By Arthur I. Zygielbaum
At the end of February 2008, there was a brief flare-up of stories about a spy satellite about to fall from the sky. The stories ended almost as rapidly as the satellite, after it was pulverized by a Navy missile. While that story faded, it is important to note that there are tens of thousands of objects in earth orbit. Some are useful—like communications satellites. Some are not—like thousands of frozen drops of nuclear reactor coolant leaking from Soviet-era satellites. Unless they are in extremely high orbits, all of these objects will eventually return to earth.
“So what keeps them up there?” The question was posed by Vice President Richard Nixon during a briefing about Soviet satellites to my father, an employee of the Jet Propulsion Laboratory, now one of NASA’s national laboratories, and a member of the briefing group. A half century later, I was asked, “Why and where do they fall down?”
First things first. Why do they stay up there? The classical explanation is fairly simple. Imagine a cannon. Load the cannon and fire a cannonball toward, say, the east. It falls back to earth due to gravity. Suppose you add some more gunpowder to the cannon. The cannonball goes a little farther. More gunpowder, more distance. Now add a whole lot of gunpowder. Eventually the cannonball falls and there is no earth to catch it. The cannonball literally falls off the ends of the earth and, given enough powder, travels out into space with its path modified by earth’s gravity. But at a magic point, the cannonball will not escape into space, but will continue to fall around the earth. It will be in orbit. The speed needed for orbiting the earth—that is, continuously falling off the edge—is about 17,000 miles per hour.
When you spin a ball at the end of a string, the string becomes taut. The ball orbits your hand. The force of the tension on the string keeps the ball from flying away. The ball’s momentum—its attempt to keep going away from you in a straight line—keeps it from coming back to your hand. The mechanics of the satellite’s orbit is just the same. The momentum that causes it to want to leave the earth is exactly balanced by the pull of the earth’s gravity. That’s what keeps the moon from leaving or crashing into the earth and the earth from crashing into or leaving the sun.
OK. I’m being a little simplistic. The moon is actually very slowly leaving the earth. It has a little more momentum than can be overcome by the pull of the earth. According to measurements made using mirrors left on the moon by Neal Armstrong and Buzz Aldrin, the moon is leaving the earth at the rate of about 1.5 inches per year: long time, then no see.
Artificial satellites (that are not in extremely high orbits) tend to fall back to earth. Once in orbit, the actual orbit or path a satellite follows is close to a constant ellipse (including a circle). But there are things that keep that path from being perfect and exactly repetitious. The gravity of the earth is not uniform. It varies depending on how massive quantities of iron and other dense materials are distributed under the earth’s crust. The sun’s gravity affects the orbit very slightly. The sun has another influence. It puts out streams of particles such as photons of light, electrons, etc., that can hit a spacecraft and impart a very small force. This stream is not constant. It varies. While always weak, sometimes it is somewhat less weak.
As an aside, that force can be significant enough to control the attitude of a spacecraft, or if you build a large enough sail, you can use the particles to “sail” to another planet. Back in 1973, the Mariner 10 spacecraft flew to Venus and Mercury. As it traveled, the spacecraft’s active attitude control system, consisting of small rocket motors, developed a leak. Clever engineers were able to adjust the position of the craft’s solar panels so that the stream of solar particles imparted the right force to help control the spacecraft. But for this intervention, so much gas would have been lost by using the little rockets that the mission would probably have been a failure. As it was, with the sun’s cooperation, the space probe was kept pointed in the right direction.
The biggest effect on an earth satellite’s orbit is the atmosphere of the earth. The atmosphere causes friction. It’s the same friction that heats up the Space Shuttle as it reenters the denser parts of the earth’s atmosphere. Wait a minute. We’ve all been told that space is a big vacuum. But it’s not a perfect vacuum. The earth’s atmosphere doesn’t just end and space begin at some point. The atmosphere just fades away as you go higher. Near the earth, there is always a tenuous atmosphere to contend with. So why not just put satellites high enough that there isn’t enough of this tenuous collection of gases to affect the satellite much? One problem is that satellites are expensive to launch. The higher, the more costly. The instruments on the satellites are also selected to perform some function, like map the earth, and can’t be too far away to do their job well. So engineers and managers select the optimal altitude their spacecraft needs—balancing scientific or observational need and the cost of getting the thing up there.
So the spacecraft is launched to an altitude above an appreciable part of the atmosphere. In an ideal world (and universe), the atmosphere would be well understood and constant. If you put the spacecraft in an orbit 1,000 kilometers up, you would expect to be able to compute what the drag of the atmosphere is on the satellite. That way one can predict that the spacecraft would be in a stable orbit for a long period of time—years, decades or centuries. The problem is that the height of the atmosphere changes. The same solar wind that saved Mariner 10 pushes on the atmosphere. It can cause bumps in what might be the perfect atmospheric globe. The size of the bumps varies with the strength of the varying solar wind. Satellites run into those bumps. It slows them down. When they slow down, they are like the cannonball without enough gunpowder. They eventually fall down.
So while good engineers and scientists can make a best guess at the lifetime of a satellite, the vagaries of the solar wind and the changes in the height of the atmosphere can play havoc with the best of estimates. Putting it all together, the atmosphere drags on satellites. It slows them down. They slowly fall to earth until they enter the denser part of the atmosphere (about 50 miles up). At that point, the satellite begins to really slow down and is heated to very high temperatures. Large satellites typically break up and burned pieces of material can hit the surface of the earth—most likely the ocean, just because there is more ocean than land.
The shape of the spacecraft also has an effect. I’m sure we’ve all flown our hands by sticking them out of a car window. If you put your hand flat against the wind, there is obviously a lot of drag. If you put your hand edge-wise, there is minimum drag. In between, your hand is lifted or pushed down. It’s like that for a satellite reentering the earth’s atmosphere. If it is big and has lots of things attached, it will slow faster than a satellite that is pointy and sleek. The satellite may break up. Lighter pieces might slow faster. Parts, like those solar panels, might fly a bit or flutter and move away from the heavier parts. As the satellite burns and breaks up, its parts are scattered.
Because of the dynamics of the atmosphere, the shape of the vehicle and the unknown orientation of the vehicle as it enters the atmosphere, predictions of impact points are difficult. Typically there is a 10 percent uncertainty just in the time of reentry. For a vehicle orbiting the earth every 90 minutes at a speed of 17,000 mph, that translates to a 5,000-mile uncertainty in the impact point.
Satellites have been unceremoniously returning to earth since the dawn of the space age. Over 100 metric tons worth of satellites, spent rocket stages and other man-made junk fall back each year. For the most part, these uncontrolled
objects burn up in the atmosphere and nothing substantial hits the earth’s surface. There are exceptions. Perhaps the most famous was the 1979 reentry of SkyLab over Perth, Australia. Many pieces of that spacecraft were collected by the residents. Fortunately, no damage was done to property and no one was injured.
Agencies launching spacecraft have learned how to de-orbit them in a controlled manner so that any surviving pieces would very likely land in an ocean. This requires that sufficient fuel and steering capability are available at the end of the satellite’s mission. By carefully tracking the satellite and firing thrusters at appropriate times, the exact point of reentry and even the trajectory once the satellite reenters can be controlled. When the Soviet Mir space station fell to earth, controllers were able to slow the station at the right point so that surviving pieces impacted the Pacific Ocean. The key is that the satellite must still be functioning and under control to accomplish this feat.
The International Space Station (ISS) slowly falls to earth because of atmospheric friction. Occasionally, its orbit is raised by a “re-boost.” Rockets on a docked Russian cargo vehicle, or the Space Shuttle, can be used to give the station a little push to move it to a higher orbit.
While a member of the congressionally chartered NASA Aerospace Safety Advisory Panel, I suggested a strategy for changing crews on board the space station that would involve leaving ISS unattended for several days. My reasoning seemed sound. This was during the time that the Space Shuttle was grounded after the Columbia accident. There is always a Russian Soyuz spacecraft docked to the ISS. It is there as an emergency escape for the crew. It will accommodate three people. A new Russian Soyuz spacecraft was going to be launched with a replacement crew. The plan was that after that Soyuz docked and after all the hugs and information exchanges, the old crew would take the old Soyuz—their escape vehicle —and return to earth. What if that older spaceship didn’t start? Then two crews would be on ISS with an escape vehicle capable of only returning one of them to earth. Not good. My suggestion was to have the current crew use their escape Soyuz and return to earth. If it didn’t work, than an unmanned Soyuz could be flown to ISS for them. If it did work, then the new crew would fly up to ISS with the new Soyuz. During the interim, the station would have no one on board.
NASA safety experts analyzed my proposal and rejected the concept. If the space station was unattended and its attitude-control system failed, it would drift in its orientation. It might even begin to spin like a massive top. There is no way that a Space Shuttle or Soyuz could dock with the celestial merry-go-round. That would mean that there would be no way to provide a re-boost. And a machine with the interior space of a Boeing 747 would fall to earth.
Last February, a spent Defense Department spy satellite the size of a school bus and weighing 9,000 kilograms (nearly 20,000 pounds) was about to rejoin the earth. The power to control the spacecraft had failed. Although much of the satellite would burn away during reentry, large chunks would remain intact. Along with the dangers inherent in big things falling from the sky, the chunks would probably have included a tank with a significant amount of highly toxic hydrazine rocket fuel. To mitigate the danger, the U.S. government intercepted the satellite with a missile. Because of the very high closing velocities, the missile’s relatively small “kinetic” payload hit the satellite with a force equivalent to many tons traveling a thousand miles per hour. This broke up the satellite and dissipated the hydrazine into space and the upper atmosphere.
Was it worth the approximately $30 million cost of shooting the intercept missile? What if no missile had been fired? Since three-fifths of the earth is covered in water, and because most land mass is sparsely inhabited, the chances are extremely low that any lives or property would have been endangered from the remains of the satellite. The Aerospace Corporation estimates that the chance of someone getting hit by a part of a satellite is one in a trillion. (The chance of being hit by lightning is about a million times greater!) But if something large or poisonous hits a populated area, it could be, in NASA parlance, a bad day.
While there is significant planning on how to get a satellite up there, unfortunately, there is frequently insufficient planning on how to bring it down safely. And, as we started this narrative, they all fall down.