Artificial Intelligence Has Discovered 1,210 New Gravitational Lenses

An example of a gravitational lens found in the DESI Legacy Surveys data. There are four sets of lensed images in DESI-090.9854-35.9683, corresponding to four distinct background galaxies — from the outermost giant red arc to the innermost bright blue arc, arranged in four concentric circles. All of them are gravitationally warped — or lensed — by the orange galaxy at the very center. Image Credit: NOIRLab CC BY 4.0

Massive objects distort space-time, and some can warp it to such a degree where the light of distant background galaxies is turned into large magnified arcs and rings. Finding them, especially the biggest ones, is not an easy task, so researchers had to deploy a new approach: a machine learning algorithm designed specifically to find them. And it was greatly successful.

The algorithm went through the exquisite map produced by the DESI (Dark Energy Spectroscopic Instrument) Legacy Imaging Surveys, the largest ever produced. It was able to discover 1,210 new lenses, almost doubling the number of gravitational lenses known to humanity. The discovery is reported in a paper accepted for publication in The Astrophysical Journal.

“A massive galaxy warps the spacetime around it, but usually you don't notice this effect. Only when a galaxy is hidden directly behind a giant galaxy is a lens possible to see,” lead author Xiaosheng Huang from the University of San Francisco, said in a statement. “When we started this project in 2018, there were only about 300 confirmed strong lenses.”

Every massive galaxy has the conditions to be a strong gravitational lens, but only 1 in 10,000 have a background galaxy aligned to our line of sight to actually see such lensing events. These peculiar objects provide insight into some of the biggest mysteries of the universe, such as dark matter and dark energy.

Examples of gravitational lenses found in the DESI Legacy Survey data. Image Credit: KPNO/CTIO/NOIRLab/NSF/AURA/Legacy Imaging Survey CC BY 4.0

Doubling up the number of subjects to study will make an investigation using gravitational lenses even more precise. It is a game-changer approach and the beginning of more intriguing findings related to gravitational lensing.

“As a co-leader in the DESI Legacy Surveys I realized this would be the perfect dataset to search for gravitational lenses,” explains study co-author David Schlegel of Lawrence Berkeley National Laboratory (LBNL). “My colleague Huang had just finished teaching an undergraduate class on machine learning at the University of San Francisco, and together we realized this was a perfect opportunity to apply those techniques to a search for gravitational lenses.”

The DESI Legacy Surveys comes from the combined observations by three ground-based observatories over 1,405 nights, together with years of data from Kitt Peak National Observatory, Cerro Tololo Inter-American Observatory, and NASA’s Wide-field Infrared Survey Explorer (WISE). Its final data release, available now, was created by the work of 200 researchers. This gravitational lens study is one of the first published from this dataset, but more are coming.

“We designed the Legacy Surveys imaging project from the ground up as a public enterprise, so that it could be used by any scientist,” said study co-author Arjun Dey, from NSF’s NOIRLab. “Our survey has already yielded more than a thousand new gravitational lenses, and there are undoubtedly many more awaiting discovery"


If you liked this story, you'll love these

This website uses cookies

This website uses cookies to improve user experience. By continuing to use our website you consent to all cookies in accordance with our cookie policy.