A team of asteroid hunters that includes researchers at the University of Washington says it has identified 27,500 new, high-confidence asteroid discovery candidates — not by making fresh observations of the night sky, but by sifting through archives of astronomical data.
The weeks-long database search was conducted by the Asteroid Institute, a program of the nonprofit B612 Foundation, in partnership with UW’s DiRAC Institute and Google Cloud.
The two institutes developed a program called THOR, which stands for “Tracklet-less Heliocentric Orbit Recovery.” THOR runs on a cloud-based, open-source platform known as ADAM (“Asteroid Discovery Analysis and Mapping”). The program can analyze the positions of millions of moving points of light observed in the sky over a given period of time, and link those points together in ways that are consistent with orbital paths.
Google Cloud’s Office of the CTO collaborated with the Asteroid Institute to fine-tune its algorithms for Google Cloud. The project analyzed 5.4 billion observations drawn from the NOIRLab Source Catalog Data Release 2.
“What is exciting is that we are using electrons in data centers, in addition to the usual photons in telescopes, to make astronomical discoveries,” Ed Lu, executive director of the Asteroid Institute, said in a news release.
Most of the 27,500 asteroid discovery candidates are in the main belt, between the orbits of Mars and Jupiter. But the candidates also include more than 100 apparent near-Earth asteroids.