Previous
abstract 
Next
abstract
We recently reported a spectrum synthesis program for binary stars (Linnell \& Hubeny, 1994, ApJ, 434, Oct.20). That program now has been extended to include the case of an optically thick accretion disk associated with either of the two stellar components. Our model of the accretion disk uses the Pringle expression (Pringle, 1981, ARA\&A, 19, 137) for $T_{\rm eff}$ values on the accretion disk face, and the results of Hubeny and Plavec (1981, ApJ, 102, 1156) for rim $T_{\rm eff}$ values. The treatment of the stellar components is the same as in our 1994 paper. The program divides the rim into NRIM latitude values, typically 9, and divides the visible accretion disk face into NRING concentric ring boundaries, typically 31. The individual rings (for both the rim and the face) subdivide into NSEG pixels, typically 101. An individual synthetic spectrum, appropriate to the local $T_{\rm eff}$ value, is attached to each pixel. For illustration purposes we have used synthetic spectra prepared from Kurucz atmospheres.
The extended program constructs a synthetic spectrum for the accretion disk face, rim, the separate stellar components, and the entire system by producing a sum of contributions, at each wavelength (with due allowance for Doppler shift), from each visible pixel on the accretion disk or the separate stellar components, weighted by the projected area of the pixel. A separate program establishes a visibility key for each pixel and cosine of the zenith angle of the observer as seen from each pixel, for the current value of orbital inclination and orbital longitude. These data combine with synthetic spectra in the spectrum synthesis program to determine line of sight light intensities at each wavelength, i.e., the contributions needed for the sum. Separate data from related programs permit a plot of the system projected on the plane of the sky. This project received partial support from NSF grant AST9020459.
