When there’s an object lying at the center of our own galaxy that could give us insights about the mysteries of gravity and evolution of the universe it helps if we can actually see it.
On Thursday, astronomers unveiled the first image of the supermassive black hole that lies at the center of the Milky Way galaxy – known as Sagittarius A* (Sgr A*). This wasn’t just any simple snap of a camera though. Black holes have long been thought to be basically ‘unseeable’. This is only the second-ever image produced of one and took a huge collaborative effort of astronomers and data scientists from around the world working together as part of the Event Horizon Telescope (EHT) project. Black holes are not only very far away but are also very large so special telescope technology and a great deal of data is required to produce an image.
For this latest endeavor, the EHT team is building on the data and methodology of their first big foray into this space. In 2019 they captured the first-ever image of a black hole, one that sits in a distant galaxy known as Messier 87 (M87*). The difference this time was that even though Sgr A* is much closer to us, it is also a thousand times smaller. This means the gases and materials flowing around the black hole which help with pinpointing and generating images of it are moving at a very high speed and so it was more challenging to get a clear image when everything was changing around it rapidly. That means image processing techniques and data analysis tools using supercomputers were going to need to be refined and improved to make this work.
Getting from process to picture is more than just aiming an apparatus at the sky but involves a sequence of steps with many pieces working together. The Event Horizon Telescope isn’t just one telescope but a network of telescopes around the globe that combine to make a virtual planet-sized dish. Observation data is collected at high speed and written to hard drives at a rate of 4 gigabytes per second. That’s a lot of data in hard drives that must be correlated together and analyzed. To do this the drives to centralized locations where supercomputers and scientists await is necessary. This cannot be done over the internet with so much data the transfers would take forever. “You can’t beat the ‘internet’ of a 747 filled with hard drives” said Shep Doeleman, founding director of the EHT, at live Harvard Center for Astrophysics panel discussion about the results. Transporting the drives by air is the most feasible option. Once the data is amassed at supercomputing locations it can be correlated and combined to produce an image of the highest resolution possible.
How much data must you compute to get an image of an object that’s 27,000 light-years away, is 4 millions times more massive than our Sun, from a network of telescopes that make a dish the size of the Earth?
The data collection actually took place over several nights back in 2017 at the various EHT locations. Then the heavy pile of hard drives were flown to the correlator sites. There data was arranged into manageable disk bays holding a capacity of 80 terabytes each (8 × 10 TB disks). All in all it’s estimated that 3.5 PB (that’s petabytes!) was collected, managed and fed into supercomputers! The supercomputers’ job (and researchers working alongside) is to clean up the data, removing noise generated by the atmosphere and instruments so all that is left is a large number of captures that can be fit together to make the final image. Helping the EHT team this time were all the insights gained from the trailblazing M87 run including being able to utilize more of the telescope’s capacity with one extra South Pole telescope helping that couldn’t see M87 at the time. The smaller scale and much faster moving nature of Sgr A* compared to M87 meant computer models were stepped up to get a crisper look at the constantly moving cosmic body.
From light-years to petabytes to just a few megabytes-sized image – a feat of science and technology and data! Here’s the final unveiled image.
From someone who also spends a lot of time looking for things in the dark, congrats to the whole Event Horizon team on this latest work.
Category: data recovery