Since 2003 we have acquired more than half a million images of a field half a degree square, containing the open star cluster M23. We shoot unfiltered images that are a few seconds in duration. These unfiltered images provide the greatest signal-to-noise ratio in the shortest time while offering unique challenges in frame normalization and calibration. We occasionally acquire images from other fields but our primary focus has been the M23 field to which we return season after season. This effort has led to a data set uniquely able to provide information on the stability of stellar signals on timescales ranging from seconds to decades.
By the end of the 2014 data season we had indisputable evidence of variability in about 60 of the approximately 1600 stars in our field of view. The vast majority of these stars are long-period variable stars and the majority of these variables would be classed as semi-regular variables. Most of these stars vary on timescales of a few tens of days to a few hundred days and have amplitudes of variation between about 0.2 and 0.5 magnitude. We are slowly deciphering the properties of these stars. Most are multi-periodic with two or more oscillation modes excited. We are searching for hints of even longer oscillation modes and for for evidence of whether these stars are on average getting brighter or dimmer. We are also interested in the stability of the different oscillation modes in a given star as well as potential outbursts these stars might show. In addition to these variables we are monitoring dozens more stars that are likely, but not yet indisputably, long-period variables. Here we have two images of a star we know affectionately as Star 356, taken several months apart. Yup, it got brighter.
We have found 6 stars in our field to be eclipsing binary systems, where one star periodically passes in front of the other from out vantage point. When this "eclipsing" occurs the apparent single star that we see gets dimmer. Timing these eclipses allows us to test several properties of the system including the rate of mass loss from the system, mass transfer from one star to the other, how the magnetic fields of the stars interact with one another and whether an unseen third star inhabits the system. Most of the eclipsing binaries we see are very close to one another, almost touching, with orbital periods on the order of a day or less. Shown below is a signal versus time graph (a light curve) for an eclipse in one of these systems.
When we started this work so many years ago we were most keenly interested in the very shortest timescale changes that might occur. Such a blip might arise from an outer solar system object blocking the light of a star briefly as it passes or a black hole microlensing a background star, causing it to appear brighter. This is long haul work with plenty of statistical testing and re-testing. We keep at it, searching for a few needles in very large haystacks. Of course, we are also sensitive to classical stellar flare events and a few apparent flares have appeared.
Read an Astronomy Magazine Web News Item describing this work.