SpaceX's recent launch of 60 low-earth orbit satellites is just the start of a program to deploy thousands of relatively small, smart, and potentially disruptive devices for supplying broadband internet to many of the 3.8 billion humans currently lacking that utility. But astronomers have been quick to express concern about what these endlessly moving, sometimes bright constellations will do for anyone hoping to study or simply experience the night sky and the cosmos beyond. 

What exactly is going on?

Several commercial entities are aiming to fill a huge gap in human data access (affecting about half our species' population) by launching a new generation of orbital broadband satellites. SpaceX is perhaps the most ambitious, with a final goal of some 12,000 'Starlink' satellites in different orbital 'shells', providing uninterrupted, high speed data to pretty much anywhere on the planet. But Amazon is not too far behind with its Project Kuiper and some 3,000 satellites, and others like OneWeb and what might eventually be 650 satellites.

Why so many satellites?

A few reasons are cost, coverage, and latency. In principle you could put up more geostationary satellites at 42,000 kilometer altitudes but these are hugely expensive, require highly directional, line-of-sight ground receivers and more powerful transmitters, and because of the finite speed of light there is an unavoidable time delay to each and every digital transaction. Alternatively, lots of low-orbit data-linked satellites can provide a dynamic 'dome-like' presence in the sky, more cheaply, and with much less time delay when some of them are not much more than a few hundred miles away from you at any moment.

What's bad for astronomy?

Short of being painted in Vantablack (which is not necessarily possible because of thermal regulation challenges in spacecraft) any artificial object will at times reflect sunlight down to the ground. This means that during portions of (early and late) night on Earth, overhead satellites hundreds of miles up can still be in daylight and appear as brilliant points of light streaking across the sky. For astronomers trying to capture deep exposure images or spectra of faint cosmic phenomena a bright satellite trail in the field of view can basically wreck the data. Other projects like the LSST plan on exploring the time domain of the cosmos - repeatedly imaging the sky night after night. Glinting, moving satellites get in the way of that. And for radio astronomy, which can seek to explore the universe in microwave bands, any satellite communications (often in similar if not overlapping frequencies) are yet another source of disastrous noise in an already noisy technosphere.

OK, but astronomers already deal with this right?

Yes they do. But what's taken people by surprise is the sudden realization that a project like SpaceX's Starlink will, once complete, put so many satellites into orbit that the odds of finding a 'pristine' piece of sky are drastically reduced - and that will hold for the foreseeable future. In other words the night sky will not be the same again. The same is true for casual astronomical observers or anyone who looks up and wants to experience the cosmos in the way our ancestors did.

But the companies got permission to do this, yes?

Well, yes, SpaceX and others in the US have been getting FCC approval and International Telecommunications Union (ITU) approval on matters like numbers and orbits - albeit pretty recently. One can only presume that someone at the FCC thought about impacts on science, and on radio interference (the Starlink satellites use two common microwave bands called Ku and Ka in the tens of gigahertz range). But there's not much evidence for that - unless it's buried in the official documents. And that's frustrating because in the US as in many countries, it's federal money that helps keep science running. So if, for example, one federal agency funds a big new ground-based telescope but another federal agency then allows activities that will reduce the efficiency or capability of that telescope it seems like there is something broken in the system.

Will it really be as bad as it sounds?

To be honest it's a bit early to say for sure. After the first 60 Starlink satellites were launched people started sharing images of the bright train of satellites moving through the night sky. It looks super cool, but is also the stuff of astronomers' nightmares. However, this is not the final orbital configuration. The satellites will spread themselves out into a planet-encircling ring with much greater spacing. Also, it seems that once their single solar panels are properly deployed the satellites are likely to present much less reflective surface to the ground - reducing their apparent brightness. On the other hand, an object doesn't need to be very bright to mess with astronomical data looking at phenomena that might be billions of times fainter. And when there are 12,000-plus of these satellites specifically designed to cover anyone's sky at any time the odds of problematic issues goes up.

Again, can't astronomers just work around this?

Probably, up to a point. Astronomers already have to deal with about 5,000 other satellites currently orbiting the Earth. If the satellite operators share positional data that could help in planning observations and sifting out natural phenomena from artificial ones. But it's not clear that data will be shared, or that the satellite self-positioning behaviors (to avoid collisions and maintain maximum telemetry integrity) will allow for precise information on where they will be at any given instant. In all cases it could mean a lot more work by the scientists, and that means more time, more money, and less real research taking place.

But what about the people?

In the most generous interpretation of commercial motivations it could be said that providing proper internet for all people is about making lives better. But of course we also know that connection and data bring problems - from social manipulation and misinformation to drastic changes in how people make a living, or don't. And of course money has to be made too. SpaceX has been pretty open about their hopes for Starlink as providing a source of revenue that can feed back into other, even grander plans, like getting to Mars. Along the way it's also many a scientist's dream that cheaper and larger space launch systems might usher in a new era of space-based astronomy, with giant telescopes far away from the maddening crowds of Earth's night skies. But there is seldom anything guaranteed in big science, and we've been waiting for decades for some dreams like this to become reality.

What should happen next?

That is indeed the multi-billion dollar question. What would be really great is if scientific needs get taken into account more than they seem to have been. It's quite hard to imagine a similarly impactful yet overlooked consequence arising in other areas. For example, what if a company unleashed a set of (entirely hypothetical) chemical compounds into the environment that complicated our ability to obtain genetic sequences of organisms - from humans to bacteria. There would be outrage, if not terror. Of course it's hard to argue that a disruption of our night skies (which have been swamped by human light sources for many of us anyway) has quite the same immediacy. Nonetheless, do we really want to live on a planet that no longer offers the same awe-inspiring cosmic view to us all? Glinting satellites can certainly produce their own feeling of awe, but are we really that species, forever obsessed with its own reflection? It seems that there should be a way to solve this, and the first step would be to try to ensure our tax dollars (or whatever currency) be put to proper use by calling for a better global plan to ensure that space-based technological advances minimize their damage to science.