The possibilities of the classification of simulated observations in the GAIA photometric system are investigated. The data simulated by Barselona group [gaia.am.ub.es] from the BaSeL-2.2 and NextGen2 libraries of synthetic spectra are used. At first, the flux distribution curves of synthetic spectra for NextGen model atmospheres [ftp.hs.uni-hamburg.de] are compared with the observed flux distribution curves of stars of few spectral classes [Straižys V., Sviderskienė Z., 1972]. It was shown, that just the atmospheric models of hotter stars (Teff > 5000 °K) predict the observed energy distribution quite well. The observations in the GAIA photometric system simulated by Barcelona group are classified using the method of comparison of the color indices of standard and investigated stars. The observations of stars of two magnitudes (G=15 mag and G=18 mag) in the C1M photometric system are investigated in more detail. The analysis of the results shows that the classification is more definite when the bank of more solid standards is used. The errors of classification are smaller for the hotter (Teff > 10000 °K) and more luminous (G=15 mag) stars. The interstellar extinction AV is defined more precisely for hotter stars and the stars with larger AV value (3 ≤ AV ≤ 5). The computing time necessary to classify one star using the method of comparison of the color indices of standard and investigated stars is measured . The classification algorithm is written in PERL. The time... [to full text]

<p>In this thesis we present a procedure by which synthetic photometry of a hydrodynamic model of star or star-like object can be calculated in a regime where the photosphere is not radially resolved. In order to properly model the unresolved photosphere, we present a method where pressure and density are integrated outward from the outermost resolved radius of the star and then interpolated in temperature-surface gravity space between a set of MARCS \citep{Gustafsson2008a} stellar atmosphere models. These interpolations are accurate to within 10\% of expected temperature values and are determined by minimizing the difference between the integrated pressure, density and surface gravity and that of the atmosphere model. Using the Monte Carlo Radiative Transfer code \texttt{radmc3d}\citep{Dullemond2012}, we produce blackbody spectra of stars and photometric light curves of equal and unequal mass detached binaries and a contact binary. Stellar blackbody spectra are accurate to better than 1\%. Resultant light curves have less scatter than existing methods, such as \texttt{shellspec}\citep{Budaj2004} and show the expected morphology. Our method allows for imaging directly from hydrodynamic simulations, with minimal user set-up. This procedure is designed with the intent of producing simulated photometry of stellar merger models.</p> / Master of Science (MSc)