A Massive Reanalysis of Microlensing Events

Researchers have reanalyzed nearly 10,000 light curves from the Optical Gravitational Lensing Experiment. The resulting catalog, which is publicly available, provides new opportunities to study black holes, exoplanets, and much more.

Making Sense of Microlensing

illustration of a gravitational microlensing event

An illustration of a gravitational microlensing event. In this case, the lensing object is a star with an exoplanet in orbit around it. Click here to see an animation of this event. [NASA]

When one astronomical object passes in front of another, the background object’s light is lensed, or focused, by the foreground object’s gravity, and we see a temporary jump in the background object’s brightness. This process of gravitational microlensing can clue us in to the presence of objects that emit little or no light, like black holes, exoplanets, and dark matter candidate objects, as they pass in front of stars or other luminous sources. Gravitational microlensing events are one of the most promising ways to find isolated stellar-mass black holes, which have long been difficult to track down.

The Optical Gravitational Lensing Experiment (OGLE) has observed more than 10,000 microlensing events since the survey began in 1992. But recording the event is just the first step to understanding what caused it. Researchers model microlensing light curves to estimate the properties of the objects involved, but many factors can complicate these calculations; our vantage point — Earth — is in constant motion, stars vary in brightness for a multitude of reasons, and instruments aren’t perfect. How can we account for all those factors and extract useful information from a microlensing light curve?

plot of the magnitude of a target of the OGLE survey as a function of time

An example of a gravitational microlensing event drawn from the sample analyzed in this work. In this event, OGLE BLG 156.7.141434, the brightness of the background source is variable. Click to enlarge. [Golovich et al. 2022]

New and Improved

That’s where today’s article comes in. In a new publication, a team led by Nathan Golovich (Lawrence Livermore National Laboratory) reanalyzed nearly 10,000 microlensing events in the third and fourth OGLE catalogs. The team’s new model accounts for Earth’s motion — which affects our perception of how quickly the background and foreground objects move relative to one another — as well as variability in the brightness of the background object and systematic instrumental effects.

This type of model has been applied to a single microlensing event, but it has never been used on a full survey because of the immense computing power it requires — Golovich and coathors used about a million hours of computer processing time to analyze their sample! The team showed that their model was able to separate the desired signal from competing factors like Earth’s motion and the variability of the object being lensed, greatly reducing sources of bias.

A Curated Catalog

plot of a microlensing event and models fit to that event's light curve

An example light curve and model fit for OGLE BLG 156.7.141434, the same event shown in the previous figure. Click to enlarge. [Golovich et al. 2022]

What does this updated catalog mean for the search for isolated black holes? Golovich and collaborators used the open-source Population Synthesis for Compact object Lensing Events tool (PopSyCLE) to simulate microlensing events and identify locations in parameter space that isolated black holes are likely to inhabit. Based on the results of these simulations, the authors estimate that 50% or more of the 390 OGLE events in that region of parameter space are most likely due to foreground black holes.

The catalog — the largest of its kind, to date — is free to anyone who wishes to use it; if you’re interested in black holes, exoplanets, or any other kind of dark object, there’s no better time to be on the hunt!


“A Reanalysis of Public Galactic Bulge Gravitational Microlensing Events from OGLE-III and -IV,” Nathan Golovich et al 2022 ApJS 260 2. doi:10.3847/1538-4365/ac5969