Sullenberger never had to. For five years he flew F-4s without drama, until in 1980 he left the air force and joined the airlines, where punching out was no longer possible but the airplanes naturally provided for more gradual descents to the ground. Indeed, jet airliners over time have become so reluctant to lose altitude that workaday descents require significant advance planning, and often cannot be performed as decisively as Air Traffic Control desires, particularly when speed reductions are also required. Sullenberger made the parallel point to me that the latest generations of airliners have grown significantly cheaper to fly per passenger-mile. This is due largely to improvements in aerodynamics, and most fundamentally to the introduction of long, thin, sophisticated wings meant to lift well at high altitude, with minimal drag, and to milk the maximum range from the fuel aboard. The engines are not quite afterthoughts—and important gains have been made there too—but to a striking degree modern airliners are coming to resemble high-performance gliders, or “sailplanes,” the ultimate in efficient flying machines. Sullenberger flew gliders years ago as a cadet at the Air Force Academy, and he worked as a glider instructor for a few summers before becoming a fighter pilot. But the sailplane experience was less useful than it might appear to have been. More relevant is the fact that modern airliners have become good gliders in their own right, and they prove it daily during routine descents with passengers aboard. During those descents the engines are throttled back to a minimum setting known as “flight idle,” at which they produce hardly any thrust at all, and, unbeknownst to the passengers, the airplanes are glided for as much as 50 miles until arriving at the desired lower altitudes, where power is again applied.
Of course, the mark of a true glider is that it has no engine at all, and therefore has no power to apply at the end of a descent. The solution in high-performance sailplanes is to find atmospheric lift, and to ride the rising currents to gain altitude and stay aloft. Because such sailplanes are capable of losing as little as 100 feet a minute, the merest lift suffices. It is routine after an initial tow to a low-altitude release position to fly full days and hundreds of miles before coming in for a landing. Indeed, sailplane endurance attempts were canceled after a Frenchman stayed aloft for 56 hours in 1952, and it was decided that this was getting to be unsafe and ridiculous. The current distance record for sailplanes is 1,870 miles, flown in Argentina through mountain waves by a German pilot named Klaus Ohlmann in 2003.
V. The Glide
It is obvious that no one will set soaring records in an airliner without power, but experience shows that a total loss of thrust is not necessarily catastrophic. There was the 1982 case, for instance, of a British Airways Boeing 747 that flew through a volcanic plume one night over Indonesia and suffered compressor stalls, surges, and the loss of all four engines at 37,000 feet. The ensuing glide (with engines harmlessly belching fire) was written up afterward as a “near-death” experience for the passengers, during which the airplane “plummeted.” But “near death” is a relative concept, and in fact the crew had more than 20 minutes of available gliding time, during which they figured they could reach a certain airport about 100 miles distant. The pilots were hardly relaxed. They were issuing Mayday calls to Jakarta Control, flying the airplane, handling the depressurization of the cabin, and struggling with procedures to re-start the engines. Nonetheless, in the midst of the glide, and with appropriate British aplomb, the captain announced to the cabin, “Good evening, ladies and gentlemen. This is your captain speaking. We have a small problem. All four engines have stopped. We are all doing our damnedest to get them going again. I trust you are not in too much distress.” The captain’s name was Eric Moody, to give credit where it is due. A few people were indeed in distress, but perhaps because this was a flight from England to New Zealand, most of the passengers seem to have matched the captain’s calm. One, an aging British woman traveling with her aged mother, turned back to a Jane Austen novel at the first sign of trouble. Apparently, she just was not going to stand for this nonsense. And sure enough, as the airplane descended below 12,000 feet, the crew was able to re-start the engines.
Engines-out airline gliding is not a sanctioned category, but records exist nonetheless. The current holder appears to be a Canadian named Robert Piché, who, if not exactly a rebel, seems to be something of a maverick. In August 2001 he was flying as the captain of a wide-bodied, twin-engine Airbus A330 at night over the Atlantic, and was a thousand miles from Europe when he ran out of fuel and lost all thrust in both engines. The airplane belonged to a Canadian charter company called Air Transat, and was carrying 306 people overnight from Toronto to Lisbon. Piché was a French Canadian par excellence—a man roughly Sullenberger’s age, who had grown up on Quebec’s remote Gaspé Peninsula in a town in the shadow of an airport, and who, like Sullenberger, had learned to fly as a teenager. While Sullenberger had gone off to the air force and a regimented airline career, Piché had moved in a different direction, becoming a bush pilot and flying every manner of old airplane on every manner of mission, including at least one run in 1983 from the Caribbean to Georgia, in a Piper Aztec, bringing in a load of marijuana. Disapprove if you will, but a run like that took nerve because it was solo and risky. Piché got caught. He served 16 months of a 10-year sentence in a Georgia prison. Afterward he returned to Canada and a checkered flying career, until in 1996, at the age of 43, he managed to hire on with Air Transat, an outfit with the decency not to hold his transgressions against him. He rose rapidly from co-pilot to captain on Lockheed L-1011s, and transitioned to the Airbus A330 in the spring of 2000, after four weeks of simulator training at the factory in France. He liked the airplane—who would not? But now, just over a year later, and far out over the Atlantic, his airplane sprang a leak and began to lose fuel—first from the right tanks, where the leak had occurred, and then, because of a valve that Piché mistakenly opened, from the left tanks as well.
It was 5:36 in the morning, local time. They were at 39,000 feet on a clear black night, with stars overhead but nothing more in sight. Piché decided to backtrack to the nearest airport, on Terceira Island, in the Azores, about 400 miles to the southwest. He made the turn, and advised Oceanic Control of the situation. It was 5:48 a.m. Around this time the senior flight attendant entered the co*ckpit to discuss the passenger services that would be required in Lisbon. Piché warned her of the low-fuel condition, and of the precautionary diversion to the Azores. She left to inform the cabin crew and secure the galleys. Several minutes later she returned, and Piché told her to prepare for a ditching. A ditching is a water landing. In a jet airliner, a ditching at night in the Atlantic means near-certain death, no matter how good the airplane is or who is flying it. The flight attendant went back and calmly instructed the cabin crew to prepare for passenger demonstrations.
It was now 6:06 a.m., a half-hour since the first sign of trouble, and still pitch-black outside. Seven minutes later, at 6:13 a.m., the right engine quietly flamed out, and the airplane began a gentle descent from 39,000 feet, still 170 miles from the airport. The passengers would have been unaware of anything unusual had not the cabin crew suddenly taken up positions in the aisles, with the lights up full bright, and begun instructing people to put on the life vests that were stored under the seats. This is not a pleasant way to be woken up on a transatlantic flight, and the mood was not exactly calm, but the passengers maintained sufficient self-restraint at least to get the vests on. As the airplane settled through 37,000 feet, Piché spotted the lights of the island about 140 miles ahead.
But at 6:26 a.m., 13 minutes into the single-engine descent, the left engine used up the last of the fuel, and the A330 became a glider. Piché responded with French profanity. They were 90 miles from the airport, trimmed to a book value for best gliding speed, and descending at about 1,200 feet a minute. From the underside of the airplane an emergency windmill known as a ram air turbine automatically extended into the slipstream and began spinning to provide backup hydraulic power and the minimum of electricity. In the cabin, the regular lights flickered and went out, and the dim emergency lighting came on. This did not go over well with the passengers. To make matters worse, the public-address system began to fail (normally a blessing, but inconvenient here). Five minutes later, as the cabin pressure leaked away, the oxygen masks automatically dropped, and this caused another round of fussing. Up in the co*ckpit, the situation was more sober, but also quite rough. Piché and the co-pilot were so busy that they did not put on their oxygen masks. They had lost most of the airplane’s electronics and flat-panel displays, and were flying with degraded controls which offered little of the assistance normally provided by the A330 to its crews. Later, Piché implied that his entire life had led up to this moment in flight—a tautology that seems to have become standard in such cases. He even included prison in the progression, and credited it for teaching him not to shy away from realities, however grim.
Not that he had a choice. At 6:31 a.m., while gliding down through the night, 27,000 feet high and 38 miles from the airport, he checked in with Approach Control, and requested that the runway lights be flashed. At 6:39 he arrived nine miles off the approach end of the runway, and 13,000 feet high. Being high allowed for plenty of maneuvering for flight-path control. It introduced room for piloting errors but also for piloting skills, and took away the element of chance. Piché swept the airplane into a descending 360-degree turn during which he extended the leading-edge slats and lowered and locked the landing gear. He straightened out from the turn at 8,000 feet on a long final approach. The runway ahead was 10,866 feet long. It was outlined in lights. Piché saw that he was high, and put the Airbus through a series of S-turns, like switchbacks, much as some F-4 pilots might have wanted to do when flamed-out, if only they had been allowed. Piché was flying at F-4 speeds or faster, though with descent rates much lower. In the cabin the flight attendants were screaming “Brace! Brace! Brace!” to the terrified passengers. The airplane crossed the runway threshold doing 230 miles an hour, slammed against the pavement about a thousand feet along, bounced back into the air, and floated for another 1,770 feet until Piché goddamned planted that airplane down to stay, and locked the brakes. The planting did not drive the landing gear through the wings, but it was hard enough to wrinkle the fuselage. The locked tires slid for about 400 feet, then abraded and deflated, leaving the airplane to grind to a halt on the ruins of its wheels. It was 6:46 a.m., at the end of a world-record, 20-minute, 34,500-foot, 90-mile, 306-person, engines-out airline glide. The passengers evacuated down the slides. Piché followed, and walked around the airplane. The wheels were destroyed. Jesus f*ck Me Mother Mary. Piché returned to a hero’s welcome in Canada, where at first he ducked the publicity, then learned to enjoy it and seek it out. He accepted multiple awards, authorized an official biography entitled Robert Piché: Hands on Destiny (available in French or English, autographed), built an official Captain Robert Piché Web site, to which you may provide your e-mail address (“Captain Piché activities: be informed!”), and set himself up as an inspirational speaker for business groups (teamwork and resolve) and schools (excitement about aviation). It turned out that Captain Piché was quite a speaker. He returned to flying, however, because flying, goddamn it, is what he best knows.
VI. The Choices
Captain Sullenberger does not swear. Furthermore, the closer he gets to airplanes, the more straitlaced he becomes. You can hear it in his transmissions to the New York radar controller immediately after losing thrust. He was calm, concentrated, and completely appropriate. The airplane had been climbing to the north and had hit the geese around 3,000 feet. Sullenberger had just taken over from Skiles. The controller knew none of it yet. He wanted to send the flight west, and then on toward Charlotte. He said, “Cactus 1549, turn left, heading two seven zero.” Sullenberger answered tightly, “Ah, this is, uh, Cactus 1549, hit birds. We lost thrust in both engines. We’re turning back towards La Guardia.” His voice was clear. He was hand-flying the turn, holding the airplane at its best gliding speed, and coordinating the re-start attempts with First Officer Skiles. Skiles is a former 737 captain who was required to fly as co-pilot again because of reductions in the US Airways ranks. f*ck it, the airlines. Come what may, this was going to be a competent operation. And the controller was cool for the game. He said, “O.K., yeah, you need to return to La Guardia. Turn left, heading of, uh, two two zero.” Sullenberger confirmed, “Two two zero.” The controller picked up the phone to La Guardia Tower. He said, “Tower, stop your departures. We got an emergency returning.”
“Who is it?”
“It’s 1549. He, ah … bird strike. He lost all engines. He lost the thrust in the engines. He is returning immediately.”
The tower was briefly incredulous. “Cactus 1549. Which engines?”
“He lost thrust in both engines, he said.”
“Got it.”
There are two runways at La Guardia, both fairly short at 7,000 feet long. They cross. Between them they provide four thresholds, or directions, to land. One of those thresholds is bounded by a busy expressway and is embedded in the neighborhoods of Queens. The other three are embedded in the upper reaches of Long Island Sound—industrial waters cluttered by bridges and causeways, and by pilings supporting the approach lights of the runways themselves. If you’re going to undershoot or overshoot a runway, you’d rather not do it here. The controller offered Sullenberger the nearest threshold to his position in the turn. He radioed, “Cactus 1549, if we can get it to you, do you want to try to land Runway 13?”
Sullenberger answered, “We’re unable. We may end up in the Hudson.” In the abstract, the judgment was a close call. Perhaps 40 seconds had passed since the impact with the birds. The airplane had just rolled out of the turn, and was descending smoothly, southwest-bound over the Bronx. Skiles was into a checklist trying to re-start the engines. Sullenberger could see La Guardia to the left side. Like all pilots he was experienced at visually projecting flight paths, even around corners, and particularly in descents. It was not obvious that if he turned directly toward the airport he would undershoot the runway. But he clearly did not have the margins of altitude that had been enjoyed for instance by Piché, and that would have allowed him to adjust the flight-path geometry to navigate a surefire descent to a safe landing. Over the months after he made the decision not to try for the runway, multiple simulations of it have been run, and not a single pilot has been able to stretch the glide to La Guardia—an outcome that would seem to justify Sullenberger’s decision to go for the Hudson instead. But that misses the point. Even if it had been shown in simulation that Sullenberger could in theory have glided to La Guardia, in practice the approach would have been a very close thing, a crapshoot in a place where undershooting the runway by 20 feet would be like undershooting it by a mile. Once you committed toward La Guardia, you either had luck on your side or you died. Who knows what Piché might have tried? But Sullenberger was not that kind of gambler.
When Sullenberger said he was unable to land on Runway 13, the controller, correctly, made no assumption about why. For all he knew, the airplane had been damaged to the extent of being difficult to fly, and in conjunction with that was perhaps even too high. He therefore responded by offering Sullenberger the same runway in the opposite direction, on the chance that Sullenberger might need to overfly the airport, then turn around and come back in. He radioed, “All right, Cactus 1549, it’s going to be left traffic to Runway 31.”
