We investigated the period variation for 79 eclipsing binary systems using 20 years (1990-2009) of EROS, Macho, and OGLE survey observations. We discovered 9 apsidal motions, 8 mass transfers, 5 period increasing and decreasing systems, 12 light-travel-time effects, 5 eccentric systems and 40 other systems showing no period variations. We select 3 representative eclipsing binary systems; EROS 1052 for apsidal motion, EROS 1056 for mass transfer, and EROS 1037 for the light-travel-time effect. We determine the period variation rate (dP/dt), orbital parameters of the 3rd body (e3, ω3, f(m3), P3, T3), apsidal motion parameters (dω/dt, U, Ps, Pa, e) and apsidal motion period by analyzing the light curves and O-C diagrams.

Productivity of the Giant Magellan Telescope is estimated based on the annual number of papers produced by the currently operating large telescopes such as the telescope at the ESO La Silla observatory, CFHT, AAT, the Magellan telescopes, ESO VLT, Japanese Subaru, the Gemini telescopes, and the Keck telescopes. We find that the amount of papers produced by a large telescope is roughly proportional to the diameter of its primary mirror. With this fact, we estimate the SCI-paper productivity of the Giant Magellan Telescope by extrapolating the productivity of the above-mentioned large telescopes. Moreover, according to the paper written in 2001 by Benn and Sanchez, the amount of highly-cited papers produced by a large telescope is roughly proportional to the light-gathering power of the telescope or the square of the diameter. Hence, we survey the productivity of Nature-class papers of the large telescopes and extrapolate the relationship to estimate the productivity of the Nature-class papers by using the Giant Magellan telescope of a filled aperture 21.4 meters in diameter. We expect that Korean astronomers will be able to produce annually 60 SCI-class papers and 20 Nature-class papers with high scientific impact by using the telescope-time corresponding to the 10% share of the Giant Magellan Telescope.