For a decade, gravitational-wave observatories have been detecting collisions between black holes across the Universe. Now, after analysing more than 150 binary black hole mergers, researchers have found evidence that these collisions can be divided into at least three distinct families, each with different characteristics and potentially different origins. 

The study, led by postdoctoral fellow Dr Sharan Banagiri from the ARC Centre of Excellence for Gravitational Wave Discovery (OzGrav) and Monash University, analysed the growing catalogue of gravitational-wave detections from the international LIGO–Virgo–KAGRA (LVK) Collaboration. Researchers found that black holes appear to cluster into three subpopulations, separated by distinct mass ranges and characterised by different spin and pairing behaviours. 

The findings, published in Physical Review Letters, suggest that the Universe may not produce merging black holes through a single dominant process, as many researchers once expected. 

“The population of binary black holes that we are discovering is complex enough that we cannot easily say it looks like one formation channel is making the vast majority of black holes,” Dr Banagiri said. 

Black holes can form in a variety of environments. Some might originate from pairs of massive stars born together that eventually collapse into black holes and merge. Others may form in dense stellar clusters where black holes dynamically capture one another, while some may grow through repeated mergers over time. 

The challenge for astronomers is that they cannot directly observe how individual black holes formed. Instead, they must work backwards from the mergers they detect through gravitational waves. 

Dr Banagiri compares the problem to finding a pile of leaves on the ground and trying to work out the different kinds of trees they came from without looking at the trees themselves. 

“You can look at the leaves and say maybe that’s a maple, maybe that’s a cherry tree. By the shape and geometry of the leaves, you can roughly say there are four kinds of trees contributing to this pile of leaves.” 

What researchers do is very similar. By analysing a large catalogue of black hole mergers, they can identify patterns in the data and group black holes with similar characteristics into distinct subpopulations. 

The researchers found that two key properties proved particularly useful: how fast black holes spin and how they pair with one another. 

Based on these characteristics, the analysis revealed three distinct subpopulations of merging black holes, separated by mass: one below about 28 times the mass of the Sun, a second between roughly 28 and 40 solar masses, and a third above 40 solar masses. 

The most massive black holes appear to spin faster and pair differently from their lower-mass counterparts. 

“Smaller black holes in binaries are relatively slower spinning, and they like to pair with other black holes that are roughly the same mass,” said Dr Banagiri. 

“Black holes that are already high mass, greater than 40 solar masses, spin faster and they like to pair with something that’s less massive.” 

One possible explanation is that some of the most massive black holes are themselves the products of earlier black hole mergers. In this scenario, known as hierarchical merging, a black hole formed in a previous merger later merges again, creating progressively larger black holes. 

While the new results are consistent with that picture, the researchers stress that more observations will be needed before any individual formation pathway can be confirmed. 

“The main discovery is that you can statistically separate the detections into different clusters,” said Dr Banagiri. 

“There are very interesting clues that are starting to become visible, but that link is not yet fully clear.” 

The work highlights how gravitational-wave astronomy is entering a new phase. Rather than simply detecting black holes, researchers are beginning to study their demographics and evolutionary histories. 

The findings do not change scientists’ understanding that black holes form when massive stars collapse. Instead, they provide new clues about what shapes the characteristics of black holes throughout their evolution.  

“The question is: what kinds of environments and what kinds of physics are dictating the properties of the black holes that we see?” said Dr Banagiri. 

As gravitational-wave detectors continue to improve and future observing runs deliver hundreds or even thousands more detections, researchers hope to refine the emerging picture of how black holes form and evolve across the Universe. 

Paper: 

Evidence for Three Subpopulations of Merging Binary Black Holes at Different Primary Masses

https://arxiv.org/abs/2509.15646

Authors: 

Sharan Banagiri, Eric Thrane and Paul D. Lasky 

Institutions: 

The ARC Centre of Excellence for Gravitational Wave Discovery (OzGrav) and Monash University