General aviation stubbornly refuses to become safer. Can training, technology and changing culture be combined to drive down accidents?
In the USA, where fixed-wing general aviation operations represent about half the global total, accident rates may have significantly improved in the past 20 years, but the figures have levelled out since 1999, according to the US National Transportation Safety Board (NTSB). Indeed, the Federal Aviation Administration has just suggested 2004 may show an upturn. The USA's experience in this respect is not unique, so aviation authorities in nations with a strong safety culture are looking at research, data analysis and technology for ways of breaking through this "glass floor" and forcing GA accident rates further down.
Predictability is one of the characteristics of fixed-wing GA accident statistics. Figures prove that pilots, in any given country or region, tend to crash for the same reasons year after year. This is the main reason for researching training methods in the hope of finding the key to eliminating accident causes that have proved resistant to change, despite years of warnings and experience. Are traditional training messages just not getting through? The US Aircraft Owners and Pilots Association's (AOPA) Air Safety Foundation (ASF) executive director Bruce Landsberg is not so sure: he thinks the most common accidents are the results of risk-taking, and there is a limit to what you can do to change the personality of particular individuals.
For example, weather is a risk to all pilots, but particularly to those who own - or part-own - their aircraft. According to aviation club chief flying instructors (CFI), club pilots who are qualified only for good weather flying tend to find the fleet grounded if the weather threatens, so the decision whether to fly is taken away. But owner/fliers, the CFIs say, are often tempted to push their luck and take off when others are grounded. The implied problem is lack of discipline rather than lack of training.
The greatest risk to pilots' lives in many countries, including Canada, South America, the UK and the USA, and most continental European countries, is continued visual flight rules (VFR) operation into deteriorating weather. The ASF is so concerned about this deadly but common trap that it is promoting on its website an alternative to "scud-running". This is aimed at those who are unexpectedly confronted by reducing visibility or are caught under a lowering cloudbase. The ASF calls it "a free terrain avoidance plan".
The objective is to climb to safety altitude - or obstacle clearance altitude (OCA) - for the area, and to fly at or above it. It might sound extraordinary to professionals that any pilot should need to be told to do this - and how to do it - but those who have only been trained for flight only in visual meteorological conditions (VMC) might not yet have been taught it. Their training simply tells them not to get into instrument meteorological conditions (IMC).
Avoiding traps
The ASF website tells pilots how to calculate the OCA from different commonly used charts. The need for this advice dramatically illustrates AOPA's degree of concern about VFR flight into IMC, because the ASF knows that although climbing quickly reduces the risk of controlled flight into terrain (CFIT) or low-visibility loss of control (LV-LOC), climbing in IMC still involves the risk of LOC for those with little or no instrument flying experience. Furthermore, flight at an unplanned higher altitude could introduce the risk of icing, depending on ambient conditions. AOPA has spent years telling pilots not to get into the deadly VFR/IMC trap, but Landsberg says those who find themselves in it need to know the least risky method of dealing with the situation.
Meanwhile, the US Federal Aviation Administration has had its successes, despite the fact that overall US GA safety appeared to stop improving about five years ago. Its most notable achievements has been with Alaska's essential GA sector. Aircraft there operate in hostile, changeable weather and mountainous terrain, and the FAA felt it had to do something to lower the high accident rate. It is now well advanced in its highly successful Capstone project . This harnesses modern navigation systems to add to pilot situational awareness, uses automatic dependent surveillance - broadcast (ADS-B) for surveillance and separation, and has applied this to an unfriendly aviation environment with spectacular results. There are more improvements to come.
In Australia, the accident rate for aviation as a whole has dropped significantly between 1993 and 2002, driven mostly by improvements in GA safety, according to the Australian Transport Safety Bureau (ATSB). But although Australia - unlike the USA - appears not to have reached its GA safety "glass floor" yet, this has not stopped the country's Civil Aviation Safety Authority (CASA) looking for ways of continuing the improvement.
Accident, fatal accident and fatality rates for GA all show strong declines over the period. The GA non-fatal accident rate was 47.6% lower in 2002 than in 1993, the fatal accident rate 54% lower and the fatality rate 49.4% down, says the ATSB. The board attributes the improvements to better safety awareness and education in the GA community, largely due to efforts by CASA to focus heavily on safety education in the sector. In 1993 there were 10.57 accidents every 100,000 hours flown and a fatality rate of 2.13. By 2002, this had dropped to 5.28 for accidents and a 0.88 fatality rate. The accident rate has declined at an average annual rate of 6.7%, says the ATSB.
Agricultural operations and the private/business sector have consistently had the highest accident rates, says the ATSB, but the former has shown dramatic improvement in the past decade as a result of industry-run programmes. Ten to 15 years ago there were seven or eight fatalities a year in agricultural aviation, but this fell in 2000 to three a year, and for the past three years there have been no fatalities.
Spraying regulation
The Agricultural Aviation Association of Australia (AAAA) had set up its own spraying regulation - Spray Safe - several years ago, and requires all operators to be accredited. Agricultural pilots must have full commercial pilots' licences, an agricultural rating and a chemical distributor's licence. Now CASA has asked the AAAA for a proposal for total self-regulation, although the agency would retain its rulemaking and surveillance role.
CASA recently launched a programme with the country's flight-training industry to see if it can reduce GA accidents further. The move follows completion of a study by the authority and the ATSB to classify Australia's GA fatal accidents in the past 10 years. The study found that 32% of fatal accidents involved controlled flight into terrain and 43% were "uncontrolled flight into terrain", says the ATSB. Poor flight planning, aircraft handling problems, and fuel starvation and exhaustion are the main causes of GA accidents in Australia, the study found, with one or more of these factors contributing to nearly two-thirds of the 196 fatal GA accidents between 1991 and 2000.
"These are areas where pilot training can be used as a preventive weapon and CASA needs to find better methods of helping the industry deliver the most effective information and skills," says CASA chief executive Bruce Byron. Byron told a March meeting of the Royal Federation of Aero Clubs of Australia that he plans to set up a panel of senior flight instructors and CASA personnel to define ways to assist the flight-training sector, including looking at revising the generic instructor's handbook.
Meanwhile, a long-term ATSB study, prompted by three mid-air GA collisions between February and July 2002, has been looking at the issue of safe separation. The rate of mid-air collisions at general aviation airports in Australia is similar to that in the USA. There were 37 mid-air collisions involving GA aircraft during the period 1961 and 2003, with about one a year since 1968. The majority, 19, resulted in fatalities. Mid-air collisions accounted for 3% of all fatal GA accidents, with 71% occurring in or near the circuit area of an airport. But the ATSB determined that there were no dominant contributing factors and that the characteristics were much the same as those in other countries that have undertaken similar studies, such as Canada and France.
The FAA has commissioned many studies over the past decade aimed at making GA cockpits simpler and improving situational awareness by making better use of planned new technology. Now the FAA's Office of the Chief Scientist for Human Factors is examining human error-related GA accidents. About 14,000 such incidents (3,000 fatal and 11,000 non-fatal) are coming under the microscope of five independent inspectors.
FAA training research
Among the 3,000 fatal GA accidents associated with aircrew error, the FAA has found that "skill-based errors, not decision errors, were the number one type of human error associated with fatal GA accidents". Skill-based errors constituted roughly 80% of the total across the years of the study (1991-8), which was more than double the percentage of decision error accidents (32%) and rule-violation accidents (33%). "Even perceptual errors, the focus of a great deal of interest over the years, were associated with less than 15% of all fatal accidents," observes the FAA study.
Also evident was the fact that the proportions of all the accident categories had hardly changed in the eight-year period, and the only significant difference between fatal and non-fatal accidents was that in the fatals there was a far higher proportion of violation-based accidents - 33% compared with about 10%.
The premise of this ongoing FAA research is that more carefully targeted training is, in the near term, the most cost-effective way of counteracting the leading causes of GA accidents in the absence of radical, widely used new technology. Then, says the FAA, when the new technology becomes broadly available, proper training to make full use of it is essential to reap the potential benefits it offers - and the study examines the kind of equipment training that will be needed. But new aircraft systems providing traffic avoidance, direct routing, real-time weather and terrain cockpit displays will not be introduced in sufficient quantities in new aircraft, or as retrofits to the current GA fleet, in time to make a significant difference to the accident rate by 2007 - its original target year - the FAA says.
Meanwhile, training research will be designed to:
reduce GA accident rates by focusing training on areas identified as leading causative accident factors;
ensure GA pilots are trained to make full use of new retrofitted aircraft systems. This recognises that widely used new technologies, like GPS, can be complex despite bringing potential safety benefits;
ensure that development of new GA aircraft systems, including cockpit interfaces, takes into account human factors and training issues;
develop evaluation and certification methods for new and emerging technologies; in the longer term this research will build on the cockpit and avionics design work accomplished under the NASA/FAA Advanced General Aviation Transport Experiments (AGATE), like synthetic vision systems-general aviation (SVS-GA). The SVS-GA is intended to overcome LV-LOC by dealing with the problem low-hours pilots have carrying out a conventional instrument scan. Pilots will be presented with an intuitive "real" display that combines aircraft attitude with a synthetic picture of navigational position and terrain.
But it is estimated that the type of aircraft envisaged by AGATE, with SVS-GA and a Small Aircraft Transportation System (SATS) will not be widely used until about 2024. SATS will provide improved aircraft systems and automatic interface with the National Airspace System.
AGATE aircraft will be equipped with a primary flight display that includes the "highway-in-the-sky" virtual VFR system, a multifunctional display and single-lever power control systems.
The FAA says, however: "The effectiveness of these new systems and the ability to achieve reduced accident rates in today's GA operational environment will only be realised if improvements and innovation in training methods and procedures accompanies the technical systems effort. Without such emphasis, the effectiveness of new systems will be severely reduced and near-term accident rates may continue at the same level for years to come."
DAVID LEARMOUNT / LONDON
Source: Flight International
