Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
×
Communications Space Technology

SpaceX's Bright Starlink Satellites Are Upsetting Astronomers (cnet.com) 199

SpaceX's Starlink satellites -- recently launched from Cape Canaveral last week -- may obstruct the view of the universe, posing serious problems for astronomers. It's not much of a problem now as there are only 60 satellites at present, but eventually that number will reach 12,000, and a megaconstellation will encircle the Earth. CNET reports: The first batch of satellites were launched from Cape Canaveral, Florida, and deployed to orbit by a Falcon 9 rocket on May 23. Each contains a single solar array, which both captures and bounces sunlight off the satellites and, as a result, can sometimes be seen from Earth. On May 25, as the drifting luminescent army of satellites zoomed overhead, Dutch satellite tracker Marco Langbroek captured their marching, posting a stunning video to Vimeo. In time, the satellites will drift apart and head to specific orbits so that satellite internet coverage can be beamed to every corner of the globe.

Will we find ourselves in a position where it's impossible to investigate the cosmos from the ground? The quick answer: not forever, no. SpaceX designed the Starlink satellites to fall back to the Earth after about five years of service, burning up in the atmosphere on their way back in. But the long answer is: potentially. Astronomers already wrangle with the problems posed by space robots and satellites circling the Earth whenever they turn their ground-based telescopes toward the stars. Bright, reflective surfaces pose a problem because they obstruct our view of the universe. More satellites equals cloudier vision, and Starlink plans to launch more satellites than ever.
Elon Musk jumped to the defense of his satellite system, tweeting how "potentially helping billions of economically disadvantaged people is the greater good," while making it clear that SpaceX plans to limit Starlink's effects on astronomy. "We care a great deal about science," Musk tweeted. He said he's sent a note to the Starlink team to reduce the amount of light the satellites reflect.
This discussion has been archived. No new comments can be posted.

SpaceX's Bright Starlink Satellites Are Upsetting Astronomers

Comments Filter:
  • The greater good (Score:2, Insightful)

    by Anonymous Coward

    For him.

    Most of the "economically disadvantaged" people are already online with cellphones. They skipped landlines and cable all together. The poorest even skipped having their own electricity first and pay per recharge at the local market. We have had several stories on slashdot over the years about subsistence farmers in the most remote places checking the weather and grain prices on cheap cellphones.

  • The time has come (Score:3, Interesting)

    by Anonymous Coward on Wednesday May 29, 2019 @02:33AM (#58670664)

    Just like the time has come for us to have a global Internet via low earth satellites, the time has come for Astronomers to move their equipment on the moon. And guess who'll help them put it there? Yes SpaceX.

    With less gravity you can have bigger mirrors too because they won't bend under the gravity like they do on earth. The far side is still pristine for radio astronomy as well. Shielded from earths radio pollution it should be good enough for everyone.

    So have you found a nice spot for the astronomers on the far side yet? Probably near a moon base hopefully (so south pole base).

    It just makes sense!

    • No, it does not.

      First, using satellites for internet isn't that great an idea as it may look. Aside of cost, which you might get under control, there's latency. To get that down, you'd first have to find a way to exceed the speed of light. And then you have things in LEO. Not even GSO. To go to the moon is a completely different beast. If you need a comparison, a LEO is a trip around Long Island in a dinghy compared to a moonshot which is closer to going from New York to Perth. You can't do that with the sa

      • by Applehu Akbar ( 2968043 ) on Wednesday May 29, 2019 @05:20AM (#58671028)

        Aside of cost, which you might get under control, there's latency.

        That's why Starlink is in low orbit, not geosynchronous.

      • by lgw ( 121541 ) on Wednesday May 29, 2019 @06:28AM (#58671180) Journal

        de of cost, which you might get under control, there's latency. To get that down, you'd first have to find a way to exceed the speed of light.

        Which is the point: the speed of light in a vacuum exceeds the speed of light in a fiber-optic cable by about 60%.

        • Re:The time has come (Score:5, Interesting)

          by fazig ( 2909523 ) on Wednesday May 29, 2019 @06:54AM (#58671284)
          That's a bit too simplistic.

          You also have to consider the distances and how such a constellation of LEO satellites that move at high speeds and therefore constantly changing distance has to coordinate their efforts in transmitting and receiving a decent enough signal up and down.
          For this these 'constellations' need to work together as a phased array antenna, which requires the information that is to be sent to you to be spread over many different satellites and the information they receive will be spread out over many different satellites as well. All of which requires constant communication between each other and computation to put things back together.
          From an engineering perspective this is a highly interesting topic. Admittedly it's bit over my head to do it all on paper beyond understanding the principles it is based on. So I'm curious and look forward to see it at work.
          • You are right that Starlink uses a LEO constellation with phased array antennas. This is very different from conventional constellations which orbit at GEO 22k miles up and get latency from the basic fact they are just really far away. The nature of fast orbits does introduce new challenges but AESA arrays are uniquely equipped to quickly track pinpoint targets. This wouldn't be possible without them. Any latency issues with starlink are going to be more due to switching than a hard wall with physics and th

      • Except SpaceX is working on a spacecraft design that allows you to go to LEO and to lunar surface "with the same dinghy", so as to speak.
      • by st0nes ( 1120305 )

        First, using satellites for internet isn't that great an idea as it may look.

        How else am I going to get broadband on my yacht?

    • by havana9 ( 101033 ) on Wednesday May 29, 2019 @04:38AM (#58670940)
      We all know how is going to happen. First of all on the moon base al the women will wear silver miniskirts and purple wigs, where the commander on the Earth base will have a blonde wig and a lot of problem with his family. Next due a weird accident we are going to throw out the moon out of the orbit, and the moon will start to travel in the outer space meeting weird plants and even the Voyager.
      • by AmiMoJo ( 196126 )

        Apparently the Andersons were convinced that wigs would become a standard part of military uniforms by 1980 (when UFO was set).

        The best thing about that show (apart from the theme music) was all the cool futuristic tech they imagined back then. Straker, head of SHADO, had all the gadgets. A drink dispenser in his office, of the kind found in modern fast food restaurants, dispensed bourbon and other liqueurs. What seemed futuristic and cutting edge turned out to be the cheap crap of the future.

        • Re:The time has come (Score:5, Interesting)

          by CrimsonAvenger ( 580665 ) on Wednesday May 29, 2019 @07:25AM (#58671362)

          What seemed futuristic and cutting edge turned out to be the cheap crap of the future.

          Which has been pretty much true for all of civilization.

          One example might be the flintlock - bleeding edge when it was first developed (when the matchlock was the cheap crap), but replaced by the caplock, then the cartridge firearm, then bolt-action, then semi-auto, then full-auto, then caseless, then. At each stage, the new thing was "cutting edge", and what it replaced was "cheap crap"....

          • by Agripa ( 139780 )

            One example might be the flintlock - bleeding edge when it was first developed (when the matchlock was the cheap crap), but replaced by the caplock, then the cartridge firearm, then bolt-action, then semi-auto, then full-auto, then caseless, then. At each stage, the new thing was "cutting edge", and what it replaced was "cheap crap"....

            Based on what I have seen, Stormtroopers should have stayed with semi-auto.

      • What the fuck was it from? plz ty
    • by gtall ( 79522 )

      Sure, the moon is the place for astronomy. It should cost, what, a few billion to get a single telescope up there, a few billion to run it, a few billion to ferry supplies and spare parts to it, a few billion for installation of said parts with people since robots cannot handle new "coincidences" very well. Why, it's good enough for everyone.

      Oh, and don't forget to miss the Chinese craft doing their thing up there, they'll want their own so as to be certain their pictures are not doctored by the Russkies or

    • Comment removed based on user account deletion
    • I have no desire to have satellite internet. I have no desire to look at electronic image of stars. I like to look at stars. I'm not sure why you think your preferences should take priority over mine, just because some egotistical billionaire agrees with you.

    • We run the LOFAR Radio telescope.
      It has over 100.000 antennas, spread over most of the EU, with distances up to 2000 km.
      It uses the high speed EU research network infrastructure to get the signals to one spot for processing.

      It took about €100 million to build. It's cheap because we used cheap parts and spend more on operational maintenance, by driving to the sites and fixing things.
      The price of one SpaceX launch.

      Running something like this on the backside of the Moon, which isn't really even big enough

  • by mrwireless ( 1056688 ) on Wednesday May 29, 2019 @02:53AM (#58670718)

    I don't see it mentioned, but one of the issues astronomers currently have is with "Iridium flares". The old Iridium satellites have solar panels at an angle that reflect sunlight back to earth. If you're an astronomer taking a picture with a highly sensitive CCD chip with a big lens in front of it, these flashes of light can damage your camera.

    This is why astronomers that I know use special apps that tell them exactly where and when these flashes will occur, so they can avoid them. For example: https://www.cloudynights.com/t... [cloudynights.com]

    I can imagine that thousands of new satellites could make it harder to avoid such flashes?

    • by Zocalo ( 252965 ) on Wednesday May 29, 2019 @03:32AM (#58670798) Homepage
      The models of Iridium satellites that created the flares are now all out of operation [skyatnightmagazine.com] (the last true Iridium flare was in late 2018) and most, if not all, have been de-orbited. There are still satellites that flare, but there's much less predictability to it and typically they do so with much less magnitude than the naked-eye visible Iridiums used to. The issue at present is more that they are all close together; once they fan out into their orbits the impact will be much lower, and since Starlink satellites are pretty small so if/when a specific satellite flares it's likely to be a fairly low magnitude affair. Going forwards though, it is still going to mean a LOT of extra dots in the sky that will need to be discounted from observations which likely sucks, but there's seldom progress (and Starlink could be revolutionary for huge areas of the world without much telco infrastructure) without some pain.
      • KInd of ironic that SpaceX was the company that launched the new non-flaring Iridium birds. There's got to be some evil plot here, but I can't figure it out...

    • There are probably no folks here old enough to remember this, but when I watched the clip of the satellites, I immediately thought of the Captain Kirk's Star Trek episode titled The Tholian Web.

      To sum it up, the Enterprise finds itself in a Bad Neighborhood, and the locals want to kick it out. So two of their ships start building a "Web" around the Enterprise, and at the end zap it out into another area of the universe.

      So . . . I'd like a Tweet from Musk explaining what his true intentions are.

      Oh, yeah.

    • by Ecuador ( 740021 ) on Wednesday May 29, 2019 @08:57AM (#58671680) Homepage

      You have it a bit backwards. The Iridium satellites were few and their flare was concentrated on a small region on the ground (hence the brightness) so we had these apps in order to be able to catch one - as they were quite spectacular. The probability of one appearing in one of your astrophoto exposures was negligible. I should know, I do astrophotography and develop one of those astro weather apps that tell you about flares. Now the SpaceX satellites will be thousands not dozens, so it's a different matter.

  • by Anonymous Coward

    These astronomers want the sky to themselves. Assholes.

    Elon is helping billions of economically disadvantaged people and will earn billions more dollars and worshipers through this. That is the greater good.

  • 12,000 micro satellites will surely create so much space junk that near earth orbits will become unusable.

    • by Anonymous Coward

      What stops these satelites smashing into each othe
      12,000 micro satellites will surely create so much space junk that near earth orbits will become unusable.

      We have an order of magnitude more ships at sea at the same time. Why are *they* not all crashing into each other constantly?

      It's because even with hundreds of thousands of ships, all much larger than these satellites, the surface of the earth is HUGE.
      Any given range of orbit in space is even more mind mindbogglingly huge.

      The launch represents the last time these satellites will ever be within tens to hundreds of miles of miles of each other.

      • by GuB-42 ( 2483988 )

        Ships, and even airplanes don't go as fast as satellites. And when they crash, they sink, they don't create a multitude of debris circling the earth faster than bullets.

        The issue with space junk is called the Kessler syndrome. As satellites crash, they create debris, and these debris can hit other satellites and other debris, creating even more. It can result in a chain reaction making orbits unusable for quite a while.

        It is serious enough to require all satellite operators to come with a space junk managem

    • 12,000 micro satellites will surely create so much space junk that near earth orbits will become unusable.

      Right, I'm sure we all still vividly remember the armageddon when we had 12000 cars driving around.

      • Implying cars never crash? If every car crash left debris traveling just as fast (fast enough to fuck you up) and in a bunch of different directions would it be any better?
    • They are equipped with their own propulsion in the form are small krypton powered hall effect thruster. So they should be able to maneuver around known debris. The trick of course is unknown debris but LEO is pretty well documented and stuff there tends not to stay in orbit all that long.

  • by Voice of satan ( 1553177 ) on Wednesday May 29, 2019 @03:43AM (#58670822)

    From TFA: "after a user suggested placing space telescopes using Starlink chassis into orbit to appease the astronomers, Musk said he "would love to do exactly that." "

    Well, Mr Labeyrie, it is time to propose Musk a pilot project of space based hypertelescopes. For those who do not know, an hypertelescope would be a lot of telescopes combining their light by interferometry. The more you have, the more light you gather. The farther they are, the better their resolving power. It has been a dream of astronomers for quite a while but would be horribly expensive and even pilot projects were axed on cost grounds.

    There are limited observations form ground telescopes in optical interferometry and some of the results are impressive. Tow Keck telescopes at Hawaii (80m apart) and four VLTI telescopes in Chile (200m apart). Without them we wouldn't have resolved Betelgeuse and (somehow) Antares.

    Imagine with a constellation of small space telescopes working together. A constellation you could improve with time !

    About optical interferometry: https://en.wikipedia.org/wiki/... [wikipedia.org]

    About astronomical interferometry in general: https://en.wikipedia.org/wiki/... [wikipedia.org]

    • Re: (Score:2, Interesting)

      by Anonymous Coward

      The size of Starlink satellites mean that it wouldn't be very useful. While interferometry can increase resolution, it must start with an antenna of a size large enough to resolve the single remote emitter that you're trying to resolve (star, black hole, etc.), because otherwise, nearby elements will add interfering noise that you have no way to cancel.

      The Wikipedia reference indicates, for optical interferometry, the high frequency of visible light requires that the light be optically combined, which very

    • by Anonymous Coward

      For interferometry, you need to know the distance between your telescopes to a precision of less than (~5% of) the wavelength of light. A typical optical wavelength is 500nm, so you need to know the distance between your telescopes to a precision of about 25nm.

      That's hard enough for Keck and the VLT: they have to deal with e.g. thermal expansion/contraction changing the separation between their component telescopes. To do it with free-flying satellites? That's a crazy-hard problem.

      Interferometry has been

    • by gtall ( 79522 )

      Can't we just shot put Elon Musk up there and tell him to phone back periodically on what he thinks he sees?

    • From TFA: "after a user suggested placing space telescopes using Starlink chassis into orbit to appease the astronomers, Musk said he "would love to do exactly that." "

      He just wants someone to pay him to build a Very Dangerous Array. [fandom.com] (Ref [schlockmercenary.com])

  • The astronomers shouldn't bitch about this and just move on, this is just the first of many MANY satellites that are going to be put into orbit around our planet. The problem already existed with the few satellites that were already around us. This 'problem' has been coming for a long time, and yet the professional astronomers haven't done anything about it to find a way around it. Not that I like a lot of satellites hanging around, but there's little one can do about it as it's progress. In a couple of dec

  • Wait until 12.000 satellites send laser messages to each other, the ground stations and a couple of billion users, _then_ you have reasons to complain.

  • Comment removed based on user account deletion
    • A proposal for that was already shot down this year. And in several countries, including the USA and I think Russia using space for advertising is illegal. So at least two of the major places you would go to get such an abomination in space wont launch it. I'd imagine most other launch capable nations would put a halt on it too.

  • by Festering Leper ( 456849 ) on Wednesday May 29, 2019 @06:26AM (#58671172) Homepage

    To maintain a network of 12,000 satellites with a 5 year lifespan will require an average of 40 launches every year (more than 3 every month) on a continuous basis, assuming 60 satellites per launch. That doesn't seem very efficient or profitable.

    • by umghhh ( 965931 )
      Citizen! I would like to point out that you did not think how the already planned launches will help in fighting the patriarchy and oppression.
    • Assume each launch costs $5 million for fuel and vehicle prep, plus $1 million for the satelites themselves. Both figures are likely to be lower, but let's be generous. At 40 launches per year that's $240 million per year.

      Given that SpaceX is projecting that the network could provide $30 billion in annual revenue once fully operational, I think they'll be ok. Even if their estimate is off by an order of magnitude the business would still be quite profitable.

      • by quenda ( 644621 )

        Assume each launch costs $5 million for fuel and vehicle prep,

        Fuel is cheap. Launches cost a lot more than $5m, even with a re-used booster. They charge $50m for "flight proven", and I doubt that is 90% profit.

        • Clearly that's a future estimate, not what's happening right now.

          The price to launch on a new one is $62M.
          The $50M price is a way to encourage customers to transition to used rockets while still having enough profit margin to recoup development costs in a reasonable amount of time.
          It's definitely going to be cheaper for internal launches.

          The current iteration of the rocket is designed for 10 launches without refurbishment, 100 launches after some occasional refurbishment.
          They're currently recovering the 9 e

        • They charge $50m for "flight proven"

          They would be stupid to charge less given that everyone else charges more than that.

          and I doubt that is 90% profit.

          Of course not. Replacing the fairings every single time on it's own costs something like $4 million. That's why they're working to recover those in the future. Amortizations of the vehicle also adds cost. The Falcon 9 first stages probably costs around $30 million to make, and none of them have flown more than 3 times. So with just those two factors we are up to a cost of $14 million per launch; and never mind recoverin

          • by quenda ( 644621 )

            So with just those two factors we are up to a cost of $14 million per launch; and never mind recovering the cost of development and the loss of earlier rockets.

            You seem to be avoiding the biggest cost: the disposable second stage. I've seen a $9m estimate floating around. Its got to be more than the fairings!
            To get marginal costs down to the sort of levels you mentioned requires full re-use, and boosters returning to base, not sea. That is their plan for the BFR.

            • That's fair. I know they were screwing around with trying to recover the second stage too but they seem to have given up on that and decided to just push forward with Starship instead. Still, even if you bump up the total cost to $20 million that gets you a per-year cost of $800 million compared to $30 billion in projected revenue. Again even if they're off by a factor of magnitude, $3 billion would be enough to make it worthwhile ... and if Starship works as expected, the cost will decrease over time.

    • With an operational Starship, you're looking at 6 launches per year.
    • by eth1 ( 94901 )

      To maintain a network of 12,000 satellites with a 5 year lifespan will require an average of 40 launches every year (more than 3 every month) on a continuous basis, assuming 60 satellites per launch. That doesn't seem very efficient or profitable.

      Well, how much does it cost per year to maintain last-mile access to every address in the world that currently has wired internet access? Wouldn't that be the test of whether it might be economically feasible? Especially since you then get every address that *doesn't* have wired access for "free"?

    • And with two other outfits planning to do the same, the competition will be considerable.

      All the rich people (i.e. those earning above the global median wage) already have internet access. Those below the average won't have much money to spend on such luxuries (they'd need electricity first, a groundstation second and content in their own languages, too). So how much would a subsistence farmer in Africa or a lost tribe in South America be willing to pay per month? $1, 1 cent ... less?

      • There are other use cases as well. Fast forward to 2021 or 2022:

        Getting dragged out into BFE on some weekend family adventure & dreading the prospect of being unable to get online for a few days? Hurricane just knocked out your area's cellular network and electricity for the next week or two? Just buy a Starlink ADVENTURER pack for $149 at Best Buy. It's a weatherproof unit the approximate size of a 6-pack that combines the phased array antenna, radio, and a wifi access point. It includes two power dong

    • Maybe with the BFR, they will launch a lot more satellites per launch.
    • Our local fiber company wanted to charge $25,000 to run fiber 2 blocks to our office. That's 1 office. Now imagine 10,000 offices around the city. That's $250m to deploy fiber.

      Comcast claims they spend $10B a year just on the US.

      40 launches a year at commercial rates is $60m * 40 = $2.4B. Assume 2,400 satellites a year * $1m each = $2.4B more.

      So Comcast $10B a year for the US vs Starlink $5B/year for the entire planet.

  • Satellite orbits are pretty predictable. So why can't you just stop taking data for an instant or two whenever a satellite *might* cross through your view. Since most astronomy collects light over a decent period of time and the satellites only occupy a small fraction of the sky why can't we just work around this only losing a tiny percent of photon collecting time.

    Now for survey telescopes it seems a bit harder but doesn't seem that devastating. I presume the actual astronomers know better than I do so

  • ...is measured in AU.
  • Clearly we need more tunnels in space.
  • ... and the early adopters often get left behind and sometimes become extinct.

    I'm thinking of the Stone Age buffalo hunters in what's now the United States.

    European invaders with Iron Age technology displaced the Early Peoples. We accept that.

    Astronomers enjoy no more respect than Native Americans.

    The ways of the Old Ones is past. Ground astronomers are fighting the Iron Horse.

    Lessons learned.

  • ...launch space based satellites
  • > Bright, reflective surfaces pose a problem because they obstruct our view of the universe.

    While opaque unreflective ones don't. Right?

  • The bodies of the satellites could presumably be painted with VantaBlack - which would prevent reflections from everything except a few components like laser apertures and such. But the solar panels can't be handled that way (duh!).

    Of course, the solar panels are supposed to ABSORB light - and not reflect it - so you'd hope that they wouldn't be a problem.

    However, clearly they are. But since this is only a problem with shallow-angle rays from the sun just before dawn and just after sunset - you kinda wo

  • Solar panel is opposite the phased array antennas and the antenna is the part facing earth. Also being a quite low orbit and tiny I'd expect them to be invisible except at dawn/dusk when they are in sunlight and the observer is in the dark. Do we know how long when that video "stream" was taken from dawn/dusk?

  • Didn't see anyone else link the vid: https://vimeo.com/338361997 [vimeo.com]

"Poor man... he was like an employee to me." -- The police commisioner on "Sledge Hammer" laments the death of his bodyguard

Working...