December 1998 Volume 41 / Number 12
The 1998 Nall Report
The AOPA Air Safety Foundation has just published the 1998 Joseph T. Nall Report. The Nall Report series has become a definitive source on fixed-wing general aviation accidents for aircraft weighing less than 12,500 pounds. There is some good news on the safety front. GA has managed to break last year's all-time safety record. For comparison in '96 there were 1,781 accidents, with 355 of these involving fatalities.
In 1997, the overall accident number went down to 1,642, of which 331 involved fatalities; the number of fatalities, however, increased slightly over the previous year. The FAA estimates that the flying hours increased slightly from 24.1 million up to 24.7 million, so the rate also becomes a record-breaker, with the fewest total and fatal accidents since recordkeeping began in 1938.
Before we break our arms in self-congratulatory backslapping, however, reflect that in many cases these accidents really weren't accidents. An accident implies the unknown. As in all recent years, the leading causes of fatalities were continued VFR into instrument conditions and low-level maneuvering flight. These rather sanitary descriptions of unfortunate situations should more properly be called judgment failures.
Are they predictable? Yes. Take a VFR pilot, add some clouds and you have a lethal combination. Are clouds predictable? Usually, but the exact location, height, or onset is difficult to forecast. But temperature and dewpoint are very reliable measurements. Are weather briefings overly conservative? Yes, because survivors of unsuccessful pilots sue the government. But, as we've said many times before, the pilots have the best observation point from which to see the clouds before they enter never-never land for VFR.
There are several points to be made — most scud-running crashes probably don't happen the first time out, as near as we can tell. The pilot may be emboldened by early successes and might get away with scud-running for months or even years. We aren't able to interview many pilots after the fact. In 1997, more than 85 percent of these incidents involved fatalities — the highest of any type of mishap. That is a sobering thought when one is tempted to cheat on the VFR regulations.
The next big fatality area is low-level maneuvering, with many flights involving low-level aerobatics and the striking of objects on the ground. We've written about this before, as well. Those of you who are frustrated fighter pilots who like flying upside down and tight turns with high G forces should consider taking aerobatic instruction from a qualified instructor in the right aircraft. It will be an experience you can both live with and enjoy. Also dangerous are wire strikes, most of which occur below 100 feet agl. The only time an aircraft needs to be that low is just after takeoff and just before touchdown. This is a transition altitude, not a cruising one.
Some things never change. In 1997, as it was in 1996, and for years before, personal flying which is the most popular kind, resulted in a highly disproportionate number of accidents. Personal flight accounted for about 42 percent of the activity and almost 67 percent of the fatal accidents. Compare that to flight instruction, which has grown to more than 20 percent of all flight hours, with less than six percent of the fatal accidents.
There are several plausible explanations for instructional safety. First, flight instructors have sharper reflexes to fend off student indiscretions and are spring-loaded to avoid danger. If you don't quite buy that, consider that student pilots and instructors operate in highly controlled conditions — usually good weather, light winds, and generally close to the airport. They fly almost exclusively fixed-gear aircraft, which fare pretty well in crashes. These are relatively benign environments, with two exceptions.
Takeoffs and landings are the most likely phase of flight in which to have a mishap, although these are seldom fatal. A typical pattern lesson will encompass more bounce-and-goes than five or six cross-countries. Go-arounds have also become an area of concern. Students are not doing well with these, and more attention should be devoted to getting the abort landing procedure down pat.
The other high-risk exposure area for instructional operations is midair collisions. High-density traffic around nontowered airports is the most likely place for two aircraft to come together. In 1997 there were 13 midairs, with 11 resulting in a fatality in at least one of the aircraft. This is down significantly from the 18 midairs that occurred in 1996. Unfortunately, these tend to be random events, and the best protection is rigorous traffic pattern discipline and an active scan. It's not a guarantee, but it helps.
Let me digress momentarily to cite a recent study by Paul Rostykus, Peter Cummings, and Beth Mueller on Risk Factors for Pilot Fatalities in General Aviation Airplane Crash Landings. It appeared in the September issue of the Journal of the American Medical Association. This is probably not the publication to leave lying in the cockpit when friends or relatives come flying. The study, covering statistics from 1983 to 1992, analyzed 8,411 crash landings that occurred during takeoff or landings and involved the loss of engine power. Flights into mountains or towers, encounters with thunderstorms, and aerobatic flights were excluded.
Out of this subset involving power loss, there were 437 pilot deaths, or 5.2 percent. There were some identifiable risk factors that increased the fatality rate. The results are about what you'd expect. Crashing without a shoulder harness is bad. If the aircraft is destroyed or catches fire, that is very bad. Off-airport landings are worse than those on the airport and fixed gear aircraft do markedly better than high-performance retractables and twins. Several factors involving the physics of higher landing speed and weight come to mind. These aircraft also spend more time on cross-country flights away from the relative safety of airports.
A few other factors were not so immediately obvious. Pilots over the age of 60 were more likely to die — possibly because of poorer physical condition or less ability to escape from a wreck (Pure speculation on that, so don't hold me to it). Less obvious is that those without current flight reviews have a risk that is twice as great as those who were legally current. Pilots with total flight time between 201 hours and 1,000 hours had more risk than those with more or less flight time. The correlation may not imply the causality, but it certainly corresponds with the concept that a little knowledge is a dangerous thing. Low-timers are usually not overconfident because they know that they don't know everything, and ditto for the veterans. The mid-hour group appears to comprise the highest risktakers.
Before deciding that perhaps you'll take up golf or boating instead of flying, there are some bright spots. A survival rate of better than 94 percent after an engine stoppage is encouraging and this study found that the numbers of crashes of this type have declined over the years. Fleet activity is tough to measure, so be careful on the inferences drawn on this last statistic.
Going back to the Nall Report, we found that business flying involved proportionately half the number of fatalities relative to the amount of flying it represented (these are flights conducted by business people who are not professional pilots). This is always a revelation to corporate risk managers who refuse to let employees fly on company business even though it could save them money and time. Generally, despite the facts, many companies decide that the risk outweighs the reward, and much of GA's potential usefulness is lost. Executive and corporate flights flown by professional pilots typically have a safety record that matches that of the scheduled airlines.
One area that the Air Safety Foundation has been watching is the number of accidents related to mechanical failures or maintenance errors. All pilots operating older aircraft should pay attention. Logic says that there should be an increase in accidents as equipment ages. Looking at one of the graphs in the Nall Report, we saw a spike in 1988 that capped a four-year rising trend. The long-awaited maintenance accident increase appeared to have arrived. But in 1989 a six-year declining trend began that dropped maintenance-related accidents down to about 20 percent of the total. In 1995 and 1996 there was anther increase, and for 1997 the NTSB reported maintenance or mechanical causal factors at a low of 15 percent, lower than it's been since 1983 when the fleet was much newer. Data is still coming in from 1997, so while the picture is close, there may be some changes. One year does not make a trend, so we'll continue to watch.
Fuel management, or rather mismanagement, continues to be a leading identifiable cause. Last year more than two airplanes a week were crashing with no danger of catching fire — no gas in the tanks. Almost everybody almost made it — usually going down within just a few miles of the destination. Despite all the bent airframes, there were only five accidents involving fatalities. The question of judgment, or lack thereof, is puzzling.
A pervasive myth that surrounds GA is the fear of fatalities when a lightplane falls onto a house. In 1997, no one on the ground near an airport was killed or seriously injured by a GA crash. This doesn't mean that it can't happen, but of the risks that face us in everyday modern life, this is one that should be relegated to the bottom of the list.
At press time the NTSB had counted seven pilot incapacitation accidents. Sadly, three involved passenger fatalities. There were three heart attacks, two carbon monoxide poisonings, one hypoxia, and one from an unidentified illness. At least one of these situations might have been better resolved if the right-seat passengers had taken an ASF Pinch-Hitter course. It is a good investment.
All told, not a bad year for safety and with record attendance at ASF programs, we are gratified that our efforts may be having some effect on those who take advantage of our publications and seminars. Our hope is that all pilots will become more involved in safety in 1999.
A full copy of the 1998 Nall Report is available by sending $2 for postage and handling to: 1998 Nall Report, AOPA Air Safety Foundation, 421 Aviation Way, Frederick, Maryland 21701. Visit the ASF Web site for a safety quiz and an online look at the Nall Report. This year's report was sponsored by two Life Hat-in-the-Ring members.
See also the index of "Safety Pilot" articles, organized by subject. Bruce Landsberg is executive director of the AOPA Air Safety Foundation.