2005ApJ...620.1043W

This paper presents the results of a comprehensive wide-field search for transiting ``hot Jupiter'' planets (gas giant planets with an orbital period in the range 1day≤P≤16days) in the globular cluster 47 Tuc. Motivated by the detection of the transit in HD 209458 and the apparent lack of planetary detections in the core of 47 Tuc by Gilliland and coworkers, this work further addresses the question of giant planet frequency in 47 Tuc by observing from the ground a 52'x52' field centered on the cluster. Hence, this work is most sensitive to the uncrowded outer regions, where the stellar densities are significantly lower than in the core, and concentrates on 21,920 main-sequence stars within 2.5 mag of the cluster turnoff (hence approaching the solar value in mass). Our work comprises the largest ground-based transit search of a globular cluster to date, incorporating a 33 night time series that allows us excellent sensitivity to detect hot Jupiter planets. Detailed Monte Carlo simulations incorporating the actual temporal sampling and photometric precision of the data predict that seven planets with orbital periods in the range 1-16 days should be present in our data set if 47 Tuc has the same planetary frequency as that observed in the solar neighborhood. A detailed search utilizing a matched filter algorithm, developed specifically for this project, found no transit events. This 3.3 σ result is consistent with the Hubble Space Telescope cluster core null detection of Gilliland and coworkers. Our result indicates that system metallicity rather than crowding is the dominant effect inhibiting hot Jupiter formation in this environment. The 33 night data set used for this result also led to the detection of 100 variable stars, including 69 new discoveries, which are presented in a companion paper.