Space junk: Catastrophe on the horizon

170815 spacejunk 2
An artist’s impression of the Kosmos 2251 and Iridium 33 collision, the result of too many objects in low-Earth orbit. – Stefan Morrell / National Geographic

Out in the blackness of space two astronauts are repairing the Hubble telescope and enjoying glimpses of their blue planet in the distance. At the same time, Russia decides to test an anti-satellite missile. Within minutes, the debris shower takes out the space shuttle and damages the International and Chinese space stations. Will the astronauts survive?

Many readers will recognise the plot of the movie Gravity. Fewer may know that on 10 February 2009, a real space nightmare came true. A satellite belonging to Iridium Communications relaying data to and from mobile phone users was passing over Siberia’s Taymyr Peninsula. At 16.56 GMT Iridium 33 was there; at 16.57 it was not. It had collided with a dead Russian military communications satellite. Weighing one tonne and travelling at a relative speed of 12 kilometres per second, the Kosmos 2251 satellite hit Iridium 33 with three times the kinetic energy of an Airbus A380. Both spacecraft disintegrated, scattering wreckage far and wide.

The loss of one of its 66 satellites was a blow to Iridium, but what caused sleepless nights for satellite operators and space agencies everywhere was the floating debris. The US Space Surveillance Network catalogued more than 2,000 fragments bigger than a grapefruit from the collision and  a much greater number of smaller ones. At the speeds required for low-Earth orbit, an object the size of a marble can cripple a satellite or punch a hole in a space station. The Iridium-Kosmos crash did not cause the kind of devastation depicted in Gravity. But there was a tense moment or two: the International Space Station had to make an evasive manoeuvre to avoid a piece of Iridium-Kosmos debris as did other satellites. Around the globe, space agencies suddenly woke up to the threat posed by the thousands of tonnes of objects – operational and dead – cruising in low orbit.

For Donald Kessler it was a Cassandra moment. He’d been predicting a catastrophe of this kind for more than 30 years. A former head of NASA’s orbital debris program, he foresaw that if we continued to launch satellites and leave their paraphernalia floating in orbit, collisions would be inevitable. Each could produce a debris shower capable of crippling other satellites in a chain reaction – a scenario that came to be known as the Kessler syndrome. If nothing is done our planet could end up ringed with a deadly debris belt spelling the end of the satellite age.

No one yet knows how to snare a piece of tumbling space hardware

and drag it towards re-entry.

Kessler’s warnings convinced space agencies to take measures such as directing defunct satellites and used upper stage rockets to re-enter the atmosphere and burn up. But this could be too little too late. Kessler and others believe that the Iridium-Kosmos collision signals that we have entered a new and more dangerous phase. A collision of this scale will likely happen more than once per decade – a Gravity scenario playing out in slow motion. If we are to avoid the Kessler syndrome there is only one solution: to hunt down the largest pieces of space junk and remove them from orbit at a rate of at least five objects per year for the next 100 years.

The problem is that no one yet knows how to snare a nine-tonne piece of tumbling space hardware and drag it towards re-entry. And the estimated cost of removing such objects is astronomical. Space agencies can be forgiven for not rushing to start the costly and unglamorous job of sweeping up our orbital back yard.

For Kessler and others, the space junk problem is akin to that of climate change. We’re dumping junk and storing up problems for the future; but scientists can’t say with absolute certainty that the dire warnings will come true.

So what do you do? Pay a fortune to tackle a problem that may or may not emerge, or do nothing and hope for the best?

Even a fleck of paint drifting through space could create havoc.

Back in the 1960s when Kessler started working for NASA, the big safety concern was space rocks – tiny micrometeoroids weighing less than a gram that drift through the Solar System. He tracked them using telescopes and radar; studied samples collected from the stratosphere by high-flying aircraft; and scrutinised returned spacecraft fro the damage they caused.

One discovery surprised him. Some of the damage to spacecraft had not been caused by micrometeoroids: it was the work of tiny man-made objects. Even a fleck of paint drifting through space could create havoc.

Kessler turned his attention to the debris problem and in 1978, he published a prophetic paper. A crowded low-Earth orbit would risk cascading collisions; the tipping point would come around the year 2000. “That set the stage for everything that’s happening today,” Kessler reflected.

170815 spacejunk 3
A piece of space debris left this hole in a panel of Solar Max, a satellite launched in 1980 to analyse solar flares. – NASA

But back in the 1970s, the threat wasn’t obvious. Kessler says he had trouble convincing authorities of the risks. “The Air Force didn’t believe it. They fought it for years.”

Ultimately, NASA listened. In 1979 they set up the orbital debris program with Kessler as its head. Its first goal: to calculate the size of the problem and ways to avoid making it worse. The US military’s North American Aerospace Defense Command (NORAD) already kept a catalogue of all objects in Earth orbit larger than 10 centimetres.

NASA’s debris program began tracking smaller objects with short-wavelength radar. They simulated what could happen in the lab, firing a hypervelocity gun at unused satellites to develop an impact model, for instance. Kessler’s back of the envelope calculations came up with similar results.

“He was very good at that sort of thing,” recalls Darren McKnight, a former NASA colleague of Kessler’s, now with Integrity Applications Inc.

Bit by bit, evidence mounted about the scale of the problem. In 1981 a Soviet military satellite quietly disintegrated a month after launch, presumably because of a collision. In 1983, a space shuttle came home with a millimetre-wide pit in one of its windows caused by a high-speed fleck of paint. “It was a slow process. Each new event added one more layer to the credibility,” Kessler says.

One of Kessler’s early discoveries was that much of the existing debris came not from satellites but from multi-stage rockets. When the bottom stage of a rocket uses up its fuel, it drops back to Earth and the next stage kicks in. The upper stage positions the spacecraft into its final orbit but once its job is done it is left to drift – often containing unburnt fuel that sometimes explodes.

NASA began making sure fuel was vented after upper stages had delivered their payloads and began steering them back towards re-entry where possible. Space agencies in Europe, Russia, Japan and China followed suit.

Defunct satellites posed another problem. In 2002 space agencies agreed to abide by a “25-year rule”. Satellite operators would have to ensure that upper stage rockets and retired satellites re-enter the atmosphere within 25 years.

Saving some fuel for a final downward burn or deploying a sail to increase drag would do the trick.

These fixes seemed to work. The growth in the number of trackable pieces of space debris slowed. The collapse of the Soviet Union in 1991 slashed the number of launches through the 1990s.

At the same time, the Sun was unusually active which heated the outer atmosphere, causing it to swell and increase the drag on debris in low orbits so that more re-entered the atmosphere. For a time it looked as though Kessler’s dire forecast would not come true.

Then on 11 January, 2007, in an event that could have inspired the movie, China carried out an anti-satellite test, firing a missile at its own Fengyun 1C weather satellite. The impact created around 3,000 pieces of trackable debris plus an estimated 150,000 pieces larger than a thumbnail – the most junk generated by a single event in the history of spaceflight.

That debris has the potential to do the sort of damage seen in Gravity, but in real life things happen much more slowly – space is, after all, a very big place. The anti-satellite test didn’t cause any immediate damage but the new swarm of debris made tracking and avoiding objects that much harder.

Then, two years later, came the Iridium- Kosmos crash.

170815 spacejunk 4
An artist’s impression of CleanSpace One, a Swiss spacecraft that will use a net to drag a nanosatellite towards re-entry. – EPFL / Jamani Caillet

“After those two events, the number of collision avoidance manoeuvres [by operational satellites] increased significantly,” says Jer Chyi Liou, chief scientist of NASA’s orbital debris program.

Something had to be done. The world’s top 13 space agencies asked the Inter-Agency Space Debris Coordination Committee (IADC) to investigate the future stability of the space environment. Six of the agencies – European, Indian, Italian, Japanese, British and American – had developed computer simulations of low-Earth orbit and they all put the same data into their models and made predictions, so that they could form a consensus view.

The six models agreed that the 25-year rule would not be sufficient to stop the increase in space debris. Even if 90% of launches complied with mitigation guidelines, the models predicted debris to grow by 30% over the next 200 years, mostly fed by collisions that would happen roughly every five to nine years.

Most of the mass of space debris still resides in large intact objects, so Kessler and others believe the best policy is to identify the largest objects in the most crowded orbits and take them down at  a rate of five per year for 100 years. “We need to put together a master plan, to change our operations to achieve a sustainable environment.

“The longer we put it off, the more expensive it will be,” Kessler says.

Removing another country’s satellite, even if
it’s dead, could be seen as a hostile act.

As yet, no one knows the best way to capture errant satellites and stages and put them out of harm’s way. Many proposals have been made but none have yet been tested in space. The cost of launching satellites is pretty steep already; the cost of finding dead ones and disposing of them safely could potentially be even steeper, especially if each retrieval mission can only take out a single object. As a result, space agencies are treading carefully, doing studies and experiments on the ground.

The technical challenges are many. Very few dead satellites and upper stages have fittings designed for retrieval operations, so the procedure will require equipment such as a grappling claw, a net or a harpoon. Junk spacecraft may also be spinning, adding further complications. “As of today, there is no economically viable or technically feasible method to allow us to do it,” says Liou.

Starting in 2009, the US military’s Defense Advanced Research Projects Agency (DARPA) invited aerospace companies to send in their concepts and held an international conference to sound out other ideas. DARPA’s report, known as the Catcher’s Mitt study, concluded that small debris (less than 5 millimetres across) could be dealt with by building shields around rockets and satellites and that large pieces (more than 10 centimetres) could be tracked and avoided. The analysis concluded the greatest threat to spacecraft comes from medium-sized debris, between five millimetres and 10 centimetres across. Because there is so much debris of that size and it is so spread out, the study could find no practical method of removing it. “The only feasible approach is to go after big things before they break into small things,” says Wade Pulliam who led the study for DARPA and is now with Logos Technologies.

Catcher’s Mitt advocated taking down between five and 10 large pieces of debris per year to stabilise the amount of medium-sized debris in orbit. It also examined the pros and cons of various removal technologies. But partly because the cost is so huge, Pulliam says we shouldn’t rush to get started.

“We have to start thinking about developing the technology but then we should leave it on the shelf. It’s not worth deploying it now.”

NASA agrees. “The sky is not falling, at least not in the foreseeable future,” says Liou. “We do have time to develop the technology … We don’t need to go out and remove debris in the next five years or so – 10 or 20 years, maybe.”

Kessler is concerned about the consequences of delaying until 2035. The 2009 IADC report proposed starting debris removal in 2020 but, he says, we’re no closer to doing that today than we were then. “It’s urgent to at least put together a plan for how to do this,” he says.

“Money is the crux of the problem,” says aerospace engineer Hugh Lewis of the University of Southampton in the UK. The issue has gone off the boil since no further collisions have taken place since the Fengyun and Iridium-Kosmos crashes. “The cheapest thing is to do nothing,” Lewis says. But this is not an option, he adds. “We should get ready, understand the probabilities, and be ready to go if we need to. ”

There are also thorny legal issues. According to international space law, responsibility for satellites in orbit rests with the country that launched them. Removing another country’s satellite, even if it’s dead, could be seen as a hostile act. Some might even view a system to remove satellites from orbit – especially one based on ground-based lasers [see box] – as a weapon.

But there are signs of progress. Researchers in Switzerland are planning to launch a small retrieval demonstration in 2018 called CleanSpace One.  A net will capture a Swiss nanosatellite and drag it towards re-entry. The European Space Agency (ESA) is also drawing up plans. Its researchers have been testing claws, nets and harpoons on the ground. “Each has pros and cons … there is no perfect solution,” says Luisa Innocenti, head of ESA’s CleanSpace office. The exact technology they will use is still taking shape, but Innocenti and her team are aiming to propose a mission costing several hundred million dollars and launching around 2021 to the ESA council when it next meets to decide budgets in 2016. Innocenti says this first mission will be ambitious, but the ESA has a good track record. “We have landed on a comet. I do not doubt that we will get there,” she says. “If we at least prove it can be done, it will be a message.”

How to clean up space junk

In 2009, the Catcher’s Mitt study from the US Defense Advanced Research Projects Agency (DARPA) examined proposed technologies for cleaning up space junk.

The study started with the premise that medium-sized space debris (between 5 millimetres and 10 centimetres across) is the biggest threat to spacecraft because it is too large to be shielded against but too small to be tracked and avoided.

The study estimated that around 20,000 new pieces of medium-sized debris are created per year, so regular removal would be needed.

Industry, space agencies and academic scientists put forward their plans for analysis by DARPA. The study’s conclusions are shown in bold below.

A number of the proposed solutions involved as yet untested technologies for launching spacecraft that don’t require fuel to chase debris, including solar-powered ion thrusters. This is because of the high cost of launching a spaceship with a heavy cargo of fuel.

There are also several other ways a device designed to capture space junk could do its job:

1 – Drag Enhancement. Attaching  a large sail or balloon to the debris so that atmospheric drag pulls it to re-entry. This method only works in lower orbits.

2 –  Solar Sail. Attaching a thin reflective sail that uses radiation pressure from sunlight to carry the debris out of orbit.

3  – Electromagnetic tether. Attaching a conducting cable several kilometres long to the object. By interacting with the Earth’s magnetic field, the tether will slow the object and take it out of orbit.

But a danger is that the tether could be cut by debris or become tangled if the object is spinning.

4  – Slingshot. A catching device fitted with a claw/net/harpoon captures the junk during a high-speed pass.

The connected device and debris spin around each other at the opposite ends of a tether. Releasing the debris at the right moment – a move like an Olympic hammer-thrower’s – will send the debris hurtling towards re-entry.

A high degree of accuracy and predictive modelling is needed for this to work.

The DARPA study


Numerous designs for orbit-cleaning ‘sweepers’ were proposed, which would absorb or simply slow down pieces of debris. Capture materials included multilayer foils, aerogel panels and layered open-cell foam. Passive sweepers cruise around debris-filled orbits catching whatever comes their way. Active sweepers identify debris and steer into its path.

Conclusion: Because debris is so widely scattered, sweepers would need to be vast in extent and launched in huge numbers, incurring enormous launch costs. Active sweepers would need large fuel supplies for frequent manoeuvres. High risk of collision with satellites or large debris. Currently impractical.


Firing a laser at debris causes some of its material to boil off at high speed, causing a thrust in the opposite direction. This can slow pieces of debris so they descend towards re-entry. Lasers could be launched into space or built on the ground and fired up through the atmosphere.

Conclusion: Debris is of irregular shape and possibly tumbling, so ensuring the thrust acts in the right direction is difficult. Powerful enough lasers do not yet exist and firing through the atmosphere bends the beam. Tracking is not yet accurate enough. Any laser system capable of downing debris would be seen as an offensive weapon by many countries. Currently impractical. 

Large debris removal

If removal of medium-sized debris is impractical then all we can do is remove dead satellites and rocket components to stop them becoming future debris. Any clean-up system would likely consist of a mothership that deploys devices capable of taking down a single object. The mothership would need to identify objects and manoeuvre close to them. It would need to grapple the object with a claw, net or harpoon and drag it down.

Conclusion: Identifying, manoeuvring close to and grappling an object are all untested challenges, especially if the object is spinning. But a bigger problem may be fuel. Launching enough chemical propellant for the mothership to chase multiple objects as well as the boosters to bring the objects down would be very costly.

Please login to favourite this article.