(ORDO NEWS) — Researchers have found more than 1,700 asteroid trails in archived Hubble data over the past 20 years.
Although many of these asteroids were previously known, more than 1,000 asteroids are not known. What good is another 1,000 asteroids? Like all asteroids, they may contain valuable information about the history of the solar system.
As time passes and more telescopes make more and more observations, their combined archive data continues to grow. Sometimes these data hide discoveries that wait for new analytical tools or new efforts by scientists before they are known. This is exactly what happened in a project called Hubble’s Asteroid Hunter.
In 2019, a group of astronomers launched the Hubble Asteroid Hunter Project. This is a citizen science project on the Zooniverse platform. Their goal was to comb through the Hubble data to find new asteroids.
The astronomers have published the results of their project in a new paper called “Hubble Asteroid Hunter I. Identification of asteroid trails in images from the Hubble Space Telescope.”
The study is published in the journal Astronomy and Astrophysics. The lead author is Sandor Kruk of the Max Planck Institute for Extraterrestrial Physics.
“One astronomer’s trash can become another astronomer’s treasure,” Crook said in a press release.
The data they were looking for was largely discarded by other observations not aimed at asteroids. In many cases, this data looked like “noise” and was removed to highlight different elements. But all this secondary, unexplored data is still archived and available.
“The amount of data in the astronomical archives is growing exponentially, and we wanted to use this amazing data,” Crook said.
The project has studied more than 37,000 Hubble composite images. They were taken between April 30, 2002 and March 14, 2021 using the Advanced Camera for Surveys and Wide Field Camera 3 aboard the Hubble Space Telescope. Most of the shots are 30 minute exposures, so the asteroid trails look like curved streaks.
These bands are at the heart of the problem: computers have a hard time detecting them. This is where the Zooniverse platform and citizen scientists come in.
“Due to the orbit and motion of the Hubble itself, the bands in the images look curved, which makes it difficult to classify asteroid trails – more precisely, it is difficult to explain to a computer how to automatically detect them,” Shandor Kruk explained.
“So we needed volunteers to do the initial classification, which we then used to train the machine learning algorithm.”
Volunteers helped. 11,482 citizen scientists participated in image classification. The Hubble Asteroid Hunter page on the Zooniverse received more than 2 million clicks, with volunteers giving 1,488 positive classifications in about 1 percent of the images.
The work of the citizens who took part in the project made it possible to train the machine learning algorithm to quickly and accurately search for other images.
The algorithm exists in the Google cloud, and after training, it helped detect 900 more objects, for a total of 2,487 potential asteroid trails in the Hubble data.
Then professional scientists played their part. Three authors of the article, including lead author Shandor Kruk, analyzed the results.
They ruled out things like cosmic rays and other objects, resulting in 1,316 Hubble images showing 1,701 trails. About a third of these were known asteroids, leaving 1,031 unidentified asteroid trails.
Follow-up observations will confirm how many are newly discovered asteroids and determine their orbits. Some of the 1031 will most likely not be confirmed, but the rest will help refine our understanding of the solar system’s asteroid population.
These asteroids have escaped detection because they are dimmer and most likely much smaller than most asteroids found from Earth.
This article is the first to be released as part of the Hubble Asteroid Hunter Project. In subsequent papers, the authors will use the curved shape of asteroid tracks to determine their orbits and distances to them.
All asteroids are the remnants of the early period of the existence of the solar system, mainly before the formation of the planets.
They are like natural time capsules that have preserved the conditions of the early system. That’s why astronomers are so interested in them and why we’ve sent spacecraft to collect samples from asteroids like Bennu and Ryugu.
“Asteroids are leftovers from the formation of our solar system, which means that we can learn more about the conditions when our planets were born,” Crook explained.
More and more researchers are using archival data like this. Browsing through existing images in search of new discoveries is cost effective and rewarding.
“Using archived data from survey campaigns whose main scientific goals lie outside the solar system is a common practice in asteroid science. Several groups have used various image archives to find and characterize SSO.”
In 2019, for example, researchers used archival images from exoplanet surveys to identify more than 1,800 asteroids, with 182 potentially open.
Astronomers want to get a complete picture of the solar system’s asteroid population, as it helps clarify the history of the solar system.
“The detailed description of small bodies in the solar system imposes constraints on various scenarios for the formation of the solar system, which allow specific predictions about the size and distribution of object orbits over time,” the authors explain.
“In particular, migrations of giant planets and collision cascades affect the size and orbital distribution of asteroids, which can be detected with specially organized observations.”
But specially organized observations are expensive and time consuming. In addition, observational proposals face stiff competition from other researchers with different goals.
“Instead, we decided to conduct such a study based on a large archival dataset,” the authors write.
Crook hinted that there are other objects in the data besides asteroids. “But there were other random finds in the archival images that we are now tracking.”
He also said that their approach is revolutionary and they intend to use it again.
“Using this combination of human and artificial intelligence to analyze huge amounts of data is a big boon, and we will use these methods in other upcoming studies, for example, with the Euclid telescope.”
As for the “…other unexpected finds…” in the images, Crook declined to say what those findings might be. He told Universe Today that the finds “… are not related to unusual asteroids, but to other finds in the data. We will report them in subsequent publications and announcements soon, so stay tuned!”
Contact us: [email protected]