New discoveries about supernovas offer clues to the expansion of the universe

Researchers from Swinburne University of Technology have contributed to a landmark study that complicates our understanding of the universe.

The work is published on the arXiv preprint server.

The Dark Energy Survey (DES), the results of which were released today, represents the work of more than 400 astrophysicists, astronomers and cosmologists from more than 25 institutions.

DES scientists collected data for 758 nights over six years to understand the nature of dark energy and measure the rate of expansion of the universe. They discovered that the density of dark energy in the universe could have varied over time, according to a complex new theory.

Dr Anais Möller from the Center for Astrophysics and Supercomputing at Swinburne University of Technology was part of the team working on this groundbreaking analysis, alongside Swinburne’s Mitchell Dixon, Professor Karl Glazebrook and Emeritus Professor Jeremy Mold.

“These results, the result of a collaboration between hundreds of scientists from around the world, demonstrate the power of cooperation and hard work to achieve major scientific advances,” says Dr. Möller.

“I’m very proud of the work we’ve done as a team; it’s an incredibly in-depth analysis that reduces our uncertainties to new levels and shows the power of the Dark Energy Survey. We didn’t just use data , but also developed pioneering methods to extract the maximum information from the Supernova survey. I am particularly proud of this, because I developed the method for selecting the supernovae used for the survey with the machine learning.

In 1998, astrophysicists discovered that the universe was expanding at an accelerating rate, attributed to a mysterious entity called dark energy, which makes up about 70% of our universe. At the time, astrophysicists agreed that the expansion of the universe should slow down because of gravity.

This revolutionary discovery, made by astrophysicists through the observation of specific types of exploding stars, called type 1a supernovae, was rewarded with the Nobel Prize in Physics in 2011.

Now, 25 years after the initial discovery, the Dark Energy Survey is the culmination of a decade of research by scientists around the world, who analyzed more than 1,500 supernovas using the tightest constraints ever on the expansion of the universe. This is the largest number of Type 1a supernovae ever used to constrain dark energy from a single study of large cosmic epochs.

The results are consistent with the now standard cosmological model of an accelerating expansion universe. However, the results are not definitive enough to rule out a possibly more complex model.

“There is still much to discover about dark energy, but this analysis can be considered the gold standard in supernova cosmology for some time,” says Dr. Moller. “This analysis also brings innovative methods that will be used in the next generation of investigations, allowing us to take a step forward in the way we do science. I’m excited to uncover more about the mystery that ‘is dark energy in the coming decade.’

Pioneer of a new approach

The new study pioneered a new approach to using photometry, with four unprecedented filters, to find supernovae, classify them and measure their light curves. Dr. Möller created the method to select these Type 1a supernovae using modern machine learning.

“It’s a very exciting time to see this innovative technology harnessing the power of large astronomical studies,” she says. “Not only are we able to obtain more Type 1a supernovae than before, but we have tested these methods extensively because we want to make more precise measurements of the fundamental physics of our universe.”

This technique requires data on Type 1a supernovae, which occur when an extremely dense dead star, known as a white dwarf, reaches critical mass and explodes. Since the critical mass is almost the same for all white dwarfs, all type 1a supernovae have approximately the same true luminosity and any remaining variations can be calibrated. So, when astrophysicists compare the apparent luminosities of two Type 1a supernovae as seen from Earth, they can determine their relative distances from us.

Astrophysicists trace the history of cosmic expansion with large samples of supernovae spanning a wide range of distances. For each supernova, they combine its distance with a measurement of its redshift, which is how quickly it is moving away from Earth due to the expansion of the universe. They can use this history to determine whether the density of dark energy has remained constant or changed over time.

The results found w = –0.80 +/- 0.18 using only supernovae. Combined with complementary data from the European Space Agency’s Planck telescope, w reaches –1 in the error bars. To reach a definitive conclusion, scientists will need more data through further investigation.

DES researchers used advanced machine learning techniques to help classify supernovas. Among data from around two million galaxies observed remotely, DES discovered several thousand supernovae. Scientists ultimately used 1,499 Type 1a supernovae with high-quality data, making it the largest and deepest supernova sample ever collected by a single telescope. In 1998, Nobel Prize-winning astronomers used just 52 supernovae to determine that the universe was expanding at an accelerating rate.