Impacts of Large Satellite Constellations on Astronomy

[su27k]

It is almost like a dyson sphere/dyson swarm (just surrounding a planet rather than a star). If we aren't smart about it, space based internet won't be a thing...and ground based astronomy...and space based astronomy or space based anything really.

Most of the debris is generated by governments, especially large ones (upper stages, big satellites) that have the potential to generate huge amount of debris if collision occurs.

And there needs to be a better justification than fortnite ping times.

That is for market to decide, you or anyone else do not have the ability to determine if a product will offer value to customers, only market can determine this.

[su27k]

Solution:
Switch to space based astronomy. [...]

This has been explained, proven and repeated ad-nauseam not to be a solution, in this very thread among other places, just a far-future declaration of intentions.

A sky full of LEO constellation satellites is also a far-future declaration of intentions. The reality is we only have ~700 satellites launched, in order to reach thousands a partially reusable launch vehicle is required, in order to reach tens of thousands, a fully reusable launch vehicle is required, neither is easy to develop.

You cannot (should not, rather) deny access to a natural resource, especially for private profit, when no options exist to do without it.

LEO constellation is not denying astronomers' access to the sky, as the SATCON-1 report shows, it merely degrades their access somewhat, with proper mitigation the degradation is minimal.

And how many times do I have to point out LEO constellation is not entirely private? Majority of the constellations are either funded or owned by national governments.

Just the single topic of hazardous NEOs would be enough to justify this, let alone with the myriad other facets on the table.

If you actually read the SATCON-1 report, the NEO search community stated that they can mitigate the effect of LEO constellation fully by building more (terrestrial) telescopes, the only question is the funding needed. This fits very well with FinalFrontier's monetary compensation scheme, I for one support such a scheme with the caveat that it shouldn't be a tax, but more like an emission credit.

Even if the US government prohibited the construction of these things other countries would just build their own eventually. There are to many advantages to this technology someone is going to build it regardless.
Short version: it's not getting banned or restricted people will have to find other solutions.

Other countries are following on the footsteps of a non-elegant, brute force solution - anyway, it would be off-topic to delve into megaconstellations' business model in this thread. However, saying "it's not going to get banned or restricted" after a few years' development of a completely new paradigm is, apart from extremely arrogant, also a bit like claiming, I don't know, dumping radioactive waste in the oceanic abysses in the 50s is just going to grow as an unavoidable consequence to energy production, or river diversion in the central Asian steppes will be the future of agriculture, so "people" will have to suck it up and quit whining about whales, disappearing seas or other minutiae in the name of "progress".

Or it's a bit like killing birds with wind turbines or solar farms destroying Joshua trees, which gets accepted without question.

I do agree that it's too early to say for certain that LEO constellation will not be banned, but all evidence pointing to it won't because: a) no such ban is on the agenda either internationally or locally; b) pretty much every space faring nation is building or planning a LEO constellation, banning it will just shoot themselves in the foot.

There will of course be restrictions, there're in fact restrictions - or rules - right now, mainly enforced by FCC, their equivalence in other countries and ITU. And FCC is planning more strict rules in the future, which for the most part the LEO constellation operators have been supportive of.

[LouScheffer]

Solution:  Switch to space based astronomy. Many have been saying for decades that we need more space based radio and optical telescopes

I agree with this on purely technical grounds, though the transition will be decades out.

Observing through the atmosphere is a major limitation.  Astronomers make this tradeoff already - they put all big telescopes, and many radio telescopes, in very inconvenient places since the atmosphere is thinner there.  No one in their right mind would put a big telescope near a major city, no matter how convenient that would be for the builders and operators.

But we have reached the limits of this strategy with ground based operation.  We're already using the best sites on Earth.  We're spending hundreds of millions on access to remote places and then reducing the effect of the remaining atmosphere, using techniques such as adaptive optics.  And after all this, we still get crappy seeing (compared what an 8m telescope could do with no atmosphere) and the atmosphere is still opaque in the UV, much of the IR, lots of low (ionosphere) and high (gasses and water) radio and mm wave bands.  It makes the theoretically simple interferometry very complex in practice.  It also adds airglow, zenith-angle dependent dispersion, frequency dependent time delays for ranging, and other limitations.  Furthermore, the bands the atmosphere screws up most are exactly the ones we'll want to look at for exoplanet studies. 

So I think over the next few decades, astronomy will shift to space.  People will think of siting a new facility on Earth the same way they now think of citing an observatory in a major city.  Sure, it's less convenient, but the alternative of working at the bottom of an atmosphere sucks even more. It's exactly the same rationale as now - building in an inconvenient spot since the observing is better.  And as time goes on, and the requirements get stricter, the workarounds for gazing through the atmosphere get harder, while meanwhile the technology for working in space is improving.  At some point meeting the requirements in not only easier in space, it's cheaper overall.   We are a long way from this today, but it's the way things will turn out.

The timescale for this is not short.  In the first half of the 20th century, astronomers build near universities (Mt. Lick, Mt. Palomar, Mt. Wilson).  70 years later, all new facilities are built at high altitude deserts.  70 years from now, all new facilities will be built in space.

[su27k]

Yes, ironic if the “game-changer” of “cheap access to space” which SpaceX and others have promised (and at least somewhat delivered) ends up destroying all access to space for everyone and for ever...

This has been discussed here previously, but these satellites will reenter in a few years if they lose control.  As will *everything* in LEO.  It would be possible to wreck the GEO ring indefinitely, but not LEO.  That’s not how LEO works.

Well, at least that is true for the two billionaire controlled constellations (Starlink and Kuiper), both uses orbits below 600km, this ensures that even if satellite lost all propulsion it will reenter in a few years (less than 5 years for Starlink, around 6 years for Kuiper). Incidentally, 600km is also strongly recommended by astronomers as the highest altitude for LEO constellation, because putting satellites above 600km will have big negative impact on optical astronomy.

However, other constellations such as OneWeb (partially owned by UK government), Telesat (partially funded by Canadian government), and some Chinese constellations (all linked to Chinese government one way or another) are planning to be deployed at 1000km+ orbits, these will take hundreds of years to naturally decay, and they'll have big negative impact on optical astronomy.

The "billionaire bad, government good" idiots are really clueless when it comes to constellations.

[matthewkantar]

Astronomers are a resourceful bunch. They figured out how to mitigate the effects of the atmosphere for Pete's sake. Though it is clear they would rather not, they will figure out how to work with 100,000 sats up there.

I wonder if Britain and China will be as concerned about minimizing the negative effects as SpaceX seems to be?

[high road]

Things are getting slightly out of hand....



It is almost like a dyson sphere/dyson swarm (just surrounding a planet rather than a star). If we aren't smart about it, space based internet won't be a thing...and ground based astronomy...and space based astronomy or space based anything really. And there needs to be a better justification than fortnite ping times.

If we add all megaconsellations ten years after this was made, we wouldn't see much of a difference. Repeat every ten years to see if the orbital debris problem escalates or even improves because more stuff gets placed up there leading to more regulation and debris removal becomes cheaper.

[Lar]

This thread has a lot of polemics and a lot of politics. In a lot of cases, from people who have been around quite a while and who presumably know better.

Going forward, do better.

Thank you.

[eeergo]

A sky full of LEO constellation satellites is also a far-future declaration of intentions. The reality is we only have ~700 satellites launched, in order to reach thousands a partially reusable launch vehicle is required, in order to reach tens of thousands, a fully reusable launch vehicle is required, neither is easy to develop.

That's just asking the frog to accept the "boiling pot" paradox. In any case, the timescale of this latter "declaration of intentions", already being heavily acted upon (700 satellites is about twice the amount of previously-existent LEO birds of ALL kinds, private and governmental, and it's probably safe to say at least as many units are ready and rearing to be launched as soon as practicable), is orders of magnitude shorter than reaching any sort of useful replacement capability for the former - kind of starting a house from the roof.

LEO constellation is not denying astronomers' access to the sky, as the SATCON-1 report shows, it merely degrades their access somewhat, with proper mitigation the degradation is minimal.

And how many times do I have to point out LEO constellation is not entirely private? Majority of the constellations are either funded or owned by national governments.

Likewise, dumping radioactive waste in the oceans in the name of "progress" in energy generation, or drying up entire seas in the central Asian steppes in the name of agricultural boosting, did not and would not have even if they had gone on, strictly denied access to those natural resources - "merely" degraded their access "somewhat". As for the public-vs-private condition of megaconstellations: they have been kickstarted, promoted as the only thinkable alternative, and deployed urgently, exclusively by private entities. Even the Chinese proposals are as private as things like these can get in that country. That governments follow the prevailing lead cannot be unexpected.

If you actually read the SATCON-1 report, the NEO search community stated that they can mitigate the effect of LEO constellation fully by building more (terrestrial) telescopes, the only question is the funding needed. This fits very well with FinalFrontier's monetary compensation scheme, I for one support such a scheme with the caveat that it shouldn't be a tax, but more like an emission credit.

That's very much NOT the conclusion of the SATCON-1 study, and I can't believe that's a good faith misunderstanding. I posted a summary of the report back in the days when it had just been released. Have *you* read it? It doesn't appear so from your comment, when you are so thoroughly misquoting the report's conclusions that read as follows:

"either the Starlink Generation 2 or the OneWeb scenario (of order 40,000 satellites) will
significantly degrade twilight near-Sun observations, especially for the LSST, [...] The satellite trail masking developed for the LSST pipeline processing is very promising, but it may also unintentionally remove trails originating from NEOs. "

(section 3, page 12 here https://aas.org/sites/default/files/2020-08/SATCON1-Report.pdf)

Sure, mostly anything is solvable with more money. If you have unlimited money to build a cutting-edge observatory in every other suitable location on Earth, including oceans, and throw a couple in space or the far side of the Moon for good measure, maybe the potentially-species-dooming issue of NEO detection being unnecessarily hindered by megaconstellations wouldn't be *that much* of a problem (although maybe the other many important research avenues would still be impacted).

Likewise, with enough money, and not even that much, proper resilient minimal-impact ground infrastructure that can be easily upgraded and passively passed on to future generations would also be feasible... why isn't this being pursued instead? Why do research institutions or governments (i.e. public funding) have to pay for a problem that is being relentlessly pushed forward by (very few) private individuals' initiative? Under this outlook, of course it's no wonder (some) satellite operators have welcomed half-hearted leaderless *suggestions* to limit their impact through a few simple, second-thought measures on the most glaringly-obvious part of their medusa-head impact tree.

[Frogstar_Robot]

Meanwhile at the United Nations Office for Outer Space Affairs, the online workshop "Dark and Quiet Skies for Science and Society" has concluded.

The following web page has links to videos of sessions, plus pdfs of presentations : http://www.unoosa.org/oosa/en/ourwork/psa/schedule/2020/2020_dark_skies.html

The plan going forward :

The online event will result in a document that describes what measures Governments and private enterprises can adopt to mitigate the negative impact of technological implementations on astronomy (e.g. urban lighting, radio broadcasting and satellite constellations' deployment) without diminishing the effectiveness of the services they offer to citizens. The final outcome document, intended to become a reference for further analysis of the situation, will be presented to the intergovernmental Committee on the Peaceful Uses of Outer Space (COPUOS) for consideration.

The involvement of COPUOS is particularly important since it is the natural international forum where these matters should be brought to the attention of the space community.

I understand that in turn COPUOS may present a resolution to the UN General Assembly for consideration.

[Frogstar_Robot]

Why do research institutions or governments (i.e. public funding) have to pay for a problem that is being relentlessly pushed forward by (very few) private individuals' initiative?

I still don't understand the fixation on a few billionaires, it has been proved ad nauseum that it is not the case. I guess it is easier to stir up hate of identifiable people rather than abstract organisations. Ironically, people like Elon Musk with a public profile do seem willing to engage, and SpaceX appears to be doing the most to help mitigate the problems with their satellites.
 Patricia Cooper, who is is Vice President for Satellite Government Affairs at SpaceX, took part in the UNOOSA "Dark and Quiet Skies for Science and Society" workshop, for example.

But you continue to push the debate into polemic and political discussion, I would like to respond to your question but we have been specifically reminded not to.

[ncb1397]

Well, at least that is true for the two billionaire controlled constellations (Starlink and Kuiper), both uses orbits below 600km, this ensures that even if satellite lost all propulsion it will reenter in a few years (less than 5 years for Starlink, around 6 years for Kuiper). Incidentally, 600km is also strongly recommended by astronomers as the highest altitude for LEO constellation, because putting satellites above 600km will have big negative impact on optical astronomy.

Not sure how accurate these numbers are. The counter-example being Astro H. Break up was March 26th, 2016. Altitude of orbit was between ~560 and ~580 kilometers. Here we are 4.55 years later and only 4 out of 13 catalogued objects have decay dates on space-track.org.

see:
https://www.space-track.org/#catalog
and input astro h in the satname column

In fact, plugging some of these object catalogue numbers into n2yo.com and some appear to still be above 500 km....

see:
https://www.n2yo.com/satellite/?s=41439#results

Do you have a source for your numbers? They may be dependent on the solar panels still being attached and acting as a drag on the spacecraft. Post break up, those numbers may not apply.

[eeergo]

Why do research institutions or governments (i.e. public funding) have to pay for a problem that is being relentlessly pushed forward by (very few) private individuals' initiative?

I still don't understand the fixation on a few billionaires, it has been proved ad nauseum that it is not the case. I guess it is easier to stir up hate of identifiable people rather than abstract organisations. Ironically, people like Elon Musk with a public profile do seem willing to engage, and SpaceX appears to be doing the most to help mitigate the problems with their satellites.
 Patricia Cooper, who is is Vice President for Satellite Government Affairs at SpaceX, took part in the UNOOSA "Dark and Quiet Skies for Science and Society" workshop, for example.

But you continue to push the debate into polemic and political discussion, I would like to respond to your question but we have been specifically reminded not to.

My apologies to moderation for going into metadiscussions again, but I do feel I have to answer such direct addresses.

I've been mostly reflecting the consensus conclusions, quite thoroughly backed up by analyses and public reports (in contrast to the opposing lobby's theses), of a sizeable part of the scientific community of interest to this forum generally and this thread's title specifically - including the recent workshop you link to, which unfortunately doesn't yet offer the relevant presentations for public download.

I fail to see why questioning the funding avenues for the mitigations, discussed openly in said reports and in this very thread by those supportive of megaconstellations, proposing alternatives, or making it explicit in the discussion who the lobbying parties are for these concepts should be polemic or overly political, in spite of repeated explicit insults on the side of posters with an opposing viewpoint I've chosen to ignore (idiot/idiotic/stupid...).  It so happens that the three main megaconstellations of relevance to these studies are privately-owned, and their leading figures highly prominent billionaires in the public domain - of course backed up by sizeable organizations and even governmental support, nothing operates in a vacuum. Where has that been "disproven ad nauseam"? That this basic truth is being polemicized upon, along with ignoring other established solid results concerning impacts to astronomy while unempathically calling for them to be accepted as they come, is IMHO what keeps driving this thread into ugly terrain.

[su27k]

Well, at least that is true for the two billionaire controlled constellations (Starlink and Kuiper), both uses orbits below 600km, this ensures that even if satellite lost all propulsion it will reenter in a few years (less than 5 years for Starlink, around 6 years for Kuiper). Incidentally, 600km is also strongly recommended by astronomers as the highest altitude for LEO constellation, because putting satellites above 600km will have big negative impact on optical astronomy.

Not sure how accurate these numbers are. The counter-example being Astro H. Break up was March 26th, 2020. Altitude of orbit was between ~560 and ~580 kilometers. Here we are 4.55 years later and only 4 out of 13 catalogued objects have decay dates on space-track.org.

see:
https://www.space-track.org/#catalog
and input astro h in the satname column

In fact, plugging some of these object catalogue numbers into n2yo.com and some appear to still be above 500 km....

see:
https://www.n2yo.com/satellite/?s=41439#results

Do you have a source for your numbers? They may be dependent on the solar panels still being attached and acting as a drag on the spacecraft. Post break up, those numbers may not apply.

Kuiper's number is from this article: "A satellite failing at Kuiper’s highest altitude will take around six years to burn up in the atmosphere."

SpaceX number is from their FCC filing (Attachment Technical Information): "Thus, even assuming an extreme worst-case scenario – i.e., the spacecraft fails while in the operational orbit (circular at 550 km), has no attitude control, and solar activity is at a minimum – the longest decay time is still only approximately 4.5-5 years."

It is true that these numbers apply to entire spacecraft, not small debris, but having a disabled spacecraft deorbit quickly is the primary way we can avoid generating smaller debris.

[su27k]

A sky full of LEO constellation satellites is also a far-future declaration of intentions. The reality is we only have ~700 satellites launched, in order to reach thousands a partially reusable launch vehicle is required, in order to reach tens of thousands, a fully reusable launch vehicle is required, neither is easy to develop.

That's just asking the frog to accept the "boiling pot" paradox. In any case, the timescale of this latter "declaration of intentions", already being heavily acted upon (700 satellites is about twice the amount of previously-existent LEO birds of ALL kinds, private and governmental, and it's probably safe to say at least as many units are ready and rearing to be launched as soon as practicable), is orders of magnitude shorter than reaching any sort of useful replacement capability for the former - kind of starting a house from the roof.

The frog and pot analogy doesn't apply, the post started this discussion is talking about monetary compensations to astronomy from constellation operators, what compensation did the frog get in the pot?

You also didn't do the math on timescale, it took a year to get to ~700 satellites, and the constellation satellite only last 5 years in orbit, this means at the current launch rate the biggest constellation can only have 3,500 satellites, no where near the 10k satellites scenario. To get to 10k+ constellation, a fully reusable launch vehicle such as Starship is required, but you yourself stated Starship won't fly to Mars until 2031 or later, so there's plenty of time ;-)

LEO constellation is not denying astronomers' access to the sky, as the SATCON-1 report shows, it merely degrades their access somewhat, with proper mitigation the degradation is minimal.

And how many times do I have to point out LEO constellation is not entirely private? Majority of the constellations are either funded or owned by national governments.

Likewise, dumping radioactive waste in the oceans in the name of "progress" in energy generation, or drying up entire seas in the central Asian steppes in the name of agricultural boosting, did not and would not have even if they had gone on, strictly denied access to those natural resources - "merely" degraded their access "somewhat". As for the public-vs-private condition of megaconstellations: they have been kickstarted, promoted as the only thinkable alternative, and deployed urgently, exclusively by private entities. Even the Chinese proposals are as private as things like these can get in that country. That governments follow the prevailing lead cannot be unexpected.

Well, since you admitted your statement "You cannot (should not, rather) deny access to a natural resource" no longer applies to LEO constellation's impact on astronomy, I don't see what more we need to discuss here.

BTW, while it is true that private companies started this round of LEO constellation building, that doesn't disprove my point, which is majority of constellations are linked to national governments. The governments do not have to follow private companies, there's no law or rule that states they have to do this, especially if as you claimed, the business case is shaky and the cost to public is high. Yet they did follow nevertheless, this tells me either the business case is good or the cost to public is not high, or both.

If you actually read the SATCON-1 report, the NEO search community stated that they can mitigate the effect of LEO constellation fully by building more (terrestrial) telescopes, the only question is the funding needed. This fits very well with FinalFrontier's monetary compensation scheme, I for one support such a scheme with the caveat that it shouldn't be a tax, but more like an emission credit.

That's very much NOT the conclusion of the SATCON-1 study, and I can't believe that's a good faith misunderstanding. I posted a summary of the report back in the days when it had just been released. Have *you* read it? It doesn't appear so from your comment, when you are so thoroughly misquoting the report's conclusions that read as follows:

"either the Starlink Generation 2 or the OneWeb scenario (of order 40,000 satellites) will
significantly degrade twilight near-Sun observations, especially for the LSST, [...] The satellite trail masking developed for the LSST pipeline processing is very promising, but it may also unintentionally remove trails originating from NEOs. "

(section 3, page 12 here https://aas.org/sites/default/files/2020-08/SATCON1-Report.pdf)

Your summary is nothing but a biased reading of the report to make LEO constellation look as bad as possible, I didn't bother to reply to it since I haven't finished reading the appendices yet. But I did look up the NEO parts of the appendices, because this does raise some concerns, here's what https://aas.org/sites/default/files/2020-08/SATCON1-WG-Tech-Reports.pdf has to say about NEO detection:

Page 76 PAN-STARRS

"Pan-STARRS consists of two 1.8-meter telescopes near the summit of Haleakala. Each
telescope has a 3.3-degree diameter field-of-view. Pan-STARRS presently spends 90% of its
observing time searching the sky for Near-Earth Objects (NEOs — asteroids or comets that
have perihelia less than 1.3 AUs). The main motivation for this is planetary defense — to find
any objects that may hit Earth well in advance of the impact — so that efforts can be made to
deflect the impact of larger objects, or to provide warnings for the impact of smaller objects.

All observations to discover Near-Earth objects require four images spaced over approximately
1 hour. Satellite trails will affect each of these four images, and regions in the sky affected in
any of these component images will be ruined for NEO detection. So it is the product of trails in
any of the four images that is damaged for science. Rough estimates suggest that 10–20% of
the imaging area may be damaged, depending on the altitude of satellites and the number, but
more detailed simulations will be required to properly assess the impact.

All-sky video monitors will not help, because there is no agility on scheduling once a one-hour
sequence has been commenced, and the field of view is large. One mitigation would be to
replace the cameras on each of the Pan-STARRS telescopes — approximate cost $6 million
each. The existing cameras use orthogonal transfer arrays, which are not ideal for NEO
searches. New cameras using modern larger CCDs could recover approximately the area of sky
that will be lost due to satellites. The cost of two cameras (one for each telescope) is $12
million. Independent of the satellite threat, Pan-STARRS is planning to upgrade the cameras,
and a proposal for funding has been submitted.

Another mitigation for the NEO search program would be to build another Pan-STARRS
telescope. The approximate cost for a new Pan-STARRS telescope (including camera) is $20
million, plus additional cost for environmental assessment work ($1 million or more); there are
suitable locations for two more Pan-STARRS telescopes on Haleakala. Adding another
telescope produces additional operations cost. If satellite trails cause a loss of 20%, then three
telescopes would produce 1.2 times as much sky coverage as two telescopes with no satellite
trails (assuming 20% loss from satellite trails)."

Page 78 Catalina Sky Survey

"Catalina Sky Survey (CSS) is the longest-running Near-Earth Object (NEO) survey and has
discovered almost half of known NEOs, including more large Potentially Hazardous Asteroids
(PHAs) than any other planetary defense survey. The broader science cases for CSS are similar
to Pan-STARRS and ATLAS (http://nesssi.cacr.caltech.edu/DataRelease). Near-Sun observations in the early evening and late morning hours are especially valuable — when the
interference from satellites in LEO will be largest.

CSS operates follow-up telescopes as well as our surveys using flexible queue scheduling. We
will incur the loss of a similar fraction of our pixels/exposures as other wide-field, large-pixel
surveys described here. Satellite trails, their optical artifacts, and electronic cross-talk introduce
large numbers of false candidate moving objects that must be scrubbed by human eyeballs
and/or machine-learning techniques. The question is: which pixels and images will be
compromised (including archival precovery images)?

Mitigation for CSS operations will depend on predicting which queued exposures will be most
affected, especially by multiple bright satellites in their initial trains, or by particularly
inauspicious multiple satellite crossings. CSS has flexibility in reordering our queues, though
similar scheduling concerns apply as for Las Cumbres Observatory. Given timely and accurate
ephemerides and having predicted an exposure at risk, we may be able to substitute a different
pointing. This will incur slewing overhead and introduce otherwise unnecessary complexity to
our systems and procedures.

Images occur in multi-field multi-exposure sets covering between 50-200 square degrees four
times in about 25 minutes. At some times of the night, every set will see multiple satellite trails.
In the case of tens of thousands of LEOsats in the higher allocated orbits, every image could
include trails. Reordering the queue may only spread the trails more evenly. An alternative
would be to increase the number of repeat exposures of each field from 4 to 5 (or more) to
compensate for clobbered pixels and exposures. This would incur at least a 25% penalty in time
and efficiency of surveying, and ultimately can only be mitigated by commissioning additional
telescopes in support of planetary defense."

Page 103 shows the complete paragraph you quoted, which put things in a different perspective for NEO detection:

"Either the Starlink2 or the OneWeb scenario will significantly degrade twilight near-sun
observations, perhaps fatally for LSST as implied by several presentations at the
SATCON1 workshop. The LSST scheduler will not point near-Sun (meaning far west
early in the evening or far east in the late pre-dawn hours) if there will be a significant
likelihood of LEOsat trails. The detrailing options for the LSST pipeline processing were
very interesting, but these will also tend to remove trails originating from NEOs. Catalina
and other ongoing NEO surveys will be similarly impacted but may not have the same
cross-talk or focus trade-offs as LSST. The NEO Surveyor Mission will cover the
near-Sun region from L-1 and will be unimpacted by LEOsats, however NEOSM will be
sensitive to larger, more distant NEOs, not the more frequent small close-approachers. It
would be useful to model / simulate twilight observations, near-sun or for illuminated
satellites high in the sky for all surveys together."

So the loss of near-sun observations for LSST can be partially compensated by NEOSM, and the overall loss for near-sun observations after all data from all surveys are aggregated is still unclear and need more research.

Page 104 Eric Christensen, Catalina Sky Survey:

"Satellite mega-constellations [probably do not] represent an existential threat to NEO
surveys, at least for programs like CSS that have a single goal of moving object
detection and operate in a truly NEO-optimized way: four exposures evenly spaced over
<1 hour, in the same filter, without strict requirements on photometry.

As for metrics: we could reasonably estimate the impact of satellite constellations on our
efficiency by counting pixels that would be lit up by a satellite — essentially treating
satellite trails as gaps that get carved in our detector. It’s fairly straightforward to
estimate how a 98% fill factor mosaic compares to a 100% fill factor mosaic when used
in a NEO-friendly cadence, for example.

There may be a tipping point where active avoidance or fancy processing techniques
pay off, but even with ~100x more satellites I’m not sure we’d reach that point. Consider
a worst-case where every survey image contains a satellite that fully crosses the image
— 5 binned pixels wide, running diagonally from corner to corner. That’s 37,000 pixels
out of 27.9 million, or about 0.13% of the detector. The fill factor drops from 100% to
99.87%. If the detection criteria were a strict 4 out of 4, then raise 99.87% to the fourth
power and your detection efficiency becomes 99.48%. Relaxing to 3 out of 4 allows you
to recover some of that, preferentially for brighter objects.

Our detection efficiency doesn’t start at 100%, but a rough estimate is that a satellite trail
in every image will cost a few tenths of a percent in detection efficiency. I’d qualify this as
“ negligible ”, nowhere near “significant” or “fatal”.

Our follow-up use case is also more forgiving than most — since we use short
exposures, all we’d have to do is drop out any images that have a trail going directly over
the object. Satellites could be present in the images but if they don’t cross the precise
position of the object, they don’t cost anything. For the worst-case survey example of a
trail crossing every survey image, scale by the FoV of a follow-up telescope (say 1/20th
the FoV of a survey telescope) to estimate one satellite per 20 images. For a 2K
detector, 14,500 pixels out of 4.2 million get carved out, or 3.5%. An average of 2
images per 40-image stack will contain a trail, but only 1 out of about 30 satellite-affected
images will touch the object of interest. I think this boils down to 1 or 2 images per night
might have to be rejected due to a coincidental passage of a satellite over a targeted
NEO — which is unlikely to make the difference between detection and non-detection in
a 10-image stack. I think the worst-case scenario of a trail in every survey image, or in
every 20th follow-up image, is only relevant if satellites are visible all night long to 1-2
meter class telescopes. If they’re visible only near twilight, then scale the overall impact
to detection efficiency down by a factor of 5-10 x."

So just like I said, the impact to NEO detection is not that high and can be compensated by building more telescopes (and it's pretty cheap to build new NEO detection telescope too, just ~$20M)

Likewise, with enough money, and not even that much, proper resilient minimal-impact ground infrastructure that can be easily upgraded and passively passed on to future generations would also be feasible... why isn't this being pursued instead?

You should ask the terrestrial internet providers and governments this question. The answer is likely that ground infrastructure would cost a lot of tax dollars, much more than the tax dollars spent on fixing LEO constellation's impact on astronomy.

Why do research institutions or governments (i.e. public funding) have to pay for a problem that is being relentlessly pushed forward by (very few) private individuals' initiative?

Well if you actually read FinalFrontier's post, he is talking about monetary compensations from LEO constellation operators to astronomy, which would solve this issue.

But even if there's no compensation, and ignore for the moment that many constellations are either owned or funded by government, assuming constellation is entirely private, there is still good justification for government to use public funding to pay for a problem that is caused by private companies: because in exchange the government will get more tax income from a successful LEO constellation, and the LEO constellation would also provide internet connections to rural areas without the need for additional government funding.

This is no different from government spending money on FAA to manage launch licenses, or spending money on FCC to manage spectrum, both are used by private companies.

Under this outlook, of course it's no wonder (some) satellite operators have welcomed half-hearted leaderless *suggestions* to limit their impact through a few simple, second-thought measures on the most glaringly-obvious part of their medusa-head impact tree.

Not sure what this means, the mitigation measures are created by astronomers and summarized in the SATCON-1 report which you quoted, now you are dismissing it as half-hearted leaderless suggestions?

[RedLineTrain]

Important to note that megaconstellations will support a great deal of commercial activity.  Anybody who is a target customer of these services knows this well.  The vast majority of the economic value will be captured by non-billionaires, so Frogstar_Robot's critique of eeergo's focus on billionaires seems particularly apt.

[Mackilroy]

Beyond that, RedLineTrain, if Starlink and other megaconstellations become profitable, there may very well be considerable pressure on ground-based ISPs to offer better services at lower prices (while higher launch rates and improving technology will help start a transition to more space-based astronomy).

[edzieba]

Analogously: who is paying, or should pay, for the impact to astronomy (and the impact to the commons right to night-sky observation, of which so much ado was made about over Starlink launch train trails) from light pollution? If nobody, and the advantages to society of wide availability of artificial lighting are worth the mitigations of putting telescopes at high altitudes in remote areas (often inaccessible, and sometimes even inhospitable), then does similar calculus not apply to the advantages to society of widespread network access - which at this point is as much an essential utility as power and water supply are?

It's not as if constellation operators are wringing their hands and cackling maniacally over recently-stomped tiny models of radio telescopes, after all. As with optical impacts and subsequent deployed mitigations, expect similar in the rest of the EM spectrum as impacts are discovered.

[eeergo]

1. The frog and pot analogy doesn't apply, the post started this discussion is talking about monetary compensations to astronomy from constellation operators, what compensation did the frog get in the pot?

2. You also didn't do the math on timescale, it took a year to get to ~700 satellites, [...] To get to 10k+ constellation, a fully reusable launch vehicle.

3. Well, since you admitted your statement "You cannot (should not, rather) deny access to a natural resource" no longer applies to LEO constellation's impact on astronomy, I don't see what more we need to discuss here.

4. BTW, while it is true that private companies started this round of LEO constellation building, that doesn't disprove my point, which is majority of constellations are linked to national governments. The governments do not have to follow private companies, there's no law or rule that states they have to do this, especially if as you claimed, the business case is shaky and the cost to public is high. Yet they did follow nevertheless, this tells me either the business case is good or the cost to public is not high, or both.

5. Your summary is nothing but a biased reading of the report to make LEO constellation look as bad as possible.

5.1. So the loss of near-sun observations for LSST can be partially compensated by NEOSM, and the overall loss for near-sun observations after all data from all surveys are aggregated is still unclear and need more research.

5.2. Page 104 Eric Christensen, Catalina Sky Survey

6. in exchange the government will get more tax income from a successful LEO constellation, and the LEO constellation would also provide internet connections to rural areas without the need for additional government funding.

7. The mitigation measures are created by astronomers and summarized in the SATCON-1 report which you quoted, now you are dismissing it as half-hearted leaderless suggestions?

Not gonna go the "wall-of-text" way, so a quick roundup:
1. Frog in the pot = you stating one has to wait and see because "only" 700+ satellites have been launched.
2. Current launch rate is not even a fraction of the manifested-for launch rate, with just one operator (and a bankrupt other), no need to wait for a megarocket.
3. Don't misquote me please. Do we need total denial of access to admit a problem?
4. ICBMs are very much a bad short-term profitable business model, both during construction and operation. Wonder why many gov's develop them? You stated megaconstellations were either owned or mainly funded by gov's, which is patently false.
5. Me quoting the conclusions of the final report is biased. You picking single cases out and claiming they're not that bad because one can imagine some level of partial mitigation, against said consensus conclusions, isn't biased. Fine. And again, this is just ONE issue that could potentially have doomsday consequences.
5.1. Partially = minority, in this case. More study = quantitatively how bad the impact will be, whereas its relevance is clear.
5.2. One individual's opinion, stated as such, consisting of back-of-the-envelope estimates, and directly contradicting two others of people in his field and instrument, plus the main conclusions in the report, in an appendix. That's some cherrypicking to prove your point.
6. That may be all good accounting practices, but doesn't even try to solve the problem, à la carbon tax.
7. Yes: they were late to the party, showed indignation upon the faits-accomplis as if somebody should have knocked at their door for permission before developing these systems, have no real insight into megaconstellations other than what they can get from companies' good will, and have comparatively little leverage. Not their job, not megaconstellation operators' fault, but it has been a weak half-hearted lobby so far.

[eeergo]

Important to note that megaconstellations will support a great deal of commercial activity.  Anybody who is a target customer of these services knows this well.  The vast majority of the economic value will be captured by non-billionaires, so Frogstar_Robot's critique of eeergo's focus on billionaires seems particularly apt.

Oh such a big IF, and so at odds with historical precedent. And if it won't?

[RedLineTrain]

Important to note that megaconstellations will support a great deal of commercial activity.  Anybody who is a target customer of these services knows this well.  The vast majority of the economic value will be captured by non-billionaires, so Frogstar_Robot's critique of eeergo's focus on billionaires seems particularly apt.

Oh such a big IF, and so at odds with historical precedent. And if it won't?

Not sure I understand this reply.  Could you please elaborate?