How the nanosatellite boom and battle for space could end in disaster for us all

Heads up. Picture: Warner Bros

Australia is back in the space race it never really joined.

Last week, ASX-listed startup Sky and Space Global successfully launched three satellites from India’s Space Research Organisation (ISRO) into an orbit of between 500km and 800km.

It was a huge deal for the Israeli company. The satellites are CubeSats measuring 10cm x 10cm x 30cm and weighing less than 10kg. They’ll form what SAS are calling its “3 Diamonds” constellation.

“Constellation.” Remember that word, because what you and I have traditionally considered it to mean – dodgy horoscopes, mainly – will change radically in the next decade, forever. To mean something like this:

Picture: OneWeb

That’s the vision of Jersey, US, startup OneWeb, which has ambitions for a network of 2000 satellites orbiting 1200km above the Earth.

For the next eight weeks, SAS will be watching its tiny constellation to see if it functions as planned, as a provider of communications services to regions across the Equatorial Belt. If it does, it will go live and start generating revenue for the fledgling tech company and its CEO, former fighter jet pilot turned space exploration project manager, Meir Moalem.

Sky and Space, Meir says, according to the existing agreements it has, expects “potential annual revenues of around $3 million” over four years.

Its first client, Sat-Space Africa, claims it can make up to $2 million a year from using “only a minor portion of bandwidth capacity” they will provide.

Commercial demonstrations and actual services will begin at the end of the eight-week testing period, which aims to validate the technology, the software and the ability to perform narrowband communications.

“However, keep in mind the goal is our equatorial constellation, which will begin deployment by Q4 2018, and this is the actual commercial space asset, with significant revenues,” Meir says.

That “equatorial constellation” involves the launch with another 200 CubeSats by 2020, with the potential to generate $US500 million a year for the company.

It’s paying Danish company GomSpace up to $75 million to deliver the bigger constellation which will “provide affordable mobile coverage to the 4 billion people” across South and Central America, Africa and Asia, as well as airliners and shipping in the regions.

SAS listed on the ASX on May 31 last year for 2 cents a share. It has since completed three fund-raising rounds, the last of which was for $10 million and was “heavily oversubscribed”.

Ahead of today’s launch, it’s trading at 24 cents – 12 times your money back in just over a year, from an Israeli startup, backdoor listed in Australia, building nanosats in the Netherlands to be launched in India to provide internet access for Africa and South America.

Sky and Space’s 3 Diamonds is go. Picture: Getty Images

It’s a global race to make the world a better place via a market expected to top $23 billion in the next decade. What could possibly go wrong?


First, here’s a quick look at how the number of the tiny things in orbit around Earth has escalated.

It’s surprisingly difficult to get a handle on exactly how many satellites are orbiting the Earth right now. Thousands have been launched, but only about 1000-1500 have orbited or are functional at any one time. It’s a continual replacement cycle – air resistance and the Earth’s gravity slowly pull them down, and they usually burn up as they spiral faster closer to the Earth, or in some cases, fall to the ground.

But many, many other pieces remain. Here’s a rough guide to how things have changed in the past decade. Key moments in space clutter, if you prefer:

  • January 11, 2007: China first put the world properly on notice about the dangers of space debris when it blew up its Fengyun-1C weather satellite using an anti-satellite (ASAT) weapon, adding 2500 pieces of debris to an estimated 13,000 pieces in orbit sized 10cm or bigger.
  • February 11, 2009: A US communication satellite collided with a dead Russian satellite. That added 2500 pieces of manmade debris, bringing it to a total of 18,000 tracked at the time by the US Space Surveillance Network.
  • October 24, 2013: In addition to the debris, Universe Today publisher Fraser Cain pegs operational satellites at 1071, “half” of those in Low Earth Orbit.
  • December 12, 2016: The Union of Concerned Scientists (UCS) put the total number of operating satellites at 1459. Of those, 180 were 10kg and under.

Basically, for the most part, there has been about 1100-1500 functioning satellites for roughly a decade or so, and 20,000 pieces of debris 10cm or bigger. The European Space Agency put the growth in debris in a GIF to make it easier for everyone:

Then, just three months later:

Here’s a couple of charts from St Louis University showing the growth in CubeSat launches alone:

Image courtesy St Louis University

You can see most of the drive comes from the switch to communications purposes since 2013.

We’re up to about 140 CubeSat launches this year. But Gunter’s ever-reliable Space Page has at least 395 more launches pegged for 2017 “with a real chance to get off the ground”.

If they do, we’ll have added about 30% of the total satellites in orbit around the Earth in 2017 alone.

We’ve only just begun

The University of Southampton’s Hugh Lewis led the first genuine warning cry about small satellites in 2014. At the time, Lewis and his team extrapolated a scenario in which 205 CubeSats were launched every year for the next 30 years. You don’t have to look too far ahead at the planned launches to realise the results are probably already irrelevant.

So this year, Lewis launched a new study to simulate 200 years of possible orbits for 300 “megaconstellations” – swarms of cheap CubeSats numbering in the thousands.

Boeing, Samsung and SpaceX are serious about those kinds of numbers, and for satellites much bigger than CubeSats.

Samsung has theorised it would need 4600 microsatellites. Boeing’s plan involves 3000 satellites, size unknown, and SpaceX founder Elon Musk has even gone so far as to apply for permission to launch 4,425 satellites the size of a Mini to blanket the Earth in high-speed internet.

OneWeb will set the next bar around this time next year, when it hopes to launch 648 small satellites to provide internet coverage, most of the capacity of which it says is already sold. Founder Greg Wyle claims OneWeb will add another 1972 small satellites to its swarm to meet the need it is seeing.

Remember, up until now, humans have got less than 1500 satellites of any size in total orbiting the Earth.

So let’s go wild and say even 20,000 satellites could be orbiting the Earth a few years from now. You’d probably pass that many cars on your way to work each week, on a single freeway.

There’s a lot more room in space, and a lot more levels to operate on. So maybe it’s understandable that we’re kind of casual about all the extra traffic on its way.

But consider this. What if in any split second, any two of those cars could become 2000 cars? Travelling at 25,000km/h?

And those 2000 cars each become 2000 more cars?


Picture: Warner Bros

You might have seen the 2013 Academy Award-winning movie where Sandra Bullock and George Clooney are left floating in space after debris wrecks their spacewalk outside the ISS.

If so, you’ll no doubt remember the hole in their unfortunate crewmate Matt Kowalski’s head. It was caused by a speck of space debris hurtling through it.

The debris was the result of the imaginary Russians blowing up an imaginary communications satellite in Low Earth Orbit (LEO). In Gravity, you get to see that debris turn parts of the ISS into even more debris, causing George Clooney to warn Sandra Bullock about even more debris when it comes back around the Earth in 90 minutes. That will in turn create even more debris and so on.

There was a bit of bad science in the 90-minute call, but the debris scenario is real. What was then called “the Gravity effect” is actually known as the “Kessler syndrome”, a scenario proposed by NASA scientist Donald J. Kessler way back in 1978.

Meir admit says “of course” having more objects in space raises the probability of a close encounter.

“However, there are a few things to remember,” he says. “Satellites’ orbits are deterministic, and we have very good knowledge of current orbits and very good predictions for the future orbits.

“(Their) orbits are separated in all four space-time dimensions, in different altitudes, inclinations and in time.”

If you modelled it realistically, from when the avalanche starts to when all the satellites are gone, would probably take weeks. You would lose everything.

Meir says SAS has an agreement with the US Department of Defense through the Joint Space Operations Center, which monitors all objects in space CubeSat size and upwards.

JSpOC can predict a collision 24-48 hours before it might happen, using a method called “conjunction analysis”, and notify asset stakeholders accordingly. Meir says the agreement also provides SAS with orbit maintenance capabilities.

That’s reassuring, but owning a solution for removing CubeSats from the race could be just as lucrative as being in it.

At its wildest end-point scenario, a Kessler syndrome event could make LEO untenable for all satellites. Like that proposed by Meir, there are arguments against this ever happening, but one Australian company isn’t prepared to to find out.

Dr Ben Greene, from the Space Environment Management Cooperative Research Centre (SERC) in Canberra, in 2014 said: “If you modelled it realistically, from when the avalanche starts to when all the satellites are gone, would probably take weeks. You would lose everything.”

In 2014, Greene, a space junk expert, set up SERC with $150 million coming in from the likes of the Australian federal government, the Japanese government, Lockheed Martin, NASA, Optus and the Australian National University.

Based at the Mt Stromlo Observatory, its mission is to find ways to manage the space junk. At the time, it received worldwide attention for working on a wild idea to “shoot” debris with lasers. The reality was the lasers would simply nudge smaller pieces into lower orbits where they would burn up.

At the time, Greene said we were “exactly on the curve that we predicted 10 years ago for the number of satellites we would lose in a year”.

He said if those losses climbed to “two of three times” that rate, the threat of an avalanche effect would suddenly come into play. At the time, people were talking of nanosatellites which “cost less than $1 million to build and launch”.

Right now, you could launch 50 CubeSats for $1 million. SpaceX, Boeing and Samsung alone “have plans” to multiply the number of larger satellites in orbit now – weighing up to half a ton – by 10 times. Not “two or three times”.

And if we’re being honest, we need to swap out “have plans” for “are in a race”.

Space is big

So when do we get alarmed?

When it comes to losing control of CubeSats, the University of NSW has had more experience than it would have cared for, and recently.

Here’s an animation of UNSW’s CubeSat UNSW-ECO launching from the International Space Station a month ago:

Even a decade ago, it’s the kind of experience students were aiming for as the peak of their future careers.

Their CubeSat joined an international program involving 27 countries launching 50 small satellites aiming to get a better understanding of the Earth’s thermosphere, about 100-1000km up.

The data they gather may one day help develop better weather prediction models, communication systems and understand the impacts of solar activity.

A week ago, UNSW had all but given up any hope of being part of it.

Along with two other Australian satellites on the trip, UNSW-ECO failed to ping back to its handlers after launch. For weeks, it sailed through space in silence, while UNSW’s Australian Centre for Space Engineering Research (ASCER) searched for it.

The problem was thought to be the satellite batteries didn’t have enough power to deploy their antennae, and what daily attempts they were making to do it drained whatever power had been gathered by their solar panels.

…intended collisions in space however, are a completely different story

ACSER had a program in hand to power the satellites down completely so they could get the full charge they needed to get the deploy-recharge cycle working efficiently. But they couldn’t communicate with the satellites in order to send it to them.

How they eventually solved the problem is quite an incredible story, and it involves a huge amount of luck, and some high-level radio and amateur astronomy nerdery from around the globe. You can read about it over at Gizmodo.

Yet even given his direct experience with an unresponsive CubeSat, relieved ACSER project technical lead Joon Wayn Cheong is confident all the right preconditions were in place to keep it safe.

“There have been significant efforts in recent research to study how to more accurately predict an object’s trajectory in space over very long periods of time,” Joon says.

“In my opinion, there has been sufficient barriers in place in the near future and plenty more emerging technologies under research and/or planned to be put in place such that the where and when of an unintended collisions in space can be accurately predicted and mitigated.”

Joon says many future nanosatellites will have in-built propulsion systems to orientate and re-position themselves, “away from a collision once notified”.

“In the case where the nanosatellites are inserted into an orbit that takes a very long time for it to decay, the same propulsion technology can also be used to de-orbit these nanosatellites at the end of its mission so that it can free up precious orbital slots for another use.”

He has faith in testing and validation procedures, and simply the fact that the cost and potential outcome of undertaking a nanosat enterprise is enough to ensure every precaution is made to keep it safe.

And consider this – right now, we know there’s approximately 22,000 pieces of space junk larger than 10cm orbiting the Earth because “larger than 10cm” is about the size limit we have the technology to track reliably.

The total number, if you add in the smaller pieces, could be up to 170 million pieces of debris in orbit around the Earth.

That sounds like a lot, but the reality is, spacecraft get pummeled by tiny junk all the time. Here’s a panel removed from the Hubble Space Telescope, showing all the debris impacts:

Orbital debris impacts. Picture: NASA/JSC

If you want to know what it sounds like, you can listen to audio from NASA’s Cassini as it passes through Saturns’s rings. These pops and crackles which are tiny particles bursting into clouds of plasma as they hit Cassini and its three Radio and Plasma Wave Science antennas:

The problem is, it’s almost impossible to quantify the level of danger space debris poses.

And as Joon says, “intended collisions in space however, are a completely different story”.

5mm warhead

“We had a situation where the Space Station was hit by a 5mm fleck of paint, they think, and it cracked one of the windows.”

David Ball works alongside Ben Green as deputy CEO at SERC, and has plenty of stories about how dangerous the millions of pieces of junk that are too small to be tracked can still spark a Kessler Syndrome effect.

“Potentially, we don’t know how much is out there, because no one’s tracking (pieces smaller than 10cm).

“We have one of the new Sentinel Earth observation satellites out of Europe, the solar array of one of those was struck by something and all of a sudden … they’ve lost power over the solar array.

“This satellite actually had a camera onboard and … they scrolled around and saw this big dent in the back of the solar array. And that can only be caused by one thing – it’s been hit by something and it’s caused the offset and the loss of power.”

Here’s the impact Ball is referring to:

Picture: European Space Agency

“The assumption these operators are making is that this is a clean physical environment they’re launching into. They spend a lot of time and money coordinating their frequency spectrum but they don’t spend a lot of time confirming the physical environment they’re going to launch into is clean.”

Ball says when we start talking about “megaconstellations” involving 4000 objects in different orbits, “your probability of collision is not zero”.

“There is a risk there,” he says, and one that isn’t being considered enough is that what goes up must eventually come down. De-orbits, be they after 12 months, five year, or 25 years, simply aren’t being considered as potential obstacles. And when we’re talking about 4000 satellites coming out of commission at once, they really should be.

The megaconstellations have to be replaced, because big business says they have to. And the new constellation has to be in place before the old one comes down.

“So the traffic management is the biggest issue you’ve got,” Ball says.

“Everyone assumes the big sky theory applies, and that worked for the aviation industry for the first few years and then we quickly found out that you need to have air traffic control.”

How do you find a speck in space?

Hitting things with lasers is easy. Japan’s space agency JAXA this month reported it successfully hit its Hayabusa2 craft with a laser from 6.6 million kilometres away, so objects orbiting the Earth are comfortably easy prey.

How to find a potentially catastrophic speck of paint is going to take some time, Ball says.

“That’s the main problem and part of the research we’re doing is to get to smaller and smaller granularity of detection and we’re nowhere near that yet in terms of the world’s capabilities.”

Using a high-powered laser and photon pressure to move debris in space is simply “the ultimate goal” and SERC has a couple of years left on that $150 million funding to run on that.

“In the last two years we’ve seen all the various disparate streams of research come together to culminate in some actual activity that could be tangible and that’s a little early to say when those events will occur,” he says.

“There’s a lot of steps to come before we put any power through the systems up to the satellites.”

But SERC is close to launching satellites itself. A dummy nanosat is pegged for launch by late 2018 with a demonstration phase beginning in 2019.

That will hopefully involves shooting ground-based lasers at them and measuring the effect it has on the satellite’s orbit.

Picture: Mt Stromlo Observatory

And it’s not like SERC are alone in their concern. Brian Holtz, the CEO of the OneWeb, told the BBC his “ambition” was “to set new standards in debris mitigation”.

“We’ve put extra hardware into the system to improve the reliability of that de-orbit process. We’re also committing to put a small adapter device on to each spacecraft that will allow those spacecraft, in the small probability that one of them dies on the way down, to be grabbed by a small chase vehicle and pulled out of orbit.”

OneWeb counts Richard Branson on its board of directors. Interestingly, Branson’s Virgin Galactic will also launch Meir’s larger constellation.

The European Space Agency, according to Ball, has also considered “using other satellites to cast nets or grab other debris”, but they’re all very early concepts, and it’s not as simple as it sounds.

“If you want to grab something you have to know where it is and how it’s behaving – is it tumbling, is it spinning, what’s the prediction on it, what’s the centre of mass – before you try to grab it,” Ball says.

“These things aren’t in stable orbit, they’re all over the place.

“You want to be ahead of the curve if you can. And space traffic management is something the world needs to look at, just as we had to get to where air traffic management got to decades ago.”

CubeSats with benefits

Space selfie! Picture: ARKYD/Kickstarter

Try, and you’ll never get across all the different experiments under way or proposed for nanosats. Roughly speaking, the majority are for experiments, particularly the effects of space on mechanics and biology, and monitoring things such as the weather and radiation levels.

There’s no denying it’s a fascinating and worthy field. Some examples include:

  • NEE-01 Pegaso – Launched in 2013 by the Ecuadorian Space Agency. The first CubeSat which successfully transmitted real-time video and broadcast it live over the internet.
  • GENESAT 1 – Launched in 2006 by NASA, GENESAT 1 is a $US6 million experiment to see how the space environment affected genetic changes in E. coli bacteria.
  • QUAKESAT – One of the first CubeSats, launched in 2003, set out to find a link between magnetic field fluctuations and the onset of earthquakes. KSAT2 is another climatology CubeSat aiming to better predict rain and tornadoes.
  • PLUME – The first English CubeSat was launched by student at the University of Leicester in the hope it can detect and examine cosmic dust particles.

And fun, which is important.

  • ARKYD – A Kickstarter funded project to get a mini space telescope into orbit and help in the hunt for asteroids. Blew past its $1 million goal with the promise of a “space selfie” for contributors.
  • STRaND-1 – The first CubeSat controlled by a smartphone (a Nexus One). It was supposed to play a selection of screams collected off YouTube, in order to test the Aliens tagline “In space, no one can hear you scream.”
  • TJ3Sat – The first CubeSat launched by high school students was a seven-year project out of Thomas Jefferson High in Virginia, US. Students could even text the satellite and get their messages transmitted back to Earth.

And there’s even experimentation to help solve exactly the problem Ball and his colleagues are working on:

  • PW-Sat – Poland’s first CubeSat launch in 2012 was used to test an atmospheric drag device which could pull other CubeSats out of their orbit and into a thicker part of the atmosphere where they could be decommissioned. NASA’s NanoSail-D2 successfully achieved this by deploying a solar sail which dragged it into the upper atmosphere and burned up.
  • ConSat-2 – Another Kickstarter project which aims to develop “self-healing” technology for satellites.

And the nanosat boom is already powering several successful startups right here in Australia and New Zealand.

Reward and risk

Gilmour Space Technologies are developing launch capability. Saber Astronautics offer ground control software and diagnostics.

New Zealand are way ahead, just last month launching a rocket with engines made from a 3D printer:

Their success is a huge kick in the pants for locals who may have been put off their space business dreams by the fact Australia is right up there with Iceland as the only OECD country without a space agency. And with the money – and capital – on offering for those who can own a piece of it, the only certainty ahead for the players is proliferation.

What “proliferation” means is perhaps the biggest uncertainty in all of this right now. Sky and Space’s Meir says a world in which every human had their own nanosatellite would be “a very interesting world”.

“Not so long ago, it was hard to imagine a world where every human will have personal computers, and today we have supercomputing power at the palm of our hands, with wearable computing becoming the new standard,” Meir says.

“As Yogi Berra said: ‘It’s tough to make predictions, especially about the future.'”

And ultimately, Ball, a veteran of the commercial satellite communications industry for 20 years, believes the nanosat movement is, “for the most part”, worthwhile.

“…the issue we see with the CubeSat push is that the CubeSats don’t have any propulsion, so they can’t position themselves to avoid collisions.”

“There’s some incredibly creative things being done at price points which were previously inaccessible to do different environmental experiments on different communications technologies,” he says. “But it is putting components up there which aren’t necessarily ‘space-rated’ components.”

By that he means in the rush to make Low Earth Orbit a lucrative business space, little consideration is being given to what impact things like temperature ranges and radiation issues are having on the work tools.

CubeSats aren’t just inhabiting Low Earth Orbits. Just this month, NASA announced three CubeSat projects had won a ride on its Orion rocket and will vie for a share of a $US5 million prize in the first-ever competition in deep space.

Some of the CubeSats will be deployed in deep space, 10 times the distance of the moon from Earth; some will orbit the moon.

And when the day comes – soon – when we’re talking in term of constellation launches in the thousands, as opposed to 5, 10 or 20 CubeSats, those orbits can be shut down in a matter of seconds, for everyone.

“It’s not a trivial problem to solve,” Ball says. “The larger sats have propulsion systems and they can manouevre so they will have obligations to de-orbit in under 25 years. But the issue we see with the CubeSat push is that the CubeSats don’t have any propulsion, so they can’t position themselves to avoid collisions.”

And once two CubeSats colliding become 4000 tiny pieces of CubeSats, “they’re all uncontrolled items”.

“The Kessler Syndrome… that’s something we have to avoid because we all use space day-to-day for things you just don’t think about any more, from turning your iPhone on and navigating to wherever you’re going and weather reports,” Ball says.

Ball says the ultimate worst case scenario is that a megaconstellation has a collision which pollutes the orbit they want to be in. That will mean no one can operate in that particular orbit “for a long, long time.”

“We’re talking decades,” he says.

“We of course keep our fingers crossed that never happens, but you can’t work on hope forever.”