I remember my worst landing.
I was flying with a colleague, a pastor at the college where I teach, from Fargo, North Dakota (FAR), to Rugby, North Dakota (RUG), to visit the geographical center of North America. We were going to have lunch at the café and take pictures at the monument. Just a lark, a sightseeing trip, on a beautiful day for flying.
It was a straight-in approach to Runway 30, an easy 3,600-by-60-foot strip, and I had the rented Cessna 172 set up perfectly. I was a very low-time pilot, but I was confident. I had this one nailed.
I have no idea why I looked. The Garmin G1000, then faultless in my mind, told me the wind speed and direction. But I looked at both the windsocks at Rugby. Despite the wind where I was, both were hanging limp and pointed in different directions. This didn’t make any sense. There was a line of trees well away from the runway, and it suddenly dawned on me that those trees were blocking the wind. Twenty or maybe 30 feet above the runway, I was going to lose my headwind.
With only a few seconds to make a decision, I was suddenly outside my experience and imagination. I knew I had to get more air moving over the wings. Add power, I thought. How much? Hell, I don’t know. Just enough.
I added power and landed a bit long and fast. Although I landed safely, I confessed to my friend, the pastor, that the landing was the worst one I had ever made. He assured me the rules of pastoral confidence would cover bad landings.
Of course, I told everybody. I was grateful my N-number would not show up in a National Transportation Safety Board accident report.
I first started reading the NTSB aviation accident reports when I was a student pilot. When I discovered the reports were online, I looked for any accidents involving the Cessna 152, the model I was flying. I wanted to know how other people got in trouble, thinking that knowledge might help to keep me safe.
Some of the stories were simply absurd, like the CFI who fell asleep on short final and then the student botched the landing. Others were so far away from my world as a prairie pilot (it remains unlikely that I will fly into a box canyon here) that they were not very instructive. But every one of them did one thing I consider essential—as essential now as it was then, when I was a student: They kept me aware. Here, they reminded me, are the things that can go wrong.
One problem with the NTSB database, as with many others, is that it’s not easily searchable in large ways. Every report is individual. You can ask it to limit by type of airplane, type of accident, weather, location, or anything else, but you have to know what you’re looking for. If you’re looking to discover trends, for generalizations that might help a broader understanding of what we’re doing wrong, you have to do it the old way. So I decided to do just that, with a legal pad, a handful of pens, and a calculator.
I decided to look at the first six months of 2014. That seemed long enough to discover trends and short enough that I would be done before retirement. Then I decided to limit the report to the type of airplane most of us fly—single engine—and only those accidents within the 50 United States and its territories. The rules became fairly simple. If the airplane was single engine (no matter what type) and had control surfaces, it was in the list.
In the first six months of 2014, there were 459 accidents that fit my search criteria. Seventy-eight of those (17 percent) were fatal, with 117 deaths. Were there trends? Were there any useful lessons to share?
Missing information. Sadly, we seem to know the least about the accidents that kill. More often than not, there is no one to interview, no one to explain the failure of machine or ability. Sometimes we get a guess, though. A pilot tries to fly over the mountains with no training in high density altitude conditions, and the airplane cannot climb as fast as the rocks. There is the overshot base to final turn, a stall, and then a very short spin. An aerobatics practice goes wrong. The weather is bad. In one case the wings came off the airplane mid-flight. More than once, the engine broke. A VFR pilot flew into IFR conditions and could not get out. A pilot hit an unmarked wind turbine. A loose fuel cap caused the airplane to run out of fuel. In one case, the pilot was taking selfies.
More than once, the propeller came off the airplane. Just once, the pilot appeared to have committed suicide. Many times, we just don’t know. And the reasons are never simple. If the engine fails, do we fault the engine for failing to produce power—or the pilot for failing to successfully glide to an emergency landing?
Broken machines. The preflight checklist calls our attention to things that might have changed since we last flew, things we need to make this next flight work, and things that might need attention—such as oil in the engine and air in the tires. More often than you might expect, the reason for an accident begins in the hangar, with a repair (or lack of repair) that we would never find on the preflight checklist.
A landing gear switch was installed upside down, which would provide a gear-down indication regardless of the true position. A propeller was improperly reinstalled after an inspection and came off during flight. A propeller spline shaft failed from lack of lubrication (the manufacturer had no stated schedule for lubrication).
The database reports a mechanic’s improper installation of a vacuum pump, another’s failure to properly torque bolts, and a third’s misrigging of a fuel mixture control. Airplane kit builders have failed at wiring, installing magnetos, and determining the amount of unusable fuel. A mechanic failed to comply with an airworthiness directive, which resulted in a loose oil filter adapter, which led to no oil, which led to a crash.
By my unscientific count, 19 reports (4.1 percent) were accidents before the airplane ever took off. Not one of them would have been caught, even with the most detailed preflight inspection. All of them should have been caught with a bit more attention in the hangar.
The bad takeoff. A total of 38 accidents (8.2 percent) fell into the category of bad takeoff. A pilot took off from a dirt runway but didn’t climb, and the gear hit a berm. One pilot tried to take off in three-foot grass, while another tried to take off on a dirt runway that had turned to sticky mud in the rain. A pilot attempted to take off from a beach and hit a buoy. More than once, a pilot rotated too soon and pulled up into a stall. A handful of students on early solo flights couldn’t steer straight down the runway.
A few airplanes took off and immediately lost power. The pilots looked for somewhere to land. Several tried to get back to the departure runway, which is almost always impossible.
More often than not, however, bad takeoffs have the same contributing factor: the wind. A tailwheel airplane rolled down the runway in a gusting crosswind and as soon as the tailwheel was off the ground the airplane weathervaned, a wing struck the ground, and the airplane headed for a ditch. Another airplane lifted off and was blown sideways into a sign.
It’s easy to assume that once the airplane is off the ground the task gets a bit easier. And in many ways, it does. But the act of getting there—the roll and rotation and early climb—need every bit of our attention.
The bad landing. The good landing is not very exciting to the kid hanging out at the airport observation area. From the outside, a good landing looks a bit boring. It doesn’t have the rush and the noise of takeoff. But from the inside, the good landing is one of the most pleasing moments in flying. It’s ballet and finesse. It requires all our attention and skill. When the main gear touch so lightly we wonder if we’re really down, every one of us smiles.
Perhaps it’s not surprising that so many accidents fall into the category of the bad landing. A student pilot landed a seaplane with the gear down (the CFI didn’t notice). Another pilot was distracted by his dog, which was loose in the cabin. One pilot’s seat wasn’t latched in the rails and slid back when the gear touched down. Many pilots landed and then hit the brakes so hard the airplane nosed over or veered off the pavement. Other pilots rolled too fast. Several pilots landed badly, knew they landed badly, but waited far too long to decide on a go-around. Tailwheel pilots landed in quartering tailwinds and ground looped fairly often. While many pilots cannot find the centerline, a couple pilots landed so short they sank into dirt.
One hundred fifty-seven accidents (34.2 percent) would fall into the category of a bad landing. Thirty-one accidents (nearly 20 percent of all landing accidents) had exactly the same problem: a bad flare—a too-high flare that resulted in a hard landing, or cracked the gear, or bent the firewall.
Loss of power/out of fuel. There are a thousand ways for an airplane to experience a sudden loss of power. All sorts of things can go wrong with the engine. That’s why inspection schedules are so rigorous. That’s why oil changes should not be delayed. One hundred one accidents (22 percent) followed a sudden loss of power, many during takeoff. Many of them are a mystery, because when the NTSB inspected and tested the engine, it found no apparent issues.
One aspect of this problem has everything to do with decision making and situational awareness. Of the 101 loss of power accidents, 45 were for the same reason: the airplane ran out of fuel. Many pilots claimed to have made preflight fuel calculations but then did not rethink when the winds changed. Several knew their fuel gauges were incorrect but took off anyway. A few pilots did not sump their tanks—they never had water in their fuel before. A handful never checked the fuel at all.
What were they thinking? A great many accidents are unique. It’s difficult to think they happened even once, much less to think they might be part of a trend. But there they are—the I-can’t-believe-its. And here is my advice:
Do not fly drunk. Do not put a nonpilot in the cockpit and tell him to press the brakes while you hand-start the engine. If you are a CFI, don’t have your seat so far back you can’t reach the controls. If you have never received any sort of aerobatic training, don’t assume you can figure it out on your own. The rotor wash of a helicopter is dangerous. Don’t run into the forklift. Check the fuel. Sump the fuel! Check the oil. Use carburetor heat. Avoid the guy on the boat (or the lawn mower). Follow the checklist.
Conclusions. Sometimes the world conspires against us. No amount of talent or skill, no amount of preflight preparation or depth of experience will help. If you are the pilot who’s carrying skydivers and one of them deploys a chute early—and it winds itself all over your tail and horizontal stabilizer—be glad you’re wearing a parachute yourself.
When a deer leaps onto the runway during the landing roll, about all you can do is expect a mess. When a herd of deer leap in your path, expect a herd of mess. There are more deer strikes than bird strikes in the NTSB database, but fortunately this type of accident is rare.
What is not rare is our tendency to respond too slowly, to overestimate our airplane, to underestimate the wind. Confidence is a common trait among pilots, although it can lead to shortcuts, bad preflight inspections and planning, and poor decisions while landing.
If the lessons of the accident database can be reduced to one common theme, it would be the implication carried by one line, repeated in various ways throughout reports. “A post-accident examination of the airplane revealed no pre-impact mechanical malfunctions or failures with the airplane that would have precluded normal operation.” In other words: pilot error. We can all fly smarter.
Related Articles
