AN ASTEROID TRAIL ACROSS THE GORGEOUS CRAB NEBULA. IMAGE CREDIT: ESA/HUBBLE, M. THÉVENOT (@ASTROMELINA)
The Hubble Asteroid Hunter project is not even three years old and it has already delivered a treasure trove of new findings. A group of citizen scientists, with the help of a machine learning algorithm, has discovered 1,031 trails in Hubble data belonging to unidentified asteroids.
In one fell swoop, the number of known space rocks has increased by about 1 percent. A fantastic result for this project, which is part of the citizen science collaboration Zooniverse. The users of this project went through 37,000 images of Hubble to identify trails.
Trails can be seen as one or multiple asteroids photobombing whatever Hubble is actually observing. A typical observation from the veteran space telescope takes about 30 minutes, and while it is observing nearby planets or distant stars, an asteroid might cross its line of sight, catching some sunlight and creating a little curved trail.
In a paper, published earlier this year in Astronomy & Astrophysics, the research team explain that they had the citizen scientists label Hubble images, find trails, etc. And those images were used to train a machine-learning algorithm to go through the archives of the space telescope and hunt for asteroids. They found 1,701 and of those 670 belonged to objects already present in the International Astronomical Union’s Minor Planet Centre database.
“Citizen science and machine learning are very useful techniques for the systematic search for Solar System objects in existing astronomy science data archives,” the research team wrote in the paper. “This work describes a method for finding new asteroids in astronomical archives that span decades; it could be effectively applied to other datasets, increasing the overall sample of well-characterised small bodies in the Solar System and refining their ephemerides.”
Ephemerides – an excellent 19-point scrabble word – are the tables that list the position in time of astronomical objects. Given that these objects were all observed in older data, it is difficult to estimate where they are and where they are orbiting. But there’s a possible fix to that.
Hubble is not fixed in space but rotates around the planet. This information, together with the shape of the asteroid trail, provides an important astronomical number the parallax, which can be used to estimate the distance of this object and thus its orbit. For longer observations, the team might even be able to estimate the rotational period and shape of the asteroids.
The work shows, once again, the power of the Hubble Space Telescope. The team estimates that the newly discovered objects are on average four times dimmer than the known asteroids in the catalog.
No comments: