In February, scientists at the LIGO observatory made history when they announced the first ever detection of gravitational waves. These ripples in the fabric of space-time came from two black holes that spun around each other several times per second before merging in a violent, energetic explosion. Now, researchers have calculated the likely origins of those black holes. A new study argues that they probably came from two massive suns that formed about 12 billion years ago — or two billion years after the Big Bang.

Researchers came up with this origin story, published today in the journal Nature, by running a complex simulation called the Synthetic Universe; it’s a computer model that simulates how the Universe may have evolved since the start of the Big Bang. "We play God," said lead study author Chris Belczynski, an astrophysicist at Warsaw University. "We have a model of the entire Universe in our computer. We populate the computer with stars from the beginning, from the Big Bang, and you let them go ahead, evolve, produce black holes, etc."

The simulation even includes a synthetic LIGO detector to determine the types of objects that the observatory would detect over time. Using this model, Belczynski and his team were able to look back to the start of the Universe and calculate the types of stars that formed the black holes that LIGO detected.

Not only has this tool been useful for tracing the origins of stars, but it has also been great for making predictions. Based on this computer model, Belczynski first estimated back in 2010 that LIGO would mostly be detecting gravitational waves coming from mergers of big black holes. It was a prediction that went against the prevailing idea at the time. "Most of the astronomy community believed that mergers of neutron stars, not black holes, would be detected first and dominate LIGO detections," said Belczynski. But so far, the three possible detections that LIGO has made — two confirmed ones and a potential third — all fit within what Belczynski predicted six years ago.

And thanks to the Synthetic Universe, Belczynski predicts that many more detections are headed our way. His models show that LIGO will pick up to 60 detections when it begins its next observation run this fall. And when the observatory is at its peak sensitivity, it could pick up to 1,000 detections of black hole mergers each year, according to the model.

Most of the black hole mergers LIGO detects will be of a certain size too, according to Belczynski’s calculations. Their combined masses will likely be between 20 and 80 times the mass of our Sun. That fits with what LIGO has detected so far. The first gravitational waves ever detected — dubbed GW150914 — came from two black holes that were 36 solar masses and 29 solar masses, respectively; combined, that’s 65 times the mass of the Sun.

Those are fairly large black holes. Normally, the black holes found in our galaxy are around five to 10 solar masses, said Belczynski. Since the black holes that created GW150914 were particularly big, they must have come from very massive stars. Black holes often form when a star dies and collapses in on itself. The bigger the star when it dies, the larger the black hole it creates. "Both stars needed to be very massive, about 40 to 100 times more massive than our Sun," Belczynski said about the stars behind GW150914.

Those stars were also probably very low in metals, meaning they didn’t have many elements heavier than helium. Stars that are high in metals produce a lot of solar wind — a constant flow of charged particles from the upper atmosphere of a star. This wind takes mass away from the star over time. "If there was a lot of wind, the wind would take a lot of mass away, and you’ll form a small black hole, not a big one," said Belczynski. Since these black holes were so big when they merged, they likely didn’t have much solar wind wearing them down over time.

That’s why Belczynski’s models indicate that these stars are probably from the early Universe. Right after the Big Bang, metals were fairly scarce; only hydrogen and helium made up the first generation of stars. But those stars created heavier metals inside of them, and whenever they died and exploded, they spread those metals throughout the Universe as gas. That metallic gas helped to create the second generation of stars, which in turn produced more metals themselves and spread them throughout the Universe when the stars died. That cycle has continued for billions of years up until now, so many recent stars have lots of metals and lose a lot of mass when they evolve.

Of course, there is a small caveat. The Universe currently does have patches of galaxies with low-metal stars. "The universe is not uniform; evolution is not uniform," said Belczynski. So it’s possible that the stars that led to GW150914 could have formed more recently, though that’s less probable. "Our prediction is they formed way back in the past, and they were spiraling around each other and they hit each other just recently," he said.

Moving forward, Belczynski’s models show that LIGO will be detecting more black hole mergers similar to this one. And the more mergers that the observatory detects, the more it validates and refines what the Synthetic Universe model has predicted about star evolution. "We are now moving toward precision astronomical science with gravitational waves," said Christopher Fryer, a research scientist at Los Alamos National Laboratory, who was not involved with the study. "The detections are already telling us about the nature of massive star evolution."

Scientists have found gravitational waves