Re: Titre vs contenu de l'article

et j'ai du mal à imaginer que ces limitations ne puissent pas être contournées par des moyens algorithmiques

Cet article explique très bien pourquoi c’est pas si simple :

The situation is a bit different for more sensitive telescopes, however, such as the Vera C. Rubin Observatory’s 8.4-meter Simonyi Survey Telescope, currently under construction in Chile. Its 3200-megapixel camera will capture about 1,000 images of the sky in 9.6-square-degree chunks — every night for 10 years. Each image can detect objects to a limiting magnitude of about 24.

It is now expected that up to 30 percent of its images will contain at least one Gen-2 satellite track. Additionally, a single bright satellite with a magnitude of 5 would not only corrupt the pixels directly under its path across the image, but also produce a 100-pixel-wide (or greater) parallel band of saturated pixels to either side. And every 30-second twilight exposure will have at least one streak across it in the Gen-2 configuration, so that up to 3 percent of the image area will have to be discarded even if the satellites are dimmed to magnitude 7, equivalent to losing several months of observing time.

D’autre part :

et au pire les orbites des satellites sont très bien documentées, ce qui permettrait de savoir à la seconde près quand le satellite va entrer dans le champ du télescope

C’est beaucoup plus compliqué que ça, surtout vue la quantité de satellites déjà en orbite et à venir à prendre en compte.

Sinon je te conseille aussi le rapport de l’American Astronomical Society (ASS) sur le sujet ou, plus technique, _Impact of the SpaceX Starlink Satellites on the Zwicky Transient Facility Survey Observations.

La connaissance libre : https://zestedesavoir.com