When black holes collide, they do not all seem to follow the same cosmic script.

A sweeping new analysis of the latest gravitational-wave catalog suggests the universe is producing merging black hole pairs through several distinct channels, not one. Some of those systems appear to come from ordinary stellar evolution, while others carry signs of stranger histories, including black holes that may already be the remnants of earlier mergers.

The study draws on GWTC-5.0, the newest release from the LIGO-Virgo-KAGRA Collaboration, which includes data through the second part of the fourth observing run. In the population analysis, researchers examined 267 compact binary merger candidates, up from 161 in the previous catalog update. Of those, 259 were classified as binary black hole candidates for black-hole-only population studies.

That expanding sample is changing the field’s focus. Instead of treating each signal as a one-off curiosity, researchers can now look for patterns across the population.

“This set of nearly 400 gravitational-wave detections from LIGO and Virgo provides us with a clear indication that the binary black hole mergers we see are forming in several different ways,” said Sharan Banagiri, a research fellow at Monash University’s School of Physics and Astronomy and the ARC Centre of Excellence for Gravitational Wave Discovery, OzGrav. “Some might form as one giant cloud of gas that collapses to give two massive stars that then become black holes. Others might be black holes that wander into each other in dense environments called clusters that are packed with stars. While others are the product of a previous generation of mergers between two black holes.”

The new catalog adds 104 binary black hole systems from the O4b observing period, plus two reanalyzed sources from O4a. That larger pool let the team test whether the black hole population is smooth, or whether it breaks into recognizable groups.

The answer points to structure.

The researchers found that the black hole mass distribution still shows a strong concentration around 10 times the Sun’s mass, along with another feature near 35 solar masses. But the updated catalog weakens some earlier hints while sharpening others. A previously reported peak in mass ratio near 0.7 is now less convincing at the population level, with the data instead leaning more toward equal-mass pairings overall.

The spin picture also remains complicated. Most merging black holes do not appear to be spinning at extreme rates. Depending on the model, about 69 to 84 percent have dimensionless spin magnitudes of 0.5 or less. Even so, the catalog now contains stronger evidence for a smaller subset with at least one rapidly spinning black hole.

That matters because spin can preserve clues about where a system came from. A black hole born from a collapsing star may not look the same, statistically, as one left behind by an earlier black hole merger.

One of the strongest patterns in the new analysis is that rapidly spinning black holes seem to cluster in two mass ranges: one between about 10 and 20 solar masses, and another above roughly 45 solar masses.

“One of the most fascinating things we’ve discovered about these new black holes is that they are spinning very fast,” Banagiri said. “The sun rotates once every 25 days. If it became a black hole and started spinning as quickly as the ones we discovered, it would be rotating several thousand times every second. So where do these rapidly-spinning black holes come from? One leading explanation is that they are ‘hierarchical’ products of a previous generation of merger between two black holes.”

That hierarchical-merger idea appears repeatedly in the results. The study found that binaries with primary masses above about 40 solar masses likely follow a different mass-ratio pattern from lower-mass systems. Their secondary companions drop off more sharply at high masses, which means the heaviest black holes are more likely to pair with lighter partners than with equals.

The team also identified transitions in spin behavior near 13.5, 20.2, and 41.6 solar masses. Above the highest of those transition points, the spin distribution becomes more consistent with what researchers expect from mergers involving second-generation black holes, remnants of earlier black hole collisions that have reentered the merging population.

The estimated merger rate for this broad, approximately symmetric high-spin subpopulation is small compared with the total binary black hole rate, around 1 to 10 percent, but it is no longer easy to dismiss as statistical noise.

The 35-solar-mass feature remains one of the more intriguing parts of the black hole spectrum. In earlier analyses, it sometimes looked like a distinct peak. With the new data, the authors say it is better described as a change in slope in the mass distribution, especially because the secondary masses fall away faster than the primaries above that range.

Its origin remains unsettled.

The paper notes that pair-instability physics has often been discussed as a possible explanation, but the observed feature is not easily matched by that process alone. Other proposed routes include chemically homogeneous evolution, stable mass transfer, Population III binaries, dynamical formation in dense star clusters, and hierarchical mergers. None yet explains every observed property cleanly.

The study also strengthens the case that the black hole population cannot be understood with one simple formation story. At least 9 to 40 percent of mergers appear to require some degree of spin-orbit alignment, while roughly 30 to 46 percent have effective spins below zero, consistent with channels that can produce anti-aligned spins. The width of the effective spin distribution also seems to vary with mass ratio, and may broaden with redshift, though those trends remain somewhat model dependent.

Assistant professor Sylvia Biscoveanu of Princeton University, a co-author and former Fulbright postgraduate scholar at Monash, said the scale of the update matters as much as any one event. “GWTC-5 represents the largest single increase in the size of the gravitational-wave catalog, including events with remarkable properties such as GW241127, which contains BHs of very different masses with clearly wobbling orbits due to tilted spins. The new catalog also contains the event with the best localisation on the sky to date, GW240615.”

Professor Eric Thrane of Monash described the moment as a turning point for the field. “We are no longer just looking at individual anomalies, instead, we are seeing a true kaleidoscope of cosmic collisions. We are pushing the edges of what we know, seeing things that are more massive, spinning faster, and more unusual than ever before.”

The immediate payoff is not just a longer event list. It is a more reliable census of how black holes form, pair up, and merge across cosmic time.

With a larger catalog, astronomers can test whether certain masses, spins, and pairings really belong to separate sub-populations, and whether those groups trace different astrophysical environments such as isolated stellar binaries, dense star clusters, or repeated merger chains.

That also sharpens the search for predicted features such as the pair-instability mass gap and helps researchers estimate merger rates with smaller uncertainties.

As detector sensitivity improves and future catalogs grow again, the field will be better positioned to move from describing unusual events to mapping the full demographic history of compact objects.

Research findings are available online in the journal arXiv.