On Sept. 29, 2013, a Cessna Citation CJ2 landing at Santa Monica Airport near Los Angeles suddenly veered off the runway and crashed through a hangar, bursting into flames and killing all aboard.
Three months later, on Jan. 5, 2014, the pilots of a Challenger 601 landing with a strong, gusting tailwind in Aspen, Colorado, lost control and crashed next to the runway, killing one of the crew members and critically injuring two others.
Then, this past May 31, a Gulfstream IV departing Hanscom Field outside Boston failed to rotate on takeoff and careened off the end of the runway before crashing in flames in a field, killing the crew and all three passengers.
While safety educators have placed increasing emphasis in recent years on runway incursions, defined as incidents in which airplanes come into conflict with other aircraft or vehicles on the ground, there is a growing realization that pilots aren’t receiving adequate training for runway excursions – which, although they occur less frequently, are far more likely to lead to serious injuries or fatalities.
How do you train for something that happens so rarely – and for which a pilot’s actions immediately after the excursion has started might not make much difference in what happens next anyway?
Based on the details that have come to light so far, the Hanscom Gulfstream crash is especially troubling. While the accident investigation has only just started, what we do know is the crew was probably facing a serious, almost unfathomable mechanical problem. At about 9:40 p.m., the pilots swung the big private jet onto Hanscom’s Runway 11 and smoothly advanced the power, catapulting the airplane down the runway for what should have been a routine flight. What happened next was anything but routine.
As the speed built, the pilot in the right seat made the rote call-outs that seasoned corporate and airline pilots practice and execute time and again, until it’s second nature. “V1,” he said, followed a moment later by “Rotate.” This is one of the most critical times in any flight. If something goes wrong now, the pilots must react instantly, relying on their training, instinct and hopefully a good pre-departure briefing to handle the emergency safely.
But the Gulfstream crew never trained for this. The pilot in the left seat, with his hands wrapped around the yoke, pulled back to raise the nose into the air – and found that he couldn’t. For some reason, the controls were jammed. The pilots quickly discussed the problem and sprang into action. By this point the airplane had reached a speed of 165 knots, hurtling toward the end of the runway. According to information gleaned from the flight data recorders, the pilots pulled the power levers into full reverse thrust and stood on the brakes. But it was too late.
The Gulfstream departed the end of the 7,011-foot-long runway at a speed of more than 100 knots. The landing gear dug into the soft earth and collapsed as the jet’s momentum carried it for another 1,800 feet before finally coming to an abrupt stop in a ball of flames.
In its preliminary report the National Transportation Safety Board noted that the elevator was deflected downward during taxi and the takeoff roll, as if the Gulfstream’s mechanical gust lock were engaged. Investigators also found that the flap handle was in the 10-degree detent, but that the crew had set the flaps to 20 degrees for the takeoff. It’s not clear what roles, if any, these anomalies might have played in the crash, but investigators did note the pilots never performed a control check prior to departure as called for on the GIV’s checklist.
Managing the Risks
Just because the pilots were unprepared for the emergency doesn’t mean this tragedy could not have been prevented. Recognizing the danger of runway excursions and the fact that they can be unavoidable, the FAA designates Runway Safety Areas (RSA) at many airports as a way to increase the margin of safety in the event of an overrun or veer-off. RSAs also provide easier access to the crash scene for emergency first responders.
At airports built before the FAA began recognizing the special hazards of runway excursions, the agency has been installing EMAS (Engineered Material Arresting System) beds composed of high-energy-absorbing concrete blocks similar in concept to the emergency truck ramps made of sand on mountain roads. These beds are being installed at runway ends to stop jets from traveling beyond the RSA. To date, EMAS beds have been installed at 47 U.S. airports, with another 15 scheduled to receive them before the end of next year.
So far the EMAS beds have stopped nine airplanes from overrunning runways, with no reported serious injuries or fatalities. The most recent happened just days after the deadly Santa Monica Citation crash when the EMAS bed at West Palm Beach, Florida, stopped a Citation Sovereign overrun that very well could have led to fatalities.
While EMAS beds were initially installed at large commercial hubs, more recently they have been appearing at smaller airports, including some of the newest coming to Ohio’s Cleveland Burke Lakefront Airport, Trenton-Mercer Airport in New Jersey and Elmira-Corning Airport in upstate New York. If you’re wondering whether your home airport or an airport you fly into has an EMAS bed, its location and size would be specified on the airport diagram and the FAA maintains a list on its website.
Some of the facts and figures surrounding runway excursions might surprise you. For instance, excursions usually happen in good weather when the runway surface is dry, even though a wet runway and darkness ratchet up the risk. An excursion on takeoff is less likely than one on landing, but it’s far more dangerous, resulting in many times the injuries and fatalities.
Overall, from 1995 through 2010, there were more fatalities worldwide caused by runway excursions than either loss of control or controlled flight into terrain. Most troubling of all is the fact that even though only 10 percent of runway excursions in this 15-year time frame were fatal, these crashes accounted for an astonishing 1,121 deaths.
Obviously a runway excursion involving a large corporate jet or airliner is going to draw a lot of attention, but overruns and veer-offs are an even bigger safety problem in light general aviation. The four major factors leading to runway excursions for all types of operations are excessive speed on approach, strong wind, incorrect threshold crossing height and improper braking. What’s so unnerving about runway excursions is that there’s often very little time to react. The sequence of events leading to an offshoot or overrun can happen seemingly without warning.
“A crosswind coupled with a wet runway, for example, is a recipe for a runway excursion,” says Al Gorthy, assistant manager of the FAA’s central region runway safety office. “But the fact is there are not many predictors.”
Gorthy, who has been holding a series of runway excursion safety seminars around the country, says those occurring during landing are easier to predict, but he agrees that by the time the pilot gets himself into the situation leading to the excursion it’s often too late to do much if anything about it. Something to keep in mind during takeoff or landing when runway excursion risk factors exist is that multiple risks, such as a wet runway combined with a strong crosswind, will more than double the risk, he says
Excursion Red Flags
According to studies of runway excursions by the Flight Safety Foundation and International Civil Aviation Organization, clear risk patterns have emerged. On takeoff in a jet, for example, the biggest danger is in trying to reject a takeoff too late in the takeoff roll. The go/no-go decision really needs to be made before V1, experts say. A whopping 45 percent of takeoff excursions occur when the pilots try to stop after reaching V1. That’s because by the time pilots reach V1, realize it and begin reacting to the emergency, the airplane is already traveling much faster.
Red flags on approach, meanwhile, include failing to recognize the need to perform a go-around. Our built-in “normalcy bias,” Gorthy notes, prevents us from preparing for an event that so rarely happens, causing many pilots to decide subconsciously they are going to land no matter what. “Runway excursions often come with warnings, but you must listen for the signs to know what they are,” he said.
General aviation pilots, and even bizjet crews, are much more likely to fall into the trap of not conducting a go-around when prudent, Gorthy adds. The reasons for this aren’t clearly understood, but it could be the mindset among certain pilots to always finish an approach with a landing, or a company culture that implies “we don’t do go-arounds.” GA pilots may also lack the procedures and training so prevalent in the airline world, where if an approach is unstabilized, the crew is taught to immediately initiate a go-around.
The whole world, of course, saw what happens if an airline crew fails to recognize the symptoms of a botched approach when Asiana Airlines Flight 214 crashed at San Francisco International Airport in July 2013. The captain flying the approach became confused about the Boeing 777’s automated systems. Perhaps affected by the fatigue of a long, intercontinental flight, the crew initiated a go-around far too late to prevent the jetliner’s tail from striking a seawall short of the runway and cartwheeling on the runway, killing three passengers.
A big lesson here is that one of the first steps we can take to prevent a runway excursion is to avoid the missteps that can lead to them in the first place. Failing that, we should follow a time-tested mantra of military pilots: Hit the softest, cheapest thing you can find as slowly as possible. It’s not a joke, either. As you may know, at high speeds impact forces increase by the square of velocity.
When to Go Around
If we’re going to mitigate the risks of landing excursions that result from an unstabilized approach, our whole thought process concerning landings needs to change, experts say. We really need to be thinking about an unstable approach as a malfunction, Gorthy advises. It’s a failure situation for which there is no published checklist. When an approach deteriorates to the point that we can no longer say with confidence that it is stable, it’s often too late to fix it. It’s time to push the power up and go around.
What constitutes an unstabilized approach? The answers will vary from airplane to airplane and pilot to pilot, but generally an approach becomes unstabilized when our speed, descent rate and/or vertical/lateral flight path fall outside of expected norms. Generally, an approach is considered stable when the aircraft is on the correct flight path; only small changes in heading/pitch are required to stay on path; speed isn’t more than VREF+20 or below VREF; and sink rate isn’t greater than 1,000 feet per minute. If an approach becomes unstabilized below 1,000 feet in IMC or below 500 feet in VMC, it’s time to execute an immediate go-around.
That all sounds good in theory, but in the real world pilots too often allow parameters to fall well outside these numbers as they vainly try to coax and cajole their aircraft back on course. According to a Flight Safety Foundation study released last year, 96 percent of unstabilized approaches flown by nonairline crews never lead to the initiation of a go-around. Instead, pilots try to fix things, literally on the fly.
Besides failing to recognize the need to go around, the biggest risk factors for a runway excursion during landing include touching down long, approaching too fast or too high, and touching down hard, accident statistics show.
While these can all result from flying an unstabilized approach, very often the fault can lie with an air traffic controller who gave us that “slam-dunk” arrival from a higher than expected altitude. Here’s the simple truth about ATC that you may or may not have come to realize: A controller may not know enough about the characteristics of individual airplanes to understand whether an instruction he gives you is going to be 100 percent compatible with our performance in every instance. All he knows is that he needs us lower, he needs us faster, and he needs both right now.
In fact, a report by the Flight Safety Foundation found that many controllers lack an awareness of the importance of stabilized approaches; they often fail to allow aircraft to fly appropriate approach speeds; they don’t always assign the proper runway based on wind speed and direction; they sometimes make late runway changes inside the final approach fix; and they sometimes fail to pass on runway condition information and wind conditions. As a result, the FAA has become more proactive through training about making controllers aware that there are vast differences between the capabilities of a C-130 and a C-150.
Know Your Stuff
That said, controllers are trained from day one to assign published arrival procedures and keep speeds realistic. In the real world, ATC may have to delay your descent because of crossing traffic, give you an unexpected or shortened final approach, or assign a landing runway with a strong crosswind or a tailwind. It’s your job as pilot in command to refuse any clearance that you feel is beyond the capabilities of your airplane, or of yourself as the captain of your ship, for that matter.
This, of course, leads us into a discussion of understanding the performance of our airplanes inside and out. After all, if we don’t know the capabilities of our aircraft, can we really expect controllers to? How well do you know your airplane? For example, how much longer will you land if you touch down at VREF+10 versus at VREF? How will a wet runway combined with a tailwind impact your ability to stay on the runway? How will a hot day coupled with a heavy load of passengers and fuel affect your ability to take off from a short runway? These are all questions we need to know the answers to long before we’re sitting in the left seat trying to decide whether to accept the controller’s slam-dunk descent clearance or an approach to a different runway that will require a 45-degree banked turn to the left.
According to the Flight Safety Foundation, one way the FAA and manufacturers could assist pilots is by creating standardized takeoff and landing data for all aircraft and all runway conditions, as well as rethinking how runway condition reports are created and disseminated.
We can help ourselves, meanwhile, through better preflight planning. For instance, if we know it’ll be raining at our destination and the winds are forecast to be strong, maybe we can choose to land at another nearby airport with longer runways. Planning for the landing, after all, really should begin well before the takeoff so we’re aware, say, of whether the runway has a VASI or precision approach, if there is an adverse runway slope that could impact performance, and whether terrain or obstacles might necessitate a steeper than normal approach.
Finally, keep in mind that while automation such as an autopilot can be a fantastic tool to lighten your workload, an airplane with an abundance of technology can be flown like a regular airplane too if need be. If the automation isn’t doing what you anticipated, or if you become confused about what it’s doing, immediately revert to flying by stick, rudder and throttle.
You might not be able to prevent every imaginable scenario that could lead to a runway excursion – such as a blown tire or a deer running across your landing path – but with some forethought about possible scenarios, proper preflight planning and adequate training, you should be able to cut your risk by a comfortable margin.