Advances in the interpretation and analysis of lunar occultation light curves

Abstract

Context. The introduction of fast 2D detectors and the use of very large telescopes have significantly advanced the sensitivity and accuracy of the lunar occultation technique. Recent routine observations at the ESO Very Large Telescope have yielded hundreds of events with results, especially in the area of binary stars, which are often beyond the capabilities of any other techniques. Aims. With the increase in the quality and in the number of the events, subtle features in the light curve patterns have occasionally been detected which challenge the standard analytical definition of the lunar occultation phenomenon as diffraction from an infinite straight edge. We investigate the possible causes for the observed peculiarities. Methods. We have evaluated the available statistics of distortions in occultation light curves observed at the ESO VLT, and compared it to data from other facilities. We have developed an alternative approach to model and interpret lunar occultation light curves, based on 2D diffraction integrals describing the light curves in the presence of an arbitrary lunar limb profile. We distinguish between large limb irregularities requiring the Fresnel diffraction formalism, and small irregularities described by Fraunhofer diffraction. We have used this to generate light curves representative of several limb geometries, and attempted to relate them to some of the peculiar data observed. Results. We conclude that the majority of the observed peculiarities is due to limb irregularities, which can give origin both to anomalies in the amplitude of the diffraction fringes and to varying limb slopes. We investigate also other possible effects, such as detector response and atmospheric perturbations, finding them negligible. We have developed methods and procedures that for the first time allow us to analyze data affected by limb irregularities, with large ones bending the fringe pattern along the shape of the irregularity, and small ones creating fringe amplitude perturbations in comparison to the ideal fringe pattern. Conclusions. The effects of a variable limb slope can be satisfactorily corrected. More complex limb irregularities could be fitted in principle with a grid search based on the standard analytical model, however this method is time consuming and does not lead to unique solutions. The incidence of the limb perturbations is relatively small, but its significance is increased with the use of very large telescopes due both to the footprint at the lunar limb and to the increased sensitivity. In general, we recommend to observe occultations using sub-pupils. This will be a necessary requirement with the next generation of extremely large telescopes.