mass Double-lined Eclipsing Binary GU Boo

TL;DR

This study investigates how star spots affect the accuracy of radius measurements in low-mass eclipsing binaries, demonstrating that proper spot modeling yields radius estimates within about 2% accuracy.

Contribution

It shows that accurate spot modeling of light curves allows for precise radius determination of low-mass stars in eclipsing binaries, confirming previous measurements within a 2% margin.

Findings

Spot modeling reduces systematic errors in radius measurements.

Derived radii agree with previous studies within 2%.

Proper modeling yields 2% accuracy in radius determination.

Abstract

GU Boo is one of only a relatively small number of well studied double-lined eclipsing binaries that contain low-mass stars. L'opez-Morales & Ribas (2005, hereafter LR05) present a comprehensive analysis of multi-color light and radial velocity curves for this system. The GU Boo light curves presented in LR05 had substantial asymmetries, which were attributed to large spots. In spite of the asymmetry LR05 derived masses and radii accurate to about 2%. We obtained additional photometry of GU Boo using both a CCD and a single-channel photometer and modeled the light curves with the ELC software to determine if the large spots in the light curves give rise to systematic errors at the few percent level. We also modeled the original light curves from LR05 using models with and without spots. We derived a radius of the primary of 0.6329 +- 0.0026 Solar-Radii, 0.6413 +- 0.0049 SR, and 0.6373 +- 0.0029 SR from the CCD, photoelectric, and LR05 data, respectively. Each of these measurements agrees with the value reported by LR05 (R1=0.623 +- 0.016 SR) at the level of about 2%. In addition, the spread in these values is about 1-2% from the mean. For the secondary, we derive radii of 0.6074 +- 0.0035 SR, 0.5944 +- 0.0069 SR, and 0.5976 +- 0.0059 SR from the three respective data sets. The LR05 value is R2=0.620 +- 0.020 SR, which is about 2-3% larger than each of the three values we found. The spread in these values is about 2% from the mean. The systematic difference between our three determinations of the secondary radius and that of LR05 might be attributed to differences in the modelling process and codes used. Our own fits suggest that, for GU Boo at least, using accurate spot modelling of a single set of multicolor light curves results in radii determinations accurate at the 2% level.