Impacts of Large Satellite Constellations on Astronomy

[Robotbeat]

If any of you remember the actual recent history of this whole topic, the astronomy community was dismissive of Starlink until it launched. They didn’t take it seriously because presumably they thought it was a pie in the sky NewSpace venture that would fail like the others. Which wasn’t totally an unreasonable expectation, as that’s what typically happened!

Starship (and Starlink) do impact ground astronomy (although mostly the survey scopes, not the high resolution narrow field of view Keck or Gemini type telescopes many of you are talking about). But it is ALSO an unprecedented opportunity. It’s a double-edged sword.

To deny the potential of Starship to enable much higher performance ( and potentially much lower cost) space telescopes of a size comparable to some of the larger ground telescopes while fearmongering about the potential of Starship is saying it’s a single edged sword, denying work by folks like LUVOIR, etc.

But we’re going to have to do better than JWST, cost-wise. Just like if these megaconstellations cost as much per satellite or per unit dry mass (ignoring launch costs) as conventional telecommunications satellites, they never would have happened, we can’t expect to get cost performance comparable to ground telescopes without a lot of work to get the cost down by leveraging the new capability. It’s not automatic. But it IS a massive opportunity.

[eeergo]

Oh please, I’m not asking for you to become a fanboi. I’m just asking for some consistency! Either be hyper-skeptical of Starship (and the idea you can inexpensively make lots of high quality aerospace hardware to fill its payload bay) or don’t. You can’t have it both ways, so make up your mind!

I think it’s perfectly rational for astronomers to be skeptical of Starship launching big cheap telescopes. But one must also be skeptical that SpaceX can launch Starship cheaply enough, with cheap enough spacecraft, to fill out Starlink v2 as they claim.

Please: Make. Up. Your. Minds.

(Astronomy is awesome, and quit putting slanderous words in my mouth. I love astronomy.)

Wow, pretty worked up today? Leave the slander in the other thread where you're using it. Nobody is quoting you as saying that literally.

I can be skeptical of, I don't know, hydrogen-powered trains, and still state that they could be a fire hazard were they to take over railway traffic. Should go without saying. It's a textbook false dychotomy. Certainly didn't need a 6-paragraph, 3-nestle-quote breathless, aggressive comment.

[DanClemmensen]

For a given set of orbits, the number of satellites needed to provide continuous ground coverage is fixed, and the only reason to use more satellites is if each satellite cannot provide sufficient bandwidth. Thus, after you reach this required number of satellites, you can either use more satellites, or you can use satellites with more bandwidth. Early constellations (OneWeb, Starlink V1.x) use the smallest satellites that can provide coverage because the $/kg is very high. To double the marketable bandwidth in high-density areas, these constellations would need to double the number of satellites, and persumably this doubles the telescope pollution. The alternative is to use higher-bandwidth satellites, but these are heavier and require bigger launchers. Starlink V2.2 will launch on Starship, and each satellite will provide roughly ten times the effective bandwidth.

The big problem for OneWeb and Starlink is that it's all-or-nothing. You do not get a new increment in marketable bandwidth for each satellite. You must launch all of the new satellites before you have continuous increased bandwidth everywhere. There are nuances involving latitude and there are some tricks you can play, but in the main, you need the whole new set before it makes much new money.

What this means for telescope pollution is that we could choose to limit the total number of satellites without limiting increases in available bandwidth to users. This is enabled by lower $/kg to orbit provided by Starship-class LVs.

Further improvements can be made by consolidating functions. We do not need separate little constellations  for earth observation, SAR, fire detection,  navigation, etc., and we sure as heck don't need to scatter a bunch of cubesats around.  If all of these satellites are considered to be pollution, then we can regulate this pollution. Of course, this completely wrecks the entire existing space economy, but it's a viable technical solution to the technical problem  of light pollution.

[Robotbeat]

There are multiple reasons to use more numerous satellites.

[Lee Jay]

Will they fund and plan a space telescope that relies on Starship before Starship has demonstrated it exists, is reliable, and is affordable by launching a bunch of V2 satellites? Of course not.

I'm not sure why anyone would plan a great observatory around Starship anyway. We just saw that designing and deploying one can take 20 years.  Will Starship be around for 20 years?  Falcon 9 is half that age and looks like it's about to be replaced.  What if you design an observatory that can only be launched on Starship and half way through the development cycle Starship gets retired in favor of a smaller, cheaper, more efficient version that's way better for everything except it can't launch your observatory?  JWST could have been moved to another launcher had that happened to it.

[Robotbeat]

… The big problem for OneWeb and Starlink is that it's all-or-nothing. You do not get a new increment in marketable bandwidth for each satellite. You must launch all of the new satellites before you have continuous increased bandwidth everywhere. …

This is demonstrably not true. Starlink has had continuous service for a while now and the first constellation is not at all complete, yet.

[Robotbeat]

Will they fund and plan a space telescope that relies on Starship before Starship has demonstrated it exists, is reliable, and is affordable by launching a bunch of V2 satellites? Of course not.

I'm not sure why anyone would plan a great observatory around Starship anyway. We just saw that designing and deploying one can take 20 years.  Will Starship be around for 20 years?  Falcon 9 is half that age and looks like it's about to be replaced.  What if you design an observatory that can only be launched on Starship and half way through the development cycle Starship gets retired in favor of a smaller, cheaper, more efficient version that's way better for everything except it can't launch your observatory?  JWST could have been moved to another launcher had that happened to it.

And if you designed telecommunications satellites the same inefficient way (“that’s just how much it costs”), megaconstellations wouldn’t be possible either.

If you’re saying we cannot afford to do massive new space telescopes that take advantage of Starship if they cost as much as JWST, you’ll find no argument from me! JWST (as wonderful as its capability is) is almost a crime when it comes to cost growth compared to the very initial cost estimates, even by traditional defense contractor standards. We MUST do better. Orders of magnitude better.

[DanClemmensen]

… The big problem for OneWeb and Starlink is that it's all-or-nothing. You do not get a new increment in marketable bandwidth for each satellite. You must launch all of the new satellites before you have continuous increased bandwidth everywhere. …

This is demonstrably not true. Starlink has had continuous service for a while now and the first constellation is not at all complete, yet.

Starlink does not have continuous coverage globally yet, but I take your point: I failed to specify what I meant by "all". I intended for "all" to apply to an "increment", not to the "constellation". Exactly what constitutes an "increment" will vary depending on the geometry of the constellation, but is certainly more than one satellite or even one plane's worth of satellites.

[DanClemmensen]

There are multiple reasons to use more numerous satellites.

Yep, and every one of those reasons will create additional light telescope pollution and space junk, and the dramatic reduction in $/kg will stimulate more satellite launches. It's not just about Starlink.

From the standpoint of flexibility, more and simpler satellites are better in almost every way. From the standpoint of telescope pollution and space junk, it's better to have fewer satellites.

[Lee Jay]

Will they fund and plan a space telescope that relies on Starship before Starship has demonstrated it exists, is reliable, and is affordable by launching a bunch of V2 satellites? Of course not.

I'm not sure why anyone would plan a great observatory around Starship anyway. We just saw that designing and deploying one can take 20 years.  Will Starship be around for 20 years?  Falcon 9 is half that age and looks like it's about to be replaced.  What if you design an observatory that can only be launched on Starship and half way through the development cycle Starship gets retired in favor of a smaller, cheaper, more efficient version that's way better for everything except it can't launch your observatory?  JWST could have been moved to another launcher had that happened to it.

And if you designed telecommunications satellites the same inefficient way (“that’s just how much it costs”), megaconstellations wouldn’t be possible either.

If you’re saying we cannot afford to do massive new space telescopes that take advantage of Starship if they cost as much as JWST, you’ll find no argument from me! JWST (as wonderful as its capability is) is almost a crime when it comes to cost growth compared to the very initial cost estimates, even by traditional defense contractor standards. We MUST do better. Orders of magnitude better.

I said nothing about cost.

[Robotbeat]

Will they fund and plan a space telescope that relies on Starship before Starship has demonstrated it exists, is reliable, and is affordable by launching a bunch of V2 satellites? Of course not.

I'm not sure why anyone would plan a great observatory around Starship anyway. We just saw that designing and deploying one can take 20 years.  Will Starship be around for 20 years?  Falcon 9 is half that age and looks like it's about to be replaced.  What if you design an observatory that can only be launched on Starship and half way through the development cycle Starship gets retired in favor of a smaller, cheaper, more efficient version that's way better for everything except it can't launch your observatory?  JWST could have been moved to another launcher had that happened to it.

And if you designed telecommunications satellites the same inefficient way (“that’s just how much it costs”), megaconstellations wouldn’t be possible either.

If you’re saying we cannot afford to do massive new space telescopes that take advantage of Starship if they cost as much as JWST, you’ll find no argument from me! JWST (as wonderful as its capability is) is almost a crime when it comes to cost growth compared to the very initial cost estimates, even by traditional defense contractor standards. We MUST do better. Orders of magnitude better.

I said nothing about cost.

Acknowledged. It had schedule delays as well. Was supposed to launch in 2006.

These two things are pretty highly correlated. Double the schedule and you’ll double the cost.

Ariane 5 didn’t exist when JWST was first formulated, either, and was almost retired before JWST launched. But other vehicles of the same class existed.

Starship isn’t the only vehicle in its class, either.

But anyway, Starship or successors will either be a thing in 10-20 years or Starlink v2 sized megaconstellations won’t be.

[meekGee]

Big space telescopes are perfectly feasible if Starship works.

ENOUGH IS ENOUGH, I CAN'T TAKE THE SILINESS ANYMORE.

Pony and unicorms are perfectly feasible if Starship works.

Seriously: this thread show perfectly what's wrong with SpaceX amazing peoplem cranked to 11. The level of obtuse reasonning and magical thinking has reached new heights.

Also complacency and arrogance. 

We should strap a bunch of BFR-Starship to EELT in Chile and send it in orbit.

I'm sickened, really. Some people really have delusion of grandeurs and live in a magical world.

What happened to this forum over the last ten years ?

Bottom line: ends justify means. Precious genius Elon Musk needs Starlink to fund his Mars plan and fill his big rockets.

According to that: SCREW astronomy.

Astronomers are stupid, they keep building ground based telescopes when Musk is littering the sky with Starlink. Stupid astronomers, get away, and put your stupid telescopes in orbit, with Magic Starship ! And if you don't want to do, Elon Musk and his amazing people graciously tell you to go to hell

This.

Same magical thinking that gave tunnel vision to those that staunchly believed STS to be a "shuttle" to space for as many people and as much cargo as you could possibly want, only much earlier in the development process, thanks to unwavering faith in a persona.

Don't forget the new argument: any interference or inconvenience to megaconstellations is unacceptable, nothing to do with interference to astronomy, which is trivial -or even beneficial: just throw your inexpensive huge telescope to space and give up Earthen limitations! Anyone who says otherwise is a liar, those professionals who oppose megaconstellations' unbridled development luddites, pundits or just trolls, or just don't exist, can't tell by now.

You're making a parody of the arguments of the "other side", and then going to town punching holes and ridiculing that parody, to nobody's satisfaction but your own.

My only question is why you even bother doing it on a public forum.  You can do this entirely on your own and save some bandwidth in the process, not to mention increasing the quality of this forum.

[Lee Jay]

But anyway, Starship or successors will either be a thing in 10-20 years or Starlink v2 sized megaconstellations won’t be.

Starship may be used to launch Starlink, diddle about either the moon a bit and be replaced with a "right-sized" version with Falcon 9's capabilities.  We don't know.

And, what other rocket is available to launch something that would barely fit on Starship?  SLS?  You think that will be around in 20 years?

[su27k]

Don't forget the new argument: any interference or inconvenience to megaconstellations is unacceptable, nothing to do with interference to astronomy

You have no idea what you're talking about:

1. Megaconstellation by regulation needs to accept interference or inconvenience in many cases, for example they need to give spectrum priority to GEO satellites, they also need to protect radio astronomy bands specified by ITU.

2. In the 12GHz case, by FCC regulation MVDDS and megaconstellation are co-primary, which means they need to share the spectrum on a first come first serve basis, so if MVDDS service was deployed first, megaconstellation will need to give priority to them, there is no "any interference or inconvenience to megaconstellations is unacceptable" argument.

3. And even if megaconstellation make use of the 12GHz before MVDDS, they (and MVDDS) still need to give priority to DBS services, so again, there is no "any interference or inconvenience to megaconstellations is unacceptable" argument.

4. In any case, this is completely incomparable to impact of megaconstellation on optical astronomy, since there is no regulation protecting optical astronomy's use of optical spectrum, unlike the 12GHz's case where there's clear FCC ruling that give megaconstellation limited protection from harmful interference by MVDDS.

[Robotbeat]

But anyway, Starship or successors will either be a thing in 10-20 years or Starlink v2 sized megaconstellations won’t be.

Starship may be used to launch Starlink, diddle about either the moon a bit and be replaced with a "right-sized" version with Falcon 9's capabilities.  We don't know.

And, what other rocket is available to launch something that would barely fit on Starship?  SLS?  You think that will be around in 20 years?

Then the risk to ground telescopes from larger Starlink v2 will also be less.

[eeergo]

Don't forget the new argument: any interference or inconvenience to megaconstellations is unacceptable, nothing to do with interference to astronomy

You have no idea what you're talking about:
[...]

But you do, professionally so:

[...]

Fact: Comparing apples-to-apples the visibility of a random piece of small debris to a 15-m reflective satellite is a fallacy

It is also a fallacy to say Starlink is 15-m long, that's the solar panel size, you have no evidence to show solar panel is the issue here. In fact it's probably not, since SpaceX is not attempting to coat the solar panel.
[...]

[Robotbeat]

Don't forget the new argument: any interference or inconvenience to megaconstellations is unacceptable, nothing to do with interference to astronomy

You have no idea what you're talking about:
[...]

But you do, professionally so:

[...]

Fact: Comparing apples-to-apples the visibility of a random piece of small debris to a 15-m reflective satellite is a fallacy

It is also a fallacy to say Starlink is 15-m long, that's the solar panel size, you have no evidence to show solar panel is the issue here. In fact it's probably not, since SpaceX is not attempting to coat the solar panel.
[...]

You just confirmed what he said.

[FutureSpaceTourist]

https://twitter.com/erinishimoticha/status/1553059526693294084

SpaceX has implemented innovative technological solutions and techniques to minimize its satellites’ effect on the night sky, and today it published new findings to share with the industry. #Starlink #astronomy

https://api.starlink.com/public-files/BrightnessMitigationBestPracticesSatelliteOperators.pdf

[su27k]

https://twitter.com/merrdiff/status/1556860597546823681

Here's the full poster! Alina did a great job fielding Qs. The main takeaway is that while Rubin can dodge sats, it's almost never worth it to try, because we waste time slewing and lose exposures (and thus survey depth). #Rubin2022 Thx @IanPaulFreeley for co-mentoring with me!

[su27k]

https://www.lsst.org/content/lsst-statement-regarding-increased-deployment-satellite-constellations

Vera C. Rubin Observatory – Impact of Satellite Constellations
NOIRLab/NSF/AURA
Revised August 25, 2022

Executive Summary

* The estimated 400,000 recent and planned Low Earth Orbit satellites (LEOsats) threaten the discovery potential of the Rubin Observatory Legacy Survey of Space and Time (LSST).
* While SpaceX is attempting to limit the visibility of their new larger V2 Starlink satellites, significant additional threats to LSST come from other LEO satellite operators. The most recent threat is from the AST SpaceMobile constellation. Each of their large, bright satellites will saturate the LSST Camera CCDs.
* Simulations of the LSST observing cadence and the full 42,000 SpaceX satellite constellation show that as many as 30% of all LSST images would contain at least one satellite streak. With constellations of 400,000 LEOsats, most images will have very bright streaks.
* Due to its rapid cadence, LSST cannot usefully avoid tens of thousands of LEOsats.
* Measurements of the brightness of the current LEOsats indicate that very bright streaks can cause residual artifacts in the reduced data if no further brightness mitigations are made.
* Darkening satellites to 7th magnitude would simplify removal of some artifacts in LSST images, but there is no guarantee most of the satellites will be limited in brightness to fainter than 7th magnitude.
* The bright main satellite trail would still be present, potentially creating bogus alerts and systematics at low surface brightness. This is a challenge for science data analysis, adding significant effort and potentially limiting discovery of the unexpected.

The Vera C. Rubin Observatory science community is concerned about the increasing deployment of Low Earth Orbit communications satellite constellations which, if unchecked, could jeopardize the discoveries anticipated from Rubin Observatory when science operations begin in 2024. There was a session at the Rubin 2022 Project and Community Workshop discussing these challenges. Because Rubin Observatory is uniquely affected by these satellite constellations, the science team is pursuing mitigation strategies to reduce the impact of the satellites on Rubin Observatory science.

The Vera C. Rubin Observatory is nearing completion, and its Legacy Survey of Space and Time (LSST) will soon offer an unprecedented, detailed view of the changing sky. Starting in late 2024, Rubin Observatory will employ the 8.4-meter Simonyi Survey Telescope and the 3200-megapixel LSST Camera to capture about 1,000 images of the sky, every night, for ten years. Each image will cover a 9.6 square degree field of view, or about 40 times the area of the full Moon. Because of the telescope's large light-collecting area, each 30-second exposure will reveal distant objects that are about 20 million times fainter than those visible with the unaided eye. This large combination of light-collecting area and field of view on the sky is unprecedented in the history of optical astronomy, which is one of the reasons Rubin/LSST was the top-ranked ground-based project in the National Academy of Science 2010 Decadal Survey of Astronomy and Astrophysics.

LSST survey images will contain data for about 20 billion galaxies and a similar number of stars, and will be used for investigations ranging from cosmological studies of the Universe to searches for potentially dangerous Earth-impacting asteroids. However, the revolutionary discoveries anticipated from the Rubin Observatory LSST could be significantly degraded by the alarming pace of new deployments of Low Earth Orbit (LEO) communications satellite constellations.

In late May 2019, SpaceX launched the first 60 of its planned Starlink constellation of 42,000 communications satellites to LEO orbits at altitudes of about 550 km. Since then, SpaceX has launched every 3 weeks and soon will deploy its larger Gen2 Starlink satellites. Many other companies, including Amazon, OneWeb, E-Space, and AST SpaceMobile have also entered the race, and the number of satellites launched may exceed 400,000 over the next decade. The negative impact of these satellites on optical astronomy depends on the number and (critically) brightness of satellites. Rubin Observatory is an extreme case for the sensitivity of astronomical observations to satellite constellations because of its unprecedented ability to repeatedly monitor the sky widely and deeply. During the nominal 30-second visit to a sky patch, satellites in 400-600km LEO orbits typically move about 15 degrees across the sky (about four times the diameter of Rubin Observatory’s field of view), and are visible a few hours after sunset and before sunrise. With 400,000 satellites orbiting Earth, tens of thousands of satellites would be visible above the horizon and it would be difficult to find a circle of 9.6 square degrees anywhere on the sky that does not contain satellite streaks. Simulations of the LSST observing cadence and the full SpaceX satellite constellation show that as many as 30% of all LSST images would contain at least one Starlink satellite trail. With the planned constellations of 400,000 satellites at 400-600 km, all images in twilight will contain streaks. The OneWeb constellation at 1200 km will be visible all night long in Chilean summer. Measurements of the brightness of the current LEO satellites in their final orbits indicate that these trails would cause residual artifacts in the reduced data. If these LEO satellites can be darkened to 7th magnitude, then a new instrument signature removal algorithm can remove some of the residual artifacts. This is challenging due to apparent non-linear crosstalk between the 16 channels on each of the 189 CCDs, the cause of which is still under study. The bright main satellite trail would still be present, potentially creating bogus alerts and systematics at low surface brightness.  Masking of these trails is not 100% perfect. This is a challenge for science data analysis, adding potentially significant effort.

If satellites were darkened to 7th magnitude, they would be 10x below saturation in LSST images. In this case, and with 1-arcminute wide software masks needed to fully mask the faint scattered light from the satellite streak, it is likely that only small fractions of pixels in the affected images—probably in the 1% to 10% range—would be rendered scientifically useless. If this estimate proves correct, the net fraction of lost LSST pixels would be in the 3% range, which corresponds to several months of observing time over the ten-year survey. In the case of the brighter streaks from the large reflective AST SpaceMobile satellites, entire CCDs will be saturated, losing 6-8% of the pixels in each exposure. There is no current understanding that most of the planned satellites will be fainter than 7th magnitude, and so the impact on LSST science may be significant.

However there is a larger challenge: because of the unprecedented large samples, LSST science will be limited by systematics rather than sample variance (area incompleteness). Of concern are various systematic effects that do not simply scale with the number of lost pixels—in other words, the residuals from these mitigation strategies on the science cases for which LSST was designed. For example, the LSST ability to detect asteroids approaching from directions interior to the Earth's orbit may be severely impacted because those directions are visible only during twilight when LEO satellites are brightest—nearly every LSST image taken at this time would be affected by at least one satellite trail. Precision cosmological studies are another example; they are very sensitive to small systematic effects, and might suffer from artifacts due to the removal or masking of the satellite tracks. At the low surface brightness of many LSST science programs, the trail is several hundred pixels wide. Perhaps the largest systematic will be the bogus alerts caused by glints from 400,000 LEOsats rolling to adjust orbits for collision avoidance.

The Rubin Observatory team is working closely with SpaceX engineers to jointly find ways to lessen the impact of the satellite streaks, though no silver bullet yet exists. It remains to be seen how faint Starlink Gen2 satellites will be as a function of sun-satellite-observer angle. In some cases, technical solutions developed by SpaceX may be adopted by other constellation operators. It is already known, however, that other operators of satellites in LEO will present a significant threat to the main mission of LSST: discovery of the unexpected.