Comparison of a ground-based microlensing search for planets with a search from space

We compare a space-based microlensing search for planets with a ground-based microlensing search originally proposed by D. Tytler. Perturbations of microlensing light curves when the lens star has a planetary companion are sought by one wide-angle survey telescope and an array of three or four follow-up narrow-angle telescopes distributed in longitude that follow events with high-precision, high time resolution photometry. Alternative ground-based programs are considered briefly. With the four 2 m telescopes distributed in longitude in the Southern Hemisphere in the Tytler proposal, observational constraints on a ground-based search for planets during microlensing events toward the center of the Galaxy are severe. Probably fewer than 100 events could be monitored per year with high-precision, high time resolution photometry, with only about 42% coverage on the average regardless of how many events were discovered by the survey telescope. Statistics for the occurrence and properties for Jupiter-mass planets would be meaningful but relatively meager 4 years after the program was started, and meaningful statistics for Earth-mass planets would be nonexistent. In contrast, the 14,500 events in a proposed 4 year space-based program (the Galactic Exoplanet Survey Telescope) would yield very sound statistics on the occurrence, masses, and separations of Jupiter-mass planets, and significant constraints on similar properties for Earth-mass planets. The significance of the Jupiter statistics would be to establish the frequency of planetary systems like our own, where terrestrial planets could exist inside the orbits of the giants.