George W. Bush’s January 14 speech at NASA headquarters, in which he set the manned space program on a new trajectory, was an oddly dissociated event. NASA administrator Sean O’Keefe stood alone at stage left with his arms hanging limply at his sides and his fingers curled, looking like an eagle that has just eaten a gratifyingly plump mouse but is having trouble digesting it. The President, adopting his customary tank-window squint, briefly praised shuttle astronauts for conducting “important research” and helping to build the International Space Station—and then enthused about the “stunning images” from NASA space telescopes and the investigations being conducted by its probes of Mars, Jupiter, and Saturn. The odd thing was that aside from Bush’s tip of the hat to the shuttle and the station—whose death warrants he was signing—all the triumphs he cited were the work of unmanned robotic spacecraft.
Which pretty much reflects the problem. NASA is two agencies—three, if you count aerospace—in one. Its unmanned programs are flying high: robotic probes have sampled the sands of Mars, mapped every planet in the solar system this side of Pluto, inspected comets and asteroids, photographed infant galaxies near the edge of the observable universe, and made incalculable contributions to terrestrial communications, agriculture, geology, and weather forecasting, all at a fraction of the cost of sending astronauts up there. Meanwhile, the manned program is stuck in low Earth orbit. As Bush noted, “In the past thirty years, no human being has set foot on another world, or ventured farther upward into space than 386 miles—roughly the distance from Washington, D.C. to Boston, Massachusetts.” It is as if sixteenth-century Spain, three decades after Columbus, lacked a single ship capable of venturing out of sight of land.
Can the Bush plan get manned spaceflight going again—and should it?
Bush was amply justified in deciding to retire the shuttle, which despite all its merits—it is, after all, the world’s only winged, reusable spaceship—never had a raison d’être and had become both an emblem and a cause of NASA’s woes. Dreamed up as a kind of hangover cure in the days following the Apollo lunar missions, when the NASA budget was shrinking from over 4 percent of the federal budget to its current level of under 1 percent, the shuttle was sold to Congress as a cost-effective way of putting humans and satellites in orbit. Taken in by NASA hype, President Nixon assured the nation that “a space vehicle that can shuttle repeatedly from Earth to orbit and back…will revolutionize transportation into near space, by routinizing it,” and President Reagan declared, following three test flights of the first space shuttle, Columbia, that shuttles were now “ready to provide economical and routine access to space.”
This was sheer fantasy, as NASA was in a position to know and ought to have admitted. Economical? The shuttle substantially raised the costs of putting vehicles in orbit, rather than reducing them: at well over $300 million per launch, it costs $10,000 per pound for the shuttle to deliver a payload into orbit—two to five times the going commercial rate. Routine? Combine two of the space agency’s own predictions—that the shuttle would fly almost weekly, which never happened, and that it would realize a 98 percent safety record, which turned out to be about right—and you’re looking at a shuttle crash every year. To have funded that sort of prospective carnage indicates that neither NASA nor the Nixon White House believed the forecasts on which they were basing their decisions.
The process of purveying such claims sent NASA down the slippery slope of believing its own press releases, a degeneration remarked upon by both the Rogers Commission, which investigated the explosion of the shuttle Challenger seventy-three seconds after launch on January 28, 1986, and the Columbia Accident Investigation Board, convened after that shuttle disintegrated while reentering the atmosphere on February 1, 2003. The Columbia board, chaired by Harold Gehman Jr.—a sixty-year-old retired four-star admiral given to the unrelenting pursuit of hard facts—has produced a clearer and more coherent report than did the Rogers Commission, but it makes an even more depressing read since so little at NASA seems to have changed during the seventeen years separating the two accidents. It concludes that the Columbia crash was “not a random event, but rather a product of…a series of political compromises that produced unreasonable expectations—even myths—about its performance.” NASA bought into these myths, at one point boasting that the shuttle was “the most reliable, flexible, and cost-effective launch system in the world” when in fact, according to the report, it was and still is “a developmental vehicle that operates not in routine flight but in the realm of dangerous exploration.”
The proximate causes of Columbia‘s demise were not particularly difficult to discern, although the Gehman board took pains to verify them and to analyze competing hypotheses. In essence, two things happened. First, eighty-two seconds after launch, a 2.7-pound chunk of insulating foam broke loose from the shuttle’s fuel tank and knocked a hole in the leading edge of its left wing. Second, although NASA soon became aware of the incident, it took inadequate steps to determine the extent of the damage and did nothing at all to protect the astronauts against what proved to be its lethal effects.
At launch, the space shuttle system consists of a reusable spaceship, or “orbiter,” about the size of a DC-9, attached to the side of a gigantic fuel tank, 154 feet long and 28 feet in diameter, containing the liquid hydrogen/ oxygen fuel that powers the orbiter’s engines. (Additional power is provided by a pair of solid rocket boosters, of which the malfunction of one blew up the Challenger.) Because hydrogen and oxygen have to be kept extremely cold in order to remain liquid rather than turn into vapor, the tank is covered with a layer of insulating foam. Keeping the foam glued firmly to the freezing, sweating tank had long been a headache for NASA engineers, and on many occasions pieces of it had come loose, hit the orbiter, and done damage. The worst such debris strike occurred during a launch in December 1988, when the shuttle Atlantis sustained a flabbergasting 707 foam hits. Mission commander R.L. “Hoot” Gibson inspected the damage with a video camera attached to the shuttle’s robotic arm. (Columbia had no such arm.) “It looked like we had been shotgun-blasted,” he recalled. “I looked at those pictures and said, ‘We are going to die’ to myself.”1 Ground control pronounced the damage acceptable and Atlantis landed without incident. It turned out to have lost a heat-protection tile, but fortunately an aluminum plate happened to lie beneath the tile. Otherwise, in Gibson’s estimation, his mission might not have made it home.
The very fact that so many shuttles survived foam strikes evidently lulled NASA into underestimating the danger. Although photos showing the debris strike during Columbia‘s final launch raised concerns on the ground just hours into the mission, the Gehman board found that NASA “declined to have the crew inspect the orbiter for damage, declined to request on-orbit imaging, and ultimately discounted the possibility of a burn-through”—that is, that there might be a hole in the wing’s thermal protection layer through which hot plasma, generated by friction when the shuttle reentered Earth’s atmosphere, could invade the wing and blowtorch it from within. Unaware of any problem, the seven astronauts left orbit on schedule, seventeen days into the mission, cheerfully shooting a last few videos showing the red glow of plasma dancing outside the flight deck windows as they began their descent into the upper atmosphere. Soon thereafter the shuttle came apart, etching epitaphic skywriting across dawn skies from California to Louisiana. A forensic analysis conducted by a Crew Survivability Working Group convened at the Gehman board’s request concluded that the module containing the crew remained intact for approximately twenty-four seconds after the orbiter broke up, during which time it fell from an altitude of approximately 140,000 feet to a little over 100,000 feet before disintegrating.
NASA’s failure to order imaging of the shuttle to look for signs of wing damage while it was in orbit is as inexplicable as it is anguishing. The Air Force has a number of telescopic cameras on Earth that are capable of taking high-resolution pictures of an orbiting shuttle. The cameras’ precise capacities are classified but a conservative estimate, based on Air Force data, is that they make visible an object the size of a golf ball at the shuttle’s distance. The Gehman board estimated that the hole in Columbia‘s wing measured 100 square inches—about the size of a bucket of golf balls. Several shuttle ground team members suggested commissioning images by such cameras. One warned in an e-mail that he and his team would “always have big uncertainties” about the damage “until we get definitive, better, clearer photos of the wing and body underside,” adding, “Can we petition (beg) for outside…assistance?”
The answer was no. Incredibly, it seems that nobody on hand appreciated what the cameras could do—although that much was common knowledge among amateur astronomers and space buffs—and efforts by engineers to find out by contacting the Defense Department directly were scotched by their superiors. On Flight Day Six, one of the NASA officials in charge of the mission, Linda Ham, told the mission management team, which she chaired, that on-orbit imaging was not being pursued because even if it revealed damage, “there is not much we can do about it.” This was not the case; as we shall see, plenty could have been done to attempt a rescue of Columbia‘s crew. But the fact that mission managers “displayed no interest in understanding [the] problem and its implications,” as the Gehman board put it, seems to have resulted from NASA’s prideful reluctance to ask for aid from other agencies, combined with its old habit of downplaying hazards it had survived in the past.
As the reentering shuttle passed over Kirtland Air Force Base in Albuquerque on the morning of February 1, scientists there took a few images of it for their own amusement, using an ancient Macintosh computer and a digital camera coupled to a 3.5-inch Questar—a portable telescope so small that it can be transported as carry-on luggage on a commercial flight. The equipment didn’t work very well and clouds got in the way, but the scientists did manage to take one overexposed photograph when the shuttle was near the horizon. Even this rude image suggested that something was going terribly wrong, but by then it was of course too late. Meanwhile the advanced “Starfire” adaptive-optics telescopes at Kirtland stood idle, NASA officials having neglected to employ them. A week after the crash, Space Shuttle Program Manager Ron Dittemore displayed the Kirtland scientists’ fuzzy photo to reporters at a press conference, offering it as an example of how ground-based imaging wouldn’t have shown anything useful. “If your eye is sharp, maybe you can draw a conclusion,” he said. “I don’t think it’s very revealing.” Watching him on live TV, I was startled to realize that although Dittemore had managed eleven shuttle missions and was immersed in the dreadful business of trying to determine what had doomed the last one, he still didn’t comprehend the contribution that ground-based imaging could have made to Columbia‘s safety.
Kevin Spear, "Foam Likely to Hit Next Shuttle," Orlando Sentinel, posted July 27, 2003.↩
Kevin Spear, “Foam Likely to Hit Next Shuttle,” Orlando Sentinel, posted July 27, 2003.↩