Last June, the Compton Gamma Ray Observatory orbited Earth for the final time, ending—prematurely, in my view—nine productive years of service to science. For nearly a decade, astronomers used this instrument to better the understanding of the universe in significant ways. They could do so because gamma rays probe the most extreme conditions of temperature, magnetic-field strength and gravity—conditions that arise near the weirdest of astrophysical objects: neutron stars, black holes and supernovae, for example. (See "The Gamma-Ray Universe" by Donald A. Kniffen, July–August 1993.)
Unlike ordinary stars, which give off mostly visible light, almost all gamma-ray sources vary in brightness and are, in one way or another, unusual. Were human eyes sensitive to these ultra-short wavelengths, every night or so a blindingly bright flash—a gamma-ray burst—would erupt overhead. Measurements from the Compton satellite showed that these events begin with extremely powerful explosions in distant galaxies.
But gamma-ray eyes would reveal more than just bursts. Within the Milky Way, one would see a sprinkling of pulsating stars. These emanations come from rapidly spinning neutron stars called pulsars— rotating beacons that periodically send energy in the direction of the earth. Observations from the Compton Observatory doubled the number of known gamma-ray pulsars.
Also along the Milky Way one would see dozens of gamma-ray sources changing irregularly in brightness. Once every now and then a gamma-ray source brightens so markedly that it outshines all others for some days to months, then fades again. These astrophysical curiosities are powered by neutron stars or by black holes. And, as the Compton satellite revealed, spread more or less evenly over the entire sky are other variable sources of gamma rays. Astrophysicists believe they are galaxies harboring massive black holes, which swallow up suns like so many raisins and send out immense jets of matter and energy.
The list goes on. Since 1991, teams of scientists using measurements from the Compton Observatory have announced scores of important discoveries. Something like half of them were made after the originally planned five-year mission of this satellite. But with such a track record of success and with most of the science instruments operating flawlessly, NASA controllers raised the orbit of Compton not once but twice, and it was not in danger of re-entering the atmosphere before 2003 at the earliest.
What happened? In short, NASA officials decided that they needed for safety's sake to crash this scientific satellite, although it was circling in a stable orbit and continuing to return valuable measurements. The long story is, of course, quite a bit more involved.
In early December 1999, after acting erratically for several months, one of three gyroscopes on board the Compton Observatory failed. These gyroscopes control the orientation of the craft. Weighing in at 17 tons, Compton was one of the most massive satellites ever launched, and NASA officials knew that, when it finally plummeted to Earth, many large pieces would survive the fiery re-entry.
At a press conference convened last March, NASA managers cited a one-in-a-thousand chance of someone being injured or killed during an uncontrolled re-entry were the Compton Observatory not brought down before another gyro failed. So, they explained, swift action was needed. This sense of urgency presumably accounts for NASA's failure to consult its own advisory committees or the international partners who had contributed instruments to the observatory.
NASA officials just went ahead and charged engineers at the Goddard Space Flight Center with finding a way to control the massive craft should another gyro fail before the re-entry could be organized. These engineers, aided by the builders of the satellite, did a fantastic job. In fact, they found a way that the Compton Observatory could be controlled without any gyroscopes at all. So NASA managers had two options: They could scuttle the satellite as quickly as possible, or they could continue the mission on two gyroscopes and use the gyroless steering tactic if need be. They estimated that the odds of someone, somewhere being harmed if they brought the craft down using the two working gyroscopes at one in 29 million. They figured that the risk for the alternate plan was about one in 4 million.
This latter estimate was somewhat uncertain, because the engineers were not given the time they needed to evaluate the chance that the gyroless steering mode would become unworkable through the failure of a critical subsystem. Nor were they given an opportunity to estimate the likelihood of such a failure precipitating an uncontrolled re-entry. At the March press conference, little was said about these two options and the comparatively small risks they entailed. Rather, the focus was on the risks involved with an uncontrolled re-entry. In any event, NASA officials ignored a third strategy entirely.
Even if all systems on board had failed, Compton would not have begun to tumble wildly. Rather, it would have settled with its heavy end pointed toward Earth. The satellite was equipped with a grapple fixture so that the Space Shuttle could retrieve it. This option was obscured at the press conference. Officials just pointed out that the gyroscopes were not designed to be serviced—conveniently ignoring episodes where astronauts have grabbed spinning satellites in their hands, have repaired satellites not designed for service in space and have even made impromptu fixes to broken equipment. Managers at NASA simply dismissed the possibility of retrieval and repair, saying it would take too long and be too risky for astronauts—assertions that are difficult to justify given that the Compton Observatory would have remained safely in orbit for at least three years, allowing astronauts ample time to plan and to train.
At the press conference announcing the decision to scrap the satellite, Edward Weiler, NASA associate administrator for space science, said, "There can be no trade-off between science and human safety." But this is exactly the trade-off NASA makes every time it launches anything. Indeed, in other contexts NASA appears perfectly willing to endanger the public to a small degree in the pursuit of science. The Galileo probe is a prime example. On its way to Jupiter, this spacecraft had to swing close to Earth two separate times. The late Carl Sagan calculated a chance of about one in 2 million of someone dying from Galileo plummeting to Earth during a fly-by, burning up in the atmosphere and spreading its on-board store of plutonium.
And NASA routinely accepts some even-greater dangers. For example, the agency often uses Delta II launch vehicles, and they let the expended second stages of these rockets just drop from the sky. NASA engineers estimate that each second stage carries a greater than one in 10,000 chance of killing someone when it eventually comes down. In fact, large pieces of these second stages have hit populated areas recently. One fell in Texas and one in South Africa. Both narrowly missed landing on houses.
In calling the press conference last March, NASA led everybody who drove to the session to take a personal risk of similar or greater magnitude to that which the spacecraft had posed to anybody living under its path. I'm no rocket scientist (just an astrophysicist keen to continue learning about gamma-ray sources), but I'm having a hard time seeing the logic. Indeed, I remain bewildered by the lack of a clear and consistent explanation for how and why the decision was made. In fact, the Compton Observatory is not an isolated example; NASA must make these sorts of decisions all the time. Still, if the agency cannot give a candid and logical accounting for its actions, it risks losing the confidence of both the scientific community and the general public.
© Stefan Dieters