Spacetime Simulations and the Discovery of Gravitational Waves
In 1916, Einstein’s general theory of relativity predicted that massive cosmic events, such as the collapse and merger of two black holes, would cause distortions in spacetime that ripple across the universe. A century later, these ripples, called gravitational waves, were physically detected on Earth—a momentous discovery by the Laser Interferometer Gravitational-wave Observatory (LIGO) that led to the 2017 Nobel Prize in Physics.
In his Nobel lecture Kip S. Thorne, one of three researchers awarded the prize for LIGO, described the importance of numerical relativity simulations to the search for gravitational wave signals. “With just pencil and paper and analytical techniques, we could not solve Einstein's equations to compute the shapes of the gravitational waves,” he explained. “We needed simulations to extract information from the colliding black holes [that produce them].” Thorne went on to discuss the decades of groundbreaking work that went into developing this capability.1
Searching for the Signal over the Decades
Among the simulations Thorne credited were those first produced in 2004 by a team of researchers at NASA’s Goddard Space Flight Center, led by astrophysicist Joan Centrella. The NASA team’s simulations of orbiting and merging black holes represented an enormous breakthrough in numerical relativity. At the time, they were the largest astrophysical calculations ever performed on a NASA computer—specifically, the Columbia supercomputer at the NASA Advanced Supercomputing (NAS) facility, located at Ames Research Center in Silicon Valley. In subsequent years, NASA supercomputers would contribute even further to the search, discovery, and characterization of gravitational waves.
[The NAS team’s] visualizations opened our eyes to what was happening as the black holes merged.
Unlike visible light, infrared, ultraviolet, radio waves, X-rays, and gamma rays, gravitational waves are not part of the electromagnetic spectrum, so they can’t be found using traditional methods of astronomy. To detect them, scientists had to first figure out exactly what sort of signal they were looking for. A fundamental part of this inquiry included the advancement of numerical relativity simulations, but for many years it was impossible to simulate orbiting and merging black holes—the type of event most likely to produce waves strong enough to detect on Earth.
As Thorne explained in his Nobel lecture, the problem was extremely difficult to solve because merging black holes are not simply events that take place inside spacetime. “You’re simulating things that are made out of warped space and time,” he said. “So, you don’t have an arena in which this is going on; you’re simulating the arena itself.”
A Breakthrough in Numerical Relativity
Back in 2004, several research teams working on the problem—including Centrella’s group—achieved the breakthrough Thorne was waiting for. The NASA researchers developed a way to achieve realistic simulations of binary black holes as they spiraled together and eventually collided, producing gravitational waves.
Their extensive calculations took 85 hours to complete across 2,032 processors of Columbia, a feat that would have taken two decades on a single processor. The scientists worked closely with the NAS Division’s Data Analysis and Visualization team, who helped them shepherd their data through the supercomputer. “Once the simulations were complete, we provided high-end visualization and data analysis so the researchers could see and understand what was going on,” said group lead Chris Henze.
The breakthrough kicked off a prolific period for numerical relativity researchers. Through simulation, they were finally able to show how gravitational waves were emitted and could characterize the resulting signatures, known as chirps, that would eventually show up in the LIGO data.
“We could not have done it without Chris and the NAS team,” Centrella said. “Their visualizations opened our eyes to what was happening as the black holes merged, helping us understand the simulations leading up to and including the result.”
Matching Simulation to Observation
For the next decade, research teams at NASA and all over the world ran thousands of possible scenarios of events that generate gravitational waves, building a large database of simulated chirps that would serve as templates for comparison with observations. Using these templates, the LIGO team was able to confirm that the chirp picked up by the observatory on September 14, 2015, was the real deal: it was a gravitational wave.
Once the detection was announced, the scientists again turned to the visualization experts in the NAS Division to develop new videos that could help communicate the findings to other researchers, and to the public. One of these, the visualization shown above, was produced by simulating the actual event that generated the historic gravitational wave—based on LIGO’s observational data—on NASA’s current Pleiades supercomputer. This simulation is the closest researchers have come to revealing the merging black holes that sent the wave rippling through the universe on its way to Earth.
The NAS team’s stunning video was recently adapted for Europe’s largest planetarium show, currently on view at the Cite des Sciences et de l’Industrie in Paris.
A New Field of Science
Far beyond their initial discovery, the ability to detect gravitational waves promises to transform our understanding of the universe by launching an entirely new field of science: gravitational wave astronomy. Unlike signals on the electromagnetic spectrum, gravitational waves are unimpeded by the dust and gas that can block our view of objects in distant galaxies. They may even allow us to peek back to the first years after the Big Bang, an era that is invisible to electromagnetic observation.
Centrella is deeply honored that her team was part of the hundred-year journey from prediction to discovery. “It’s very exciting to know that we contributed in a significant way to opening the gravitational wave window on the universe.”
—Michelle Moyer, NASA/Ames Research Center
1Kip S. Thorne – Nobel Lecture. NobelPrize.org Nobel Media AB 2019. Accessed Fri. 5 Apr 2019.
