It's happened only three times, as far as we know: three occasions when two sizeable objects in Earth orbit have accidentally collided at high speed. The first, or the first to be clearly identified, was in July 1996, when a fragment from a long-disintegrated Ariane rocket stage hit a French military satellite called Cerise at nine miles a second. The impact cut in half a 20-foot boom extending from the spacecraft, thereby adding another bit of junk to the 12,000 pieces currently being tracked in space. Cerise kept operating, which was lucky. Lab tests conducted last year in Japan, of impactors striking smaller satellites at much slower speeds, smashed the targets into smithereens.
Another collision happened five years earlier, although it wasn't understood until recently. A defunct Russian navigation satellite (Cosmos 1934) ran into a fragment from another Cosmos while both were circling in similar 600-mile-high orbits. Tracking networks soon noticed two new pieces of debris produced by the collision.
The third occurred in January 2005, when a fragment from a Chinese rocket struck a U.S. rocket stage that had been orbiting for 31 years, breaking off three pieces of debris from the larger object. Lucky again.
Most experts on space debris say that in terms of both the frequency of these events and their severity, our luck will soon run out. Collisions among pieces of debris in space beget more debris, and certain orbits—particularly the polar orbits favored for weather satellites and other Earth observers—will eventually become hazard zones. It's not an emergency yet, says Nicholas Johnson, the chief scientist at NASA's orbital debris program office at the Johnson Space Center in Houston. But it's time to act.
In an influential article published in Science magazine in January 2006, Johnson and his NASA colleague Jer-Chyi Liou argued that only "the removal of existing large objects from orbit" can keep the problem from getting worse.
The scientists used computers to simulate the proliferation of debris in Earth orbit over the next 200 years, assuming no more satellite launches—a hypothetical best case. Their models predict that, up until 2055, the creation of new debris from collisions will be balanced by the disappearance of old junk, which burns up in the atmosphere as its orbit decays. After 50 years, though, as more collisions occur, the creation of debris will start to predominate. The simulations predict 18 collisions over the course of 200 years, each yielding hundreds or thousands of fragments that exacerbate the risk. Even if we never launched another satellite—and of course we will—Johnson and Liou wrote, "The current debris population in [low Earth orbit] has reached the point where collisions will become the most dominant debris-generating mechanism."
Although the scientists conclude that it's time to start cleaning up, they also acknowledge that "no single [cleanup] technique appears to be both technically feasible and economically viable."
Two years later, that's still the case: No one knows how to begin removing orbital debris. "No easy or cheap solutions have yet been identified," Johnson says flatly. It isn't for lack of ideas. Well-meaning inventors have come forward with all kinds of schemes for clearing out space junk: space flypapers, sweepers, robot garbage scows. Take, for example, U.S. patent no. 4,991,799, filed in 1990, for a propeller-like sweeper that would ram into small particles and knock them from a threatening orbit. Or patent no. 6,655,637, filed in 2002, for a robot that could grab space junk with "inflatable fingers."
"Some of the ideas are technically outlandish, some are technically feasible," says Johnson. The problem, almost always, is cost. "If you want to spend tens of millions to retrieve a single rocket body, you can do it," he says. "But it doesn't make any sense economically."
So Johnson and other debris experts from Europe, the United States, and Japan are working on a comprehensive study for the International Academy of Astronautics that will evaluate cleanup options. Results are due next year.
The ideas tend to come in two sizes: systems for clearing out particles smaller than 10 centimeters (four inches), and schemes for "de-orbiting" large objects like whole rocket bodies, usually by pushing them down to lower orbits, where they burn up due to increased atmospheric drag.
Fragments between 1 and 10 centimeters in size will penetrate most spacecraft, according to the Aerospace Corporation's Center for Orbital and Reentry Debris Studies, and more than 100,000 are estimated to be circling Earth. (Pieces even smaller than a centimeter can cause damage, as NASA space shuttle managers know; they've had to replace more than 60 shuttle windows, dinged by tiny particles.) In the 1990s, NASA and the U.S. Air Force Space Command studied a concept called Orion, sometimes called a "laser broom," designed to eliminate small debris. A ground-based laser would be aimed at each object until pressure from the beam, coupled with the reaction force from material ablating away from the target, sends it into a lower orbit.
Orion, though, "turned out to be not all that easy technically," says Johnson. And with an estimated cost of $500 million, "it was certainly not within anybody's budget." The system would have required its own tracking network, since current space surveillance cameras track objects only down to 10 centimeters. Engineers would have to work out a system that imparted enough momentum to move a chunk of debris and that would be sure to lower instead of raise its orbit. "There are lots of little gotchas in the Orion final report," Johnson says. "There's a reason why it's been sitting on the shelf for more than a decade."
Okay; can something be done about the bigger pieces? Over the long term, removing large objects like empty rocket stages is the most effective way to reduce the likelihood of collisions, since such objects account for most of the total surface area that could be hit by smaller pieces. According to Johnson and Liou's calculations, de-orbiting just five large objects a year could reduce the overall collision risk significantly, and help stem the proliferation of junk. Removing 20 objects a year would reduce orbiting debris from 65,000 pieces by the year 2200 (if we do nothing) to just 18,000—not much more than what exists today.
One way to bring down a large piece of junk is to attach a small rocket engine and fire it toward Earth. Rocket engines are expensive, though, so space junk junkies are turning their attention toward tethers—among the cheapest methods of space propulsion.
Tethers sound impossibly great, like perpetual motion machines. Simply by attaching a single thin wire, which extends downward from a satellite for several miles, you can lower its orbit. No fuel required. As the wire drags through Earth's magnetic field, it generates a current, which acts as a natural brake against the orbital motion. Or, attach a "momentum tether" to a spacecraft, and it can be flung from one orbit to another, even from one tether to another, like Tarzan swinging from vine to vine. Again, no fuel required. Several experiments have already flown, on both U.S. and European spacecraft.
There's just one trouble with space tethers. "They all fail!" Johnson laughs.
Despite the well-understood physics and the "no reason this shouldn't work" assurances from proponents, something always seems to go wrong. Last year alone, two space tether experiments returned the kind of mixed results that have become frustratingly common for the technology. In April, the Multi-Application Survivable Tether, built by Tethers
Unlimited, Inc. of Bothell, Washington, got hung up while one tether-deploying satellite was separating from a companion satellite. In September, a tether called YES2, built by European students, apparently extended to its full length of 18 miles in orbit, a new world record. But only apparently. The YES2 team had to piece together what happened, because the satellite attached to the end of the tether disappeared and hasn't been heard from since.
Rob Hoyt, president and chief scientist at Tethers Unlimited, admits that the glitches and half-successes haven't exactly inspired confidence. Yet he is not alone in believing that tethers will, after more flight experiments, eventually be certified for real work, including debris removal. The Japanese space agency JAXA is working on designs for a small satellite that could attach a tether to a piece of space junk to remove it from orbit. A tether test is planned in space, says project engineer Satomi Kawamoto, although no date has been set.
Johnson agrees that tethers should be able to handle the de-orbiting job, and says their first use may be as kits attached to new satellites as a means to dispose of them safely at mission's end. Hoyt's company has just such a system in mind—the Terminator Tether—which he hopes could be priced at under half a million dollars.
Installing a tether on the ground before launch is one thing. Attaching a tether, or any kind of de-orbit package, to a crumbling, tumbling rocket stage in orbit is another—particularly if the target has nothing for the package to hold on to, because it wasn't designed to be touched ever again by human or robot after being put in space.
JAXA engineers are looking at several options for wrangling such uncooperative targets. In one scheme, the junk removal satellite would dampen any tumbling motion of a large object by shooting small projectiles at it—ice pellets work nicely—before moving in to attach a tether.
In 2003, Tethers Unlimited designed a system called GRASP (Grapple, Retrieve, And Secure Payload), which used a net made of Kevlar yarn to snare a small object and steady it enough for a tether to be attached. With funding from the Defense Advanced Research Projects Agency, the company got as far as testing a prototype during short stretches of weightlessness on a zero-G airplane flying parabolic arcs. It worked, says Hoyt, but DARPA hasn't come through with money for a follow-on test.
And that, as usual, is the rub. No public or private entity has volunteered to research, let alone build, an operational junk removal system.
One would expect satellite owners and insurers to take an interest, but relatively few satellites in polar orbits are privately owned, and assigning blame for orbiting debris collisions is still a fuzzy area of the law. Who was at fault, asks Johnson, when an old rocket fragment and the Cerise satellite ran into each other in 1996? Neither object had the ability to maneuver, and the fragment had been orbiting for years before Cerise was launched. "There are no rules of the road in space," he says. And because collisions have been extremely rare, insurance premiums haven't risen enough to press satellite owners into doing something about the debris problem.
And so the junk multiplies. The last year alone has seen the two worst space-junk-producing events in history. Last February, leftover propellants in a two-ton Russian Briz-M rocket stage caused it to explode, producing about 1,000 new pieces of orbital debris. Fragments from rocket stage explosions make up the majority of space junk, and most satellite-launching nations have now learned to vent leftover propellant from used rockets so they don't turn into orbiting bombs like the Briz.
But the worst case of orbital littering ever was deliberate. On January 11, 2007, as part of an anti-satellite test, China slammed some kind of impactor into its own Fengyun 1C weather satellite. By year's end, 2,500 pieces of debris were being tracked. Last summer one of those speeding fragments forced NASA to hastily move its $1.3 billion Terra Earth-viewing spacecraft out of the way. These kinds of evasive maneuvers are no longer uncommon for large spacecraft, including the International Space Station.
Expect to see satellites undertake more and more debris-dodging maneuvers. And hope that somebody finds a way to lower the cost of building and deploying space junk collectors—in case those satellites can't dodge quickly enough.