A new telescope will find billions of asteroids, galaxies and stars
By The Economist
On April 15, at 8pm local time, the Vera Rubin Observatory recorded its very first photons of starlight. At first, the images that filled the screens in the control room on Cerro Pachón, 2500m high on the foothills of the Andes in northern Chile, looked like a field of snowy static on an old television. But, zoomed in, the spots soon resolved into an uncountable number of stars and galaxies floating between enormous, wispy clouds of dust, like tiny multicoloured flecks of paint spattered across a vast black wall.
“There was this huge amount of cheering and screaming, people were getting teary-eyed,” recalls Alysha Shugart, a physicist who watched the events unfold on the night. “Those little photons had no idea of the red carpet that was rolled out for their reception.”
This image, comprised of 678 separate images taken by NSF–DOE Vera C. Rubin Observatory in just over seven hours of observing time, shows the Trifid nebula (top) and Lagoon nebula, both of which are thousands of light years away from Earth.Credit: NSF-DOE Vera C. Rubin Observatory
The arrival of those photons – many from ancient stars and galaxies and which had been travelling across the universe for billions of years – marked a neat moment of symmetry. It had been exactly ten years since work had started on Cerro Pachón to build the observatory; it also marked the start of a ten-year project – the legacy survey of space and time (LSST) – that will see the Rubin telescope repeatedly take ultra-high-resolution pictures of the entire night sky of the Southern Hemisphere every three or four days. Rubin will see more detail about the cosmos, and unlock more of its unknowns, than any machine that has come before. It will collect so much information – trillions of data points on more than 40 billion new stars, galaxies and other cosmic objects – so quickly that it will transform astronomy in its wake.
In its first year alone, it will double the amount of data collected so far by every other instrument in the history of optical astronomy. It will collect 20 terabytes of raw image data every night and, over the course of the LSST, will produce more than 500 petabytes of images and analysis. For the first time astronomers will also have a decade-long time-lapse of the night sky.
That last part is what has scientists most expectant. Astronomical observatories until now have focused on taking detailed snapshots of tiny points in the night sky. But “the sky and the world aren’t static,” says Yusra Al-Sayyad, a researcher at Princeton University who oversees Rubin’s image-processing algorithms. “There are asteroids zipping by, supernovae exploding.” Many of those fast or transient objects can only be seen by big observatories if they happen to be pointed in exactly the right direction at exactly the right time.
“Today we don’t really have a very full, wide and deep picture of the universe,” says Leanne Guy, a physicist at Rubin.
Rubin will fix that gap. Its 1.7m-long, 3200-megapixel camera – the biggest digital camera ever built – has an enormous field of view, equivalent to an area of sky covered by 45 full Moons.
The camera will be fed starlight reflected off a primary mirror that is 8.4m wide and which took scientists at the University of Arizona seven years to grind into its unique shape. Despite their size, the mirrors, telescope and the giant silver dome that houses it can all move together extremely fast. The telescope will be able to take an image every 30 seconds and its “brain” – a piece of software known as the scheduler – will use machine-learning algorithms to automatically work out the best places to point the camera, every night, as it attempts to cover as much of the sky as possible while also avoiding obstructions, such as clouds or satellites streaking overhead. Over the course of a decade, each point in the sky will be photographed around 800 times.
In an image released this week by the Rubin team, for example, stitching together ten hours of observations, astronomers identified more than 2000 asteroids in the solar system that had never been seen before (including seven near-Earth asteroids). For comparison, around 20,000 asteroids are discovered in total every year by all other ground and space-based observatories. During the LSST, Rubin will conduct the most detailed census yet of millions of as-yet-unknown objects in the solar system, including tripling the number of known objects that could come near to the Earth and finding around 70 per cent of asteroids classed as “potentially hazardous”, ie, bigger than 140m wide. If, as some scientists reckon, there is a ninth planet hidden in the clouds of rocks somewhere far beyond Neptune, Rubin will find it.
Celestial surveillance
The census-taking will stretch far beyond the solar system. Because the LSST camera will keep coming back to the same point in the sky many times during its decade-long survey, astronomers will be able to combine many images of the same location. The fainter an object, the farther away and older it is likely to be and, therefore, hundreds of stacked images will eventually reveal the very earliest stars and galaxies.
By recording details – such as the colours, shapes, positions and movements – of more than 17 billion stars and 20 billion galaxies, Rubin is expected to produce a catalogue of the night sky that cosmologists can then use to build their most detailed picture yet of the early universe and examine how it has evolved over time. That will be crucial for two of the prime goals of the Rubin observatory – understanding the nature of dark matter and of dark energy.
It is this dark universe for which Rubin was first conceived in the late 1990s. The observatory’s namesake, Vera Rubin, was an American astronomer who, in the 1970s, made her name by measuring that the stars at the edge of the nearby Andromeda galaxy were moving just as fast as those at the centre, impossible if only normal matter was present. Her discovery provided evidence of the existence of “dark” matter, which cannot be seen and interacts with normal matter only through gravity.
Two decades later, scientists discovered an even bigger hole in the universe – a mysterious substance was found to be accelerating the rate at which space was expanding. Dark energy, it turned out, made up 68 per cent of the mass in the universe and dark matter made up around 27 per cent. Only around five per cent comes from the familiar “normal” matter that makes up stars, planets, dust and everything on Earth.
Understanding how the invisible dark universe behaves depends on better observations of the visible one. One of the ways in which Rubin’s LSST will help is by measuring how the light from very distant galaxies is distorted by the gravitational force of the matter between them and Earth. These measurements will give astronomers details about how matter is arrayed in the universe and also how it is moving. Both are important clues to the nature of the dark universe
The study of dark energy, in particular, will get a boost. The phenomenon was discovered in the 1990s when scientists were studying the movements of the handful of supernovae known to exist in the Milky Way at the time. Rubin will, according to the scientists working there, be a “supernova factory”, potentially discovering billions more of these exploding stars, providing cosmologists with a vastly bigger data set to study more deeply and precisely, and with much better statistics, the way that dark energy behaves.
Pioneering astronomer Vera Rubin, who helped find powerful evidence of dark matter, died in 2016 aged 88. She is pictured here in the 1970s.Credit: AP
Rubin’s data will not stay on the mountaintop in Chile. Less than ten seconds after the LSST camera’s shutters close every day, everything will be transferred, through dedicated optical fibres, to computers at the SLAC National Accelerator Laboratory in California (backups will go to data centres in France and Britain). At SLAC, an automated process will first clean up the images and carry out an initial analysis that will look for objects that have, say, appeared for the first time or significantly changed position or brightness since the previous night. These changes – there will probably be millions per night – will be quickly winnowed down (by more specialised algorithms) into a priority list and passed on to other observatories around the world who can then follow up with more detailed direct measurements of their own. All of this will happen autonomously.
“There’s absolutely no way any human being could go through these alerts by eye,” says Dr Guy. “There’s no way.”
The LSST is scheduled to begin at Rubin in October. In the meantime, the instruments sitting on Cerro Pachón will continue to be tested, re-tested and calibrated. Though Rubin’s primary mission is set for now, the scientists who have built the observatory know that what they ultimately have at their disposal is a discovery machine. “What I’m most excited about seeing from Rubin in the long term,” says Dr Guy, “are the things we’ve never even thought about.”