Safety board blames pilots, controllers for USAir crash


WASHINGTON -- The National Transportation Safety Board concluded yesterday that USAir Flight 1016 crashed in Charlotte, N.C., on July 2 because air traffic controllers failed to pass along crucial weather information, the crew failed to recognize signs of wind shear and, at the crucial moment, the apparently disoriented captain told his first officer to push the nose of the plane down when it should have been climbing.

The findings were approved at a seven-hour meeting to establish the cause of the crash, which killed 37 people and injured 20 others aboard the USAir DC-9.

Besides listing ways the crash could have been prevented, the board reiterated a 1990 recommendation that child safety seats be required on all airliners.

A 9-month-old baby who died in the crash would probably have survived if she had been in a child safety seat, said Nora Marshall, a staff investigator.

The chairman of the safety board, James E. Hall, seemed as tonished by the profusion of weather information that was available on the ground but not in the cockpit: For example, a departing plane had radioed the tower that there was a storm over the airport; USAir had evacuated its ramp area because of lightning; and the controllers' radar showed severe weather.

The crew members of Flight 1016 were not told these things, investigators found, although they did know, as did the controllers, that two USAir planes waiting to take off had waited on the runway.

"So, if a storm's on top, it closes the ramps, holds the planes, shows level 3 on the radar, there's no requirement that this be given to the pilot?" Mr. Hall asked during the hearing. "I find that . . . upsetting."

Besides faulting the Federal Aviation Administration controllers and the actions of the crew in the crash, the board found problems with training at USAir and with the FAA's oversight of training.

An FAA audit of pilot training at USAir in 1993 turned up 51 cases where it was not clear whether the pilot had wind-shear training. But the FAA's principal operations inspector for the airline never followed up, the board said.

The board's investigation showed that the crew had the possibility of wind shear in mind toward the end of the short hop from Columbia, S.C.

But the captain, Michael Greenlee, and the first officer, James Hayes, both of whom survived, testified later that while some clues were present, all of the wind-shear drills that they had run in the USAir simulator included turbulence, but there was none preceding the crash.

On the hot July evening, the plane's radar spotted a thunderstorm 15 miles ahead and the pilots asked air traffic control for permission to fly around it; the controller said that the plane would be turning away for approach before it reached the spot. But the plane hit the cell, experiencing a sharp increase and then decrease in airspeed.

But wind shear alone did not cause the crash, the board found. James Ritter, a performance expert on the staff, said that when the plane hit the ground it was flying about 40 knots above the speed at which a stall warning system sounds. The wind shear was diminishing, he said, and some of the plane's forward motion could have been used to climb.

As outlined by investigators, the final moments of the flight occurred this way: The captain decided to abort the approach to the runway, not because of wind shear but because of sharply decreased visibility. With the plane enveloped in clouds, the first officer, at the controls, put the plane into a climb, the nose angled up 15 degrees, and accelerated.

Investigators said the captain appeared to become disoriented, suffering from what they called "a perceptual illusion" created by the inability of the middle ear to differentiate between the plane speeding up and the plane tilting back.

Called somatogravic illusion, and known among pilots as elevator illusion, the effect is believed to have been responsible for at least two other crashes in the last two years, the safety board said.

The USAir captain had the problem, investigators said, because, inside the cloud, he had no visual reference. The first officer probably did not suffer from the problem because his attention was focused on the artificial horizon and other instruments, investigators said.

The captain, the cockpit voice recorder shows, told the co-pilot "down, push it down" referring to the nose of the plane. The captain said later he did not remember making the statement and the first officer said he did not hear it.

But the flight recorder shows that the first officer worked the controls to push the nose 5 degrees down.

If the plane had not nosed down, recovery, or flying through the wind shear, "was possible," Mr. Ritter said.

The Air Line Pilots Association sharply disputed that conclusion and cited a study by NASA engineers that suggested that the plane would have crashed anyway.

H. Keith Hagy, the union's manager of engineering and accident investigation, said: "Bottom line, the pilots shouldn't have been there in the first place," and that they would not have been if they had received proper information from the ground.

In its meeting, the board cataloged missed opportunities to warn the plane, or for the crew to have followed procedures more closely and avoided the crash.

Among the problems were a National Weather Service expert in Atlanta who had to survey more than 100,000 square miles and could not keep up with all the weather activity.

The plane carried a wind-shear alert system, but the manufacturer, Honeywell, had programmed it not to sound an alarm while the flaps were being retracted, because that action can create turbulence that the system could mistake for wind shear, setting off a false alarm.

The USAir crew was retracting the flaps as part of the missed approach procedure.

Copyright © 2020, The Baltimore Sun, a Baltimore Sun Media Group publication | Place an Ad