Independent reconstruction recovers the ring around M87*
An independent, peer-reviewed analysis recovered the ring around the black hole M87* from existing Event Horizon Telescope observations. A method using measurements resistant to telescope calibration errors reproduced the broad shape across datasets from 2017, 2018 and 2021. The result supports the ring’s reality while finer features, including its width and central brightness, remain sensitive to reconstruction choices.
Science··Night
Existing telescope data yield the ring again
An independent analysis has recovered the ring surrounding M87*, the supermassive black hole at the centre of the galaxy Messier 87. The peer-reviewed Nature paper reanalyses publicly available Event Horizon Telescope measurements from 2017, 2018 and 2021 rather than presenting new observations. Its broad ring shape agrees with the familiar image, despite reaching that reconstruction through a different method.[1], [2]
The telescope combines radio observatories across Earth into a virtual aperture of planetary scale. That aperture is not completely filled: large gaps remain between the measurements supplied by individual telescopes. An image must therefore be reconstructed from sparse information. Calibration and imaging choices can influence the outcome, explaining why some earlier reanalyses challenged the ring while many others recovered it.[1]
Calibration-resistant measurements constrain the reconstruction
The new approach uses closure invariants, combinations of telescope measurements that preserve information about shape while cancelling station-based calibration errors. A generative model then produces candidate reconstructions from those observables. Its training includes non-astronomical images. A further test excluded geometric shapes from training and still recovered the ring, probing the influence of simple shape preferences in the training material. Avoiding a manually prescribed ring does not remove all the statistical assumptions learned by the model.[1]
The team compared several fields of view and many individual reconstructions. The ring persisted most clearly in the best observing datasets, whereas poorer coverage left greater ambiguity. Ring width, central brightness and surrounding diffuse emission were more sensitive to reconstruction choices. These finer details cannot all be read as fixed intrinsic properties of the black hole’s surroundings merely because the overall ring is robust.[1]
Angular diameter supports a conditional mass estimate
The measured angular diameter is 42.2 ± 1.5 microarcseconds. With assumptions about distance and physical scaling, the analysis infers a mass of 6.5 ± 0.5 billion solar masses. The angular measurement and the mass inference are different steps: converting apparent size into mass requires those additional inputs. The analysis does not constrain the magnitude of black hole spin. Public data and code retain a route for revisiting the comparison of methods and the smaller, less stable image features.[1]
Related columns
For more information on this topic, you can read the related columns.
