Realisation of a fully-deterministic microlensing observing strategy for inferring planet populations

Within less than 15 years, the count of known planets orbiting stars other than the Sun has risen from none to more than 400 with detections arising from four successfully applied techniques: Doppler-wobbles, planetary transits, gravitational microlensing, and direct imaging. While the hunt for twin Earths is on, a statistically well-defined sample of the population of planets in all their variety is required for probing models of planet formation and orbital evolution so that the origin of planets that harbour life, like and including ours, can be understood. Given the different characteristics of the detection techniques, a complete picture can only arise from a combination of their respective results. Microlensing observations are well-suited to reveal statistical properties of the population of planets orbiting stars in either the Galactic disk or bulge from microlensing observations, but a mandatory requirement is the adoption of strictly-deterministic criteria for selecting targets and identifying signals. Here, we describe a fully-deterministic strategy realised by means of the ARTEMiS (Automated Robotic Terrestrial Exoplanet Microlensing Search) system at the Danish 1.54-m telescope at ESO La Silla between June and August 2008 as part of the MiNDSTEp (Microlensing Network for the Detection of Small Terrestrial Exoplanets) campaign, making use of immediate feedback on suspected anomalies recognized by the SIGNAL-MEN anomaly detector. We demonstrate for the first time the feasibility of such an approach, and thereby the readiness for studying planet populations down to Earth mass and even below, with ground-based observations. While the quality of the real-time photometry is a crucial factor on the efficiency of the campaign, an impairment of the target selection by data of bad quality can be successfully avoided. With a smaller slew time, smaller dead time, and higher through-put, modern robotic telescopes could significantly outperform the 1.54-m Danish, whereas lucky-imaging cameras could set new standards for high-precision follow-up monitoring of microlensing events. (C) 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim