AnswersPublished20 sources
How CSIRO Independently Reconstructed M87*’s Black Hole Ring
CSIRO researchers used GenDIReCT and triangular closure invariants to reconstruct M87 from Event Horizon Telescope data, recovering a ring like structure 42.2 ± 1.5 microarcseconds across. The study estimates a black hole mass of 6.5 ± 0.5 billion solar masses under its stated scaling assumptions; the result strengt...
Published byEdited with GPT-6 LunaImages generated with GPT Image 2
Research answer
Key takeaways
- CSIRO researchers used GenDIReCT and triangular closure invariants to reconstruct M87 from Event Horizon Telescope data, recovering a ring like structure 42.2 ± 1.5 microarcseconds across.
- The study estimates a black hole mass of 6.5 ± 0.5 billion solar masses under its stated scaling assumptions; the result strengthens the case for the ring but is not, by itself, a standalone test of general relativity...
- Rather than combining telescope measurements in pairs, the method uses calibration resistant combinations from telescope triangles and a generative machine learning imaging pipeline.[3][4][10]
AI Prompt
openai.comCreate a landscape editorial hero image for this Studio Global article: How did CSIRO researchers Samuel Lai, Nithyanandan Thyagarajan, O. Ivy Wong and Foivos Diakogiannis independently reprocess the original Eve. Article summary: The CSIRO team independently reconstructed M87* using GenDIReCT, a generative machine-learning method driven by triangular interferometric “closure invariants,” and recovered a ring-shaped shadow with a diameter of 42.2 . Topic tags: general, academic, general web, education. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts wi
A CSIRO team has independently reconstructed the ring-like structure around M87* using the Event Horizon Telescope’s data and a different imaging approach. Published in Nature on November 11, 2026, the study reports a diameter of 42.2 ± 1.5 microarcseconds—consistent with a ring-like morphology in the original landmark image.1
8
18
The analysis matters because it changes how the measurements are combined and how the image is reconstructed. It is an independent check of the result using the same telescope data, rather than a new observation of M87*.3
8
Why reconstructing the image is difficult
The Event Horizon Telescope is a very-long-baseline interferometry network. Its measurements sample an astronomical source indirectly, and the network’s sparse aperture coverage makes turning those measurements into an image a challenging reconstruction problem.18 Telescope and atmospheric effects can also complicate the data and the calibration needed to interpret it.
10
That means an image is not simply a photograph assembled from pixels. Researchers must infer the source’s structure from the measurements, using a reconstruction method that fits the data.
The key change: measurements from telescope triangles
Many interferometric analyses combine measurements from pairs of telescopes. Lai, Thyagarajan, Wong and Diakogiannis instead used closure invariants built from triangular combinations of telescope measurements.3
10
These quantities preserve information about the source’s morphology while being intrinsically insensitive to station-based calibration errors. In practical terms, they give the reconstruction a way to use the data while reducing its dependence on some of the calibration effects that can complicate very-long-baseline observations.3
18
How GenDIReCT turns those measurements into an image
The team used GenDIReCT, a generative deep-learning method designed to reconstruct images directly from observed closure invariants.4 The method addresses the inverse problem: finding an image that is consistent with the measurements, even though those measurements do not directly provide a complete picture.
2
4
An account of the team’s approach says the model was trained on terrestrial images such as cats, dogs and cars, rather than black-hole images.36 That detail describes the model’s training examples; it does not mean the reconstruction is free of every modeling choice or assumption.
What the reconstruction found
The Nature paper reports that the method reproduced a ring-like morphology with an angular diameter of 42.2 ± 1.5 microarcseconds. Under the paper’s stated nominal scaling parameters, that diameter corresponds to a black-hole mass of 6.5 ± 0.5 billion solar masses. The paper also reports an inclination of about 20 degrees from the approaching jet.1
The researchers’ analysis also describes measurements of the ring’s width, orientation, azimuthal brightness asymmetry and central brightness fraction.31 These features help characterize the reconstructed structure beyond its overall diameter.
What the result does—and doesn’t—show
Recovering the ring-like structure with a different set of observables and a different reconstruction method strengthens the independent validation of the EHT result.3
8 It helps address the concern that the ring might be an artifact of one particular imaging approach.
17
But this is still a reanalysis of the EHT data, not a second observation, and the result should not be described as proof of general relativity on its own. The supported conclusion is narrower: an independent analysis recovered a ring-like structure at the reported scale, adding evidence that the original image captured a real feature of M87*.1
3
8
Studio Global AI
Continue your research
This page includes a source-backed answer you can continue inside Studio Global.
People also ask
What is the short answer to "How CSIRO Independently Reconstructed M87*’s Black Hole Ring"?
CSIRO researchers used GenDIReCT and triangular closure invariants to reconstruct M87 from Event Horizon Telescope data, recovering a ring like structure 42.2 ± 1.5 microarcseconds across.
What are the key points to validate first?
CSIRO researchers used GenDIReCT and triangular closure invariants to reconstruct M87 from Event Horizon Telescope data, recovering a ring like structure 42.2 ± 1.5 microarcseconds across. The study estimates a black hole mass of 6.5 ± 0.5 billion solar masses under its stated scaling assumptions; the result strengthens the case for the ring but is not, by itself, a standalone test of general relativity...
What should I do next in practice?
Rather than combining telescope measurements in pairs, the method uses calibration resistant combinations from telescope triangles and a generative machine learning imaging pipeline.[3][4][10]