Statistics and Simulations of Transit Surveys for Extrasolar Planets

Abstract

The yields from transit surveys can be used to constrain the frequency and statistical properties of extrasolar planets. Conversely, planet frequencies can be used to estimate expected detection rates, which are critical for the planning and execution of these surveys. Here I review efforts to accomplish these two related goals, both of which generally require realistic simulations. Early attempts to predict planet yields generally resulted in overly optimistic detection rates that have not been realized. I point out where these estimates likely went wrong, and emphasize the strong biases and sensitivity to detection thresholds inherent in transit surveys. I argue that meaningful comparisons between observed and predicted detection rates require proper calibration of these biases and thresholds. In the few cases where this has been done, the observed rates agree with the results from radial velocity surveys for similar stellar environments. I then go on to describe recent, detailed calculations which should provide more accurate rates, which can be critically compared to observed yields. Finally, I discuss expectations for future all-sky synoptic surveys, which may have the sensitivity to detect hundreds or thousands of close-in transiting planets. Realizing the enormous potential of these surveys will require novel methods of coping with the overwhelming number of astrophysical false positives that will accompany the planet detections.