No abstract

Stellar masses and ages are not directly observable parameters, and the methods used to determine them are based on the calibrating relations. In particular, the mass-luminosity relation, based on the masses of less than 200 well-studied binaries, is virtually the only way to estimate the mass of single stars. Thus, the development of methods for estimating stellar masses with accuracy comparable to direct methods is a problem of vital importance.Here, we describe a method for estimating stellar masses and ages, which is based on the geometric similarity of evolutionary tracks for the stars at the same evolutionary stage in the Hertzsprung-Russell (HR) diagram. To examine the proposed approach, it has been applied to various test data sets. Application of the method, using synthetic stellar spectra Basel Stellar Library (of theoretical spectra; BaSeL), demonstrates that it allows determination of masses and ages of stars with a predictable distribution of uncertainties.This statistical approach allows us to demonstrate the viability of the method using it on the set of double-lined eclipsing binaries with intermediate-mass and low-mass components which allows us to compare calculated characteristics with observational ones. As a result, the uncertainties of the stellar masses estimated with the proposed method are comparable with the accuracy of ones obtained from direct observations. This allows us to recommend the method for mass estimates of masses of single stars by the localization in the HR diagram.As for the ages, the estimates for intermediate-mass stars are more reliable, while those obtained for low-mass stars are very uncertain, due both to slower movement of these stars in the HR diagram with age at stages close to the main sequence and to certain disagreements between theoretical models for this mass range.Well-known correlation between mass of the star and its observational characteristics (radius, luminosity, temperature) provides observational relations like 'mass-luminosity', 'mass-radius', etc.

Virtual Observatory facilities allow users to make a fast and correct cross-matching of objects from various surveys. They yield multi-color photometry data (color indices) on registered objects and make it possible to determine stellar parameters. A method of catalogue cross-matching, as well as its application to various areas in the sky and preliminary results of stellar parameterization, are discussed in the paper. We also make a critical analysis of modern stellar spectral atlases.

Detection of a composite flux in photometry can serve as an indication of a photometrically unresolved binarity and can contribute to the parameterization of the components of binary systems. A main goal of the present study is to develop a method of automatic photometric detection of binaries, based on multi-color photometry, theoretical stellar spectral energy distributions and general understanding of binary evolution. In particular, we consider an ultraviolet photometry where, in combination with optical and infrared photometry, interstellar reddening can be easier distinguished from temperature reddening.The following procedure is applied to achieve the declared goal. One can compose possible pairs of components, based on evolution concept. This can be done for various stages of binary evolution. Theoretical spectral energy distributions and response functions of ultraviolet photometric bands in appropriate sky surveys allow us to compute color-indices of such pairs, when they are unresolved, as well as of single stars. Usage of an interstellar extinction law gives us theoretical color-indices of reddened objects, both single stars and unresolved binaries. When plotted on a multidimensional color space, they allow us to indicate areas, where unresolved binaries can be easily separated from single stars, and identify binaries among objects, crossmatched in photometric surveys.