The extension of the standard thin, flat disk eclipse mapping algorithm to a truly three-dimensional light curve fitting program provides a useful tool for (i) modelling light curves of nearly any shape of accretion disk and any brightness structure on the disk and the secondary star, and (ii) finding the brightness structure of the three dimensional accretion disk surface and the secondary star by fitting the light curves, accounting for primary and secondary eclipses, or any other change in brightness during the binary orbit. This algorithm allows a vast range of binary star and accretion disk geometries to be studied, which facilitates easy exploration of various geometries in order to understand observed light curves. However, caution is required when interpreting the results since the large number of possible geometries aggravates the problem of finding a unique and correct solution.
Figure 6: Three examples of a reconstructed radial temperature profile using
standard eclipse mapping for the case of a thick, flared accretion disk.
The theoretical profile is shown by the solid curves (note that the
lower two curves are displaced by 0.3 and 0.6 dex
relative to the top curve for
clarity). The orbital inclination and disk opening angle are
specified in the diagram
This program was used to construct light curves of flaring disks in eclipsing systems. Fits of these eclipse light curves with the incorrect assumption of a flat and thin disk are shown to still accurately reproduce the disks' radial temperature profile, provided that the inner disk is not obscured by the outer disk rim. This shows that standard eclipse mapping is capable of reproducing the disks' radial temperature profile even in the case of flared disks, contrary to what has been suggested by Smak (1994).
The most important usage of this light curve fitting method is expected to be in situations where both the secondary star and the accretion disk contribute substantially to the light curve, as is often the case at infra-red wavelengths. Also when the disk shape is obviously important, such as for example in high inclination systems, or in the case of tilted or warped disks, this program may serve as a helpful tool in fitting and predicting light curves.
Acknowledgements
I thank Vik Dhillon for many useful discussions, and two referees for valuable comments on the manuscript.