But these conversations did not focus on light pollution, a problem presented by the reflective surfaces of proposed satellite constellations such as Starlink. At first, SpaceX said the complication would be minimal, and the new committee is trying to assess the impact and actively find solutions.
“So far, they’ve been quite open and generous with their data,” Dr. Lowenthal said. “But they have not made any promises.”
A spokeswoman from SpaceX said the company was taking steps to paint the Earth-facing bases of the satellites black to reduce their reflectiveness. But Anthony Tyson, an astronomer at the University of California, Davis, said that wouldn’t solve the problem.
Dr. Tyson is the chief scientist for the Large Synoptic Survey Telescope — a 27-foot, billion-dollar telescope under construction in Chile that will scan the entire sky every three days. The survey, the world’s largest yet, will help astronomers better understand dark energy, dark matter, the origin of the Milky Way and the outer regions of the solar system. But because it is designed to scan faint objects, it is expected to be greatly affected by the satellites.
Dr. Tyson’s simulations showed that the telescope would pick up Starlink-like objects even if they were darkened. And they wouldn’t just affect a single pixel in a photograph. When there is a single bright object in the image, it can create fainter artifacts as well because of internal reflections within the telescope’s detector. Moreover, whenever a satellite photobombs a long-exposure image, it causes a bright streak of light that can cross directly in front of an object astronomers wish to observe.
“It’s really a mess,” Dr. Tyson said.
Knowing how challenging it would be to correct these interrupted images, Dr. Tyson decided the best step forward was to set the telescope to avoid Starlink satellites. While simulations based on the earlier 12,000-satellite total suggested that would be possible, SpaceX’s application for 30,000 additional satellites upset the calculations.
“We’re redoing the models now just to see what’s visible at any one time — and it’s really quite frightening,” said Patrick Seitzer, a professor of astronomy emeritus at the University of Michigan, who has been running similar analyses to determine how many satellites will be visible and when.