A few years ago, a research team dedicated to finding killer asteroids before they kill us came up with a cool trick.
Instead of scanning the sky with telescopes to look for asteroids, scientists scoured old photos of the night sky and wrote an algorithm that discovered about 100 asteroids that had been overlooked in those images.
On Tuesday, scientists from the Asteroid Institute and the University of Washington unveiled an even larger bounty of 27,500 newly identified solar system bodies.
That's more than what all the world's telescopes discovered last year.
“This is a huge change,” said Ed Lu, executive director of the institute, which is part of the B612 Foundation, a non-profit organization Dr. Lu helped found, about the way astronomy research is conducted.
The discovery includes about 100 near-Earth asteroids, which are space rocks passing within Earth's orbit. None of the 100 are likely to hit Earth anytime soon. But the algorithm could prove to be a key tool in spotting potentially hazardous asteroids, and the research supports “planetary defense” efforts being undertaken by NASA and other organizations around the world.
Most of the space rocks the lab has identified lie in the main asteroid belt between the orbits of Mars and Jupiter. Others, known as Trojans, are trapped in Jupiter's orbit. The search also discovered smaller worlds even further away, known as Kuiper Belt objects, beyond Neptune's orbit.
“There’s a lot of good science here,” said Dr. Lu, a former NASA astronaut who noted that the key to future astronomical discoveries may not be more hours of observation with telescopes, but more powerful computers that can already process the vast amount of observational data. .” he said. We have gathered.
Historically, astronomers have discovered new planets, asteroids, comets and Kuiper Belt objects by imaging the same swath of the sky multiple times in a single night. The patterns of distant stars and galaxies do not change. However, much closer objects within the solar system will move noticeably within a matter of hours.
Multiple observations of a moving object, called a “tracklet,” provide enough information to sketch its path and give astronomers a good idea of where to observe and fix its orbit on different nights. .
Other astronomical observations inevitably include asteroids, but only at a single time and location, not the multiple observations required to construct a track.
The 412,000 images in the digital archive of the National Laboratory for Optical Infrared Astronomy (NOIRLab) contain approximately 1.7 billion dots of light that appear in just one image.
The algorithm used in the current study, known as Tracklet-less Heliocentric Orbit Recovery (THOR), can connect a point of light visible in one image with another point of light in another image taken on a different night. Another telescope reveals that these two points are actually the same object. An asteroid that usually changes position as it orbits the sun.
THOR's ability to identify asteroid candidates from different images is a difficult computational task that would have been impossible not long ago. However, Google Cloud, a distributed computing system, was able to perform the calculation in about 5 weeks.
“This is an example of what’s possible,” said Massimo Mascaro, technical director in the office of Google Cloud’s Chief Technology Officer. “We can’t even quantify how much opportunity there is in terms of the data already collected, and analyzing it with the right calculations could lead to more.”
Dr. Lu said improved software tools have made it easier to utilize computing power. When scientists no longer need huge software engineering teams to retrieve data, “that's when really interesting things can happen,” he said.
The THOR algorithm could also transform the operations of Chile's new Vera C. Rubin Observatory, which is expected to begin operations next year. The 8.4-meter telescope, funded by the National Science Foundation and the Department of Energy, repeatedly scans most of the night sky to track changes over time.
Currently, the Rubin telescope will scan the same part of the sky twice a night in a cycle designed to detect asteroids. THOR eliminates the need for a second pass through the telescope, allowing it to cover twice the area.
“Most science programs are willing to move from the default cadence of two observations to one observation per night,” said Zeljko Ivezic, a professor of astronomy at the University of Washington who serves as director of Rubin Construction.
This algorithm could increase the number of asteroids that Rubin can discover, which would be enough to meet a mandate passed by Congress in 2005 to find 90% of near-Earth asteroids that are at least 460 feet in diameter.
“The latest estimate is that it’s around 80%,” Dr. Ivezic said. He said, “With THOR, we could probably get it up to 90%.”