Keith Epstein

Feds Shrug Off a Life-Saver for Commuter Planes

Safety device will be required...eventually

Plain Dealer – March 4, 1993

SERIES: DEADLY DELAYS – RED TAPE COST LIVES

Last of a five-part series

By Keith Epstein

Downeast Airlines commuter flight smashes into woods just short of an airport in Maine, killing 17.

1985: Henson Airlines commuter flight preparing to land crashes into a Virginia mountaintop, killing 14.

1989: Aloha IslandAir commuter flight nearing a Hawaiian airport hurtles into the slopes of a volcano, killing 20.

Each accident happened on routine approach to airport runways, with experienced crews at the controls. Each time, without realizing it, the pilot was flying too low.

During the last 17 years, in almost identical circumstances, at least 36 commuter airliners have crashed, claiming 127 lives.

For years, federal safety officials investigating the crashes have advocated a solution – a simple device known as a ground proximity warning system. It alerts a pilot who flies too low with a warning to “pull up!” Developed in the 1950s, the devices provide extra protection when pilots err or instruments fail.

The warning system has been required on large jets since 1975. It has saved an estimated 8,500 lives. And at $14,600, it’s a relatively inexpensive addition to a $3 million aircraft.

But for more than two decades the federal government has refused to require the warning systems where most needed – on commuter flights.

“How many deaths, how many people does it take before something is actually done?” asked Kevin St. Germain, the latest turboprop pilot to crash on approach. He lost a friend, his co-pilot, the full use of his shattered legs and his confidence.

“If I’d had even a few seconds’ warning,” said the 10-year veteran of commercial aviation, “at least I could have reacted.”

It happened just after 6 a.m. one morning in January 1992. As the CommutAir flight from Plattsburgh, N.Y., neared Saranac Lake, St. Germain began his routine descent over New York’s snow-encrusted Adirondack Mountains.

He couldn’t see through the mist, so he checked his speed and altitude using standard equipment. He checked his gyroscope. The localizer. Everything seemed normal.

It wasn’t.

Shrouded in clouds, the mountains lay 1,200 feet closer than he suspected. Only three miles from the runway, the plane smashed into Blue Hill. The aircraft sliced open a wide swath in the trees, shedding parts, clipping branches and exploding into bits.

When he came to in the snow, 250 feet away, fragments of metal lodged in his shivering limbs, St. Germain could recognize only the tail section. It was in flames.

He and the only passenger survived. His co-pilot died.

“A man three feet from me didn’t make it,” St. Germain said. “The Lord was with me. I guess I had an angel on my shoulder.”

He would have been better off with a ground proximity warning system in the cockpit.

Susan Coughlin, a member of the National Transportation Safety Board, said in an all-too-familiar refrain: “It appears that a GPWS would have prevented the accident.”

Fatigued, stressed or just human, men and women at the controls of the nation’s aircraft sometimes make mistakes. Even when they don’t, their instruments may. Data supplied by gauges in the cockpit can be insufficient or flawed.

In such situations, pilots unknowingly can stray too close to the ground.

With a warning to “pull up!” – even just seconds before impact – they usually can ascend, averting disaster. Thus, experts view the warning device as a valuable emergency backup.

In 1971, the safety board, discerning the pattern of needless deaths and destruction on passenger planes, urged the alarms on “all carrier aircraft.”

Eighteen years ago, they were first mandated on large jets, where the alarms quickly proved themselves. Latest estimates show the devices probably have saved at least 120 large planes.

Still, commuter airlines opposed the devices for years.

Recalls Charles O. Miller, who retired in 1974 as the safety board’s top aviation investigator: “I kept telling them, ‘Look, I can show you the voice cockpit records. The pilots (in fatal crashes) were … surprised (to be flying so low). They were human and needed help.’

Two years ago, the Federal Aviation Administration finally recognized the need on commuter aircraft. The Air Line Pilots Association agreed. “Similar casualties can be expected in the future if GPWS’s are not installed,” the agency declared.

It decided to require the devices.

But not for another two years – in some cases, not for another four years.

“Excessive” delay, the pilots association complains. “They should be required as soon as possible.”

That’s not the government’s way.

The FAA itself proposed requiring the devices in 1967.

Within a few months, under pressure from the aviation industry, the agency changed its mind. Not even a 1970 crash of a Southern Airways DC-9 ferrying a Marshall University football team back from a game swayed the agency. Seventy-five people died.

The safety board, which can only suggest remedies, urged the FAA to “evaluate … ground proximity warning devices.” But the aviation administration said it preferred to “continue emphasis on crew vigilance.”

After two more DC-9 incidents in 1971, the safety board again urged installation of ground proximity warning systems and, again, the FAA said the existing system was “safe and adequate.”

After FAA Administrator Alexander P. Butterfield visited a factory that made the alarms, he vowed to require them on all airliners “sooner than originally estimated.”

In 1975, when the FAA required them on large jets, the agency exempted commuter airliners. Under pressure from the airline industry, it decided to study whether the devices posed “an undue financial burden” on commuter airlines.

In July 1987 – 17 years after first urging the alarms – the safety board decided the matter was hopeless, categorizing the aviation agency’s action “unacceptable.”

It was the safety board’s way of officially giving up.

For many years, despite the advice of many experts, the FAA refused to abandon its official notion that commuter aircraft were “less susceptible” to accidents in which pilots flew too low.

Between 1975 and 1985, the safety board counted 31 crashes of scheduled commuter passenger flights, and 79 deaths. Not once had the flight crew been aware of flying so close to the ground. “Many of which could have been avoided,” the safety board concluded.

The bureaucracy also knew by then that the alarms had saved many lives on large jets – averting as many as 18 near-accidents in five years. Clearly, they had provided “the sole warning in six reported instances that otherwise would most probably have ended in disaster,” researchers reported in 1981 at an Ohio State University symposium.

By 1985, the risk should have been obvious, but the FAA continued to say “present rules provide an adequate level of safety.” Still planes flew into the ground. The best-known was in August when a Bar Harbor Airlines flight from Boston crashed in Maine, killing Samantha Smith, the schoolgirl famous as a peace ambassador, and seven others.

In 1988, two more airliners crashed – one on approach to Durango, Colo., the other at Raleigh-Durham, N.C. Dead: 21.

By the following year, another government study – echoing earlier studies – caught even the industry’s attention. A Transportation System Center study found that during the previous decade the ground proximity alarms would have prevented at least 18 – and perhaps 24 – of 27 accidents, sparing 64 people from death and 22 from serious injuries.

The report was a bombshell.

The Aerospace Industries Association of America made the warning system one of its top priorities. The industry urged the FAA to “encourage” their use on planes not yet required to have them.

Influenced by the report and the thawing of industry attitudes, the aviation agency agreed to “consider” requiring the alarms – action the safety board termed “acceptable.”

But the FAA failed to start the rule-making process until one year later, in April 1990. By then, four more airliners had crashed, – one ferrying congressional black leader Mickey Leland, D-Texas, and 15 others to an African refugee camp in August 1989; and others that crashed in Molokai, Hawaii; Pasco, Wash., and Elko, Nev.

Each time, safety experts concluded the pilots would have had time to avert the accident. Even the FAA acknowledged that “similar casualties can be expected” unless commuter airlines get the alarms.

The bureaucracy plodded along. Early last year, the safety board, in a budget report to Congress, again stressed that “ample evidence” demonstrated the need for the alarm.

Almost a month after Kevin St. Germain’s crash in the Adirondacks, the transportation secretary finally signed off on the requirement; but it took until mid-March for another agency, the Office of Management and Budget, to give the go-ahead.

In March 1992, the FAA published the final rule, but gave airlines two to four years to comply. It also offered an alibi. Officials said they had believed until then that commuter planes “had a greater ability to respond quickly in emergency situations” than large jets.

In Congress on the day the rule was published, Rep. James L. Oberstar, D-Minn., chairman of a House Public Works aviation subcommittee, asked the FAA’s top rule-maker, Associate Administrator Anthony J. Broderick, why the agency took so long to require the devices.

“Frankly,” said Broderick, “I just think it’s the workload. … the analysis we have to do.”

Oberstar persisted. Haven’t some of these arguments been going on for years? he asked.

“I will agree,” said Broderick. “It took too long. Everything takes too long nowadays.”

GRAPHIC: BOX: A SIMPLE DEVICE COULD SAVE LIVES; SOURCE: Federal Aviation Administration, Sundstrand Data Control Inc.; PD Graphic; PHOTO BY AP/MARK KURTZ; Downed trees mark where a small commuter plane went down in the Adirondacks on its way from Plattsburgh, N.Y., to Saranac Lake, N.Y. The pilot didn’t realize how close he was to cloud-shrouded mountains, but a warning device would have alerted him in time.