The microlensing event MACHO 97-BLG-41 was first alerted by the MACHO team on 18 June 1997 as a possible short duration event toward the direction of the Galactic Bulge. On 29 June 1997, MACHO reported that the event had not dimmed as expected and on 2 July 1997, PLANET reported that its brightness was beginning to increase. On 23 July 1997, PLANET issued an electronic alert that the background source star was likely to cross a caustic in the lens magnification pattern within about 24 hours, as was later observed by many teams.
As the background star passes behind the binary lens, the bending of light rays by the gravitational field of the binary causes different amounts of light to reach the telescope at different times. This generates peaks in the light curve, a plot of the amount of light received as a function of time. The PLANET team light curve for MACHO 97-BLG-41 is shown below, with data from different PLANET observing sites coded in different colors. (Green = ESO/Dutch 0.9m in Chile; Red = SAAO 1m in South Africa; Blue = Canopus 1m in Tasmania; Yellow = CTIO 1m in Chile; Magenta = Perth 0.6m in Western Australia). The changing observed brightness of the background star is shown; each time mark represents five days on the horizontal scale.
The small peak near day 619 and the large peak near day 654 are due to the source crossing regions of very high magnification, called caustics, on the sky. Such caustic regions are not created by a single lens, but 1, 2 or 3 caustic regions can be caused by double lenses. No static (motionless) double lens can produce caustics of the right shape, separated by the right amount at the right time to explain the light curve of MACHO 97-BLG-41. But since all binary stars must orbit their common center of mass, the PLANET team decided to study whether or not orbiting lenses could rotate the caustic regions enough to explain the light curve. A successful rotating binary model was indeed found, and the resulting light curve is indicated by the line passing through the data points in the plot above. A sketch of the rotating binary and its rotating caustics is shown in the plot below. Click on the figures for a closer view.
The double lens is indicated by the two dots near the center of the frame; their positions near the first and second caustic crossings are shown in blue and red, respectively. The straight line shows the path taken by the background star through the magnification pattern of the lens. The two inset boxes show expanded views near the first caustic crossing (left) and second (right). The source (small open circle) first crosses the triangular caustic, but by the time that it has reached the larger central caustic, both the small triangular caustics and the central caustic have moved. This movement is caused by the movement of the binary lens from the blue position to the red position during that time. This movement also alters the shape of the light curve that would normally be caused by a motionless double lens in such a way that the observed light curve of MACHO 97-BLG-41 is generated. The amount of rotation suggests that the double lens orbits once in every 1.5 years or so. This is the first time that rotation has been measured in a microlensing event. The exact mass and distance to the binary are not known, but the most likely combinations for our solution yield orbital velocities small enough that the system is quite comfortably bound.
The PLANET team model also appears to be consistent with data available from the MACHO and GMAN teams for MACHO 97-BLG-41, even though it was not used in the modeling in any way and covers regions of the high peaks where no PLANET data exist. This comparision is shown below.
To read the technical details concerning MACHO 97-BLG-41 and the method we used to find this rotating binary solution, download our preprint at astro-ph/9910307 or the published paper in the Astrophysical Journal (2000, 534, 894).