Caustic Crossing Microlensing Event by Binary MACHOs and Time Scale Bias

Caustic crossing microlensing events provide us a unique opportunity to measure the relative proper motion of the lens to the source, and so those caused by binary MACHOs are of great importance for understanding the structure of the Galactic halo and the nature of MACHOs. The microlensing event 98-SMC-01, occurred in June 1998, is the first event for which the proper motion is ever measured through the caustic crossing, and this event may be caused by binary MACHOs as we argue in this Letter. Motivated by the possible existence of binary MACHOs, we have performed the Monte Carlo simulations of caustic crossing events by binary MACHOs and investigated the properties and detectability of the events. Our calculation shows that typical caustic crossing events have the interval between two caustic crossings ($t_{\rm cc}$) of about 5 days. We argue that with the current strategy of binary event search the proper motions of these typical events are not measurable because of the short time scale. Therefore the proper motion distribution measured from caustic crossing events suffers significantly from {`}time scale bias{'}, which is a bias toward finding long time scale events and hence slowly moving lenses. We predict there are two times more short time scale events ($t_{\rm cc}\le 10$ days) than long time scale events ($t_{\rm cc}\ge 10$ days), and propose an hourly monitoring observation instead of the nightly monitoring currently undertaken to detect caustic crossing events by binary MACHOs more efficiently.