Discussion
Currently watching an IAU-CPS talk given by Olivier Hainaut.
His simulations show that 30,000 sats at 350-600km causes the same astronomy research data loss as 6,000 sats at 1200 km. This is because higher altitudes are visible for a larger fraction of the night, and the sats move more slowly across the image so they cause brighter streaks even though they are farther away
@sundogplanets I'd be very curious (and sad) to see how "dark sky" maps change based on those light pollution increases. Does everyone lose The Milky Way?
Reflect orbital passes this threshold with basically any more than one satellite. SpaceX's million sats cross this threshold just with scattered light, even ignoring the streaks they will cause. Scattered light starts to get above the 3% loss level with tens of thousands of sats. We're quickly getting there :(
Points out that cross-talk in Vera Rubin Observatory makes it so you get multiple trails for sats brighter than mag 7. So Rubin gets to 3% loss much faster than other observatories.
@sundogplanets The sad truth is that it’s these facts are not rocket science.
Also talks about the point I've made repeatedly: for planetary defense (tracking potentially hazardous asteroids), twilight is an extremely important time to observe, and this is perhaps the most valuable observing time, and when satellites are brightest and data loss is highest. How do you put a price on a very low probability but extremely high mortality event (an asteroid impact)? We have the tech to avoid this (thanks, DART!) but need good orbits as early as possible.
So, what Olivier is starting to try to do is put a price on the data loss. This is smart! But hard. He's got the first steps (quantifying potential losses and how much telescope time costs for ESO) very well laid out in his most recent paper here: https://arxiv.org/pdf/2604.09427
I guess when the people you're arguing with only care about money, this is an extremely smart tactic. I hate that (how do you quantify awe and scientific discoveries?) but I totally understand why he's doing this. Good work!
Alejandro Borlaff has entered the discussion, to remind everyone that the satellite streaks are also visible in space telescopes, and not only that, each lost pixel costs WAY more money than for ground-based telescopes. https://iopscience.iop.org/article/10.3847/1538-3881/ae71ce/pdf
So the techbros "just put your telescopes in space!" is totally ridiculous and unhelpful.
@sundogplanets presumably if they're getting good enough to launch that many satellites, putting telescopes in space, including higher orbits, will also be easier/cheaper.
@tewalds Space telescopes are generally 10x more expensive for a 10x smaller mirror with the technology frozen at about 10 years old by the time it launches, and you can't fix it or change anything. This is not a solution.
@sundogplanets That is how it's been, but will that remain true once launch costs drop dramatically (which we must assume if we're also assuming space datacenters), and rockets get big enough to launch big (but not enormous telescopes)?
@tewalds @sundogplanets We're not getting space datacentres, though. The idea's utterly impractical. Hence, this is a bit like suggesting the space telescopes be put into orbit by friendly unicorns.
@denisbloodnok @sundogplanets Agreed, it's very unlikely that we'll get space datacenters for a whole bunch of reasons. I'm just saying that if we do get even small space datacenters, we'll probably also get cheaper space telescopes.
Install bonfire.cafe
Get the full app experience
Light pollution from 5000 Reflect Orbital satellites, the scattered light from those will increase light pollution worldwide 20-30%. 50,000 would increase light pollution by 200-300%. That sucks.
Points out that most research telescopes lose about 3% of time to technical problems. Starting with this comparison point, he says that when research telescopes start losing 3% of their data to satellites, that's when the loss is unacceptable.