by Brant Houston | Investigative News Network, Robert McClure | InvestigateWest, and Kevin Crowe | inewsource
A six-month examination of more than 150,000 reports filed by pilots and others in the aviation industry over the past 20 years reveals surprising and sometimes shocking safety breaches and close calls at local, regional and major airports throughout the country.
A consortium of journalists working at six nonprofit investigative centers across the U.S. — including inewsource — reviewed the records with Investigative News Network, of which they are members, and National Public Radio. To put the confidential reports into context, the journalists did extensive data analysis of the reports and conducted scores of interviews with pilots, air traffic controllers and aviation safety experts.
This review of the little-explored NASA records shows that the wide variety of problems translates into more than 130 near-mishaps and lapses reported on an average day, most happening unbeknownst to the flying public or those living near the airports.
And that’s just what’s publicly revealed; the actual number of reports is about five times as much, but budget constraints limit what NASA can explore in depth and some of the reports are not considered substantial enough to include. NASA maintains and analyzes the database, known as the Aviation Safety Reporting System, on behalf of the Federal Aviation Administration.
Nearly two-thirds of the reports list “human factors” as the root cause, according to an analysis by inewsource in San Diego.
NASA’s database is public, but is little-known and seldom explored by journalists. Some aviation experts and critics say the FAA itself, which contracted with NASA to create the database, does not pay enough attention to the results.
But FAA officials last year told congressional investigators that they would not otherwise know about 95 percent of the incidents revealed through NASA and a related system run by the airlines that feeds into NASA’s database.
While NASA and FAA assert that no formal statistical conclusions can be drawn from the database because of a presumed self-reporting “bias” and incomplete data, aviation safety experts and NASA maintain that it is a rich trove of information that helps prevent accidents and increase safety.
Indeed, the picture that emerges from an examination of the reports would cause any layman to ask: Is our air-transportation system as safe as we’ve been led to believe?
“Is it safe? Yes,” said Terry von Thaden, a Ph.D. air-safety researcher and private consultant at the University of Illinois Urbana-Champaign, who created a special system to measure the safety culture of aviation organizations.
But she adds: “Does stuff happen every day that you don’t want to know about? Yes. The fact that it works always amazes me.”
The state centers’ investigations, which examined records as far back as1988, found:
- In Colorado, reports by air-traffic controllers of major safety lapses have surged in the last year, even as the Government Accountability Office and Department of Transportation inspector general warned that FAA efforts to train a new generation of controllers to deal with record retirement levels is lagging, especially in Denver.
- In the New York City region, consistent problems were reported at smaller Connecticut airports because controllers at the regional level were late in “handing off” control of incoming airplanes to the local airports.
- In Massachusetts, the period from 1990 to 2010 saw about 500 cases in which aircraft were either damaged or pilots had to take evasive or emergency action to prevent accidents or damage to objects on the ground.
- In Wisconsin, as well as many other states, authorities are contending with an unusual problem: Deer, turkeys, coyotes and other wild animals are turning up on runways right in the path of planes.
- In San Diego, there were at least 45 accounts since 2000 by pilots who said they had to bank, dive or climb – sometimes violently – to avoid colliding with aircraft or terrain that in some cases came as close as 30 feet.
- And in the state of Washington, safety breaches were recorded more than two times a week on average – including an air-taxi pilot operating on three hours’ sleep who nearly flew into a plane leaving SeaTac International Airport near Seattle.
“Yes, we were tired and hurried, but that can be no excuse,” the pilot wrote in his report to NASA. “Thankfully this did not cause a (near-midair collision) or worse.”
NPR, meanwhile, has used NASA data to focus on the challenges pilots have had with automation systems on the flight deck, where they may see a loss of some of their manual flying skills. Also pilots may not fully understand the systems or get confused by them during flight.
NASA operates the system for the FAA because NASA has no ability to punish pilots and air-traffic controllers, who often are confessing to errors that otherwise could endanger their FAA licenses. The idea is to pinpoint causes of accidents and near-accidents so that the entire system can be made safer.
NASA, which says it has a small, under-funded staff that reviews the reports, removes the identity of those reporting and issues alerts on serious problems it finds. The database contains only 20 percent of reports filed, although the National Academy of Public Administration recommended in an FAA-funded study that 40 percent to 100 percent of the reports be processed.
While flight-related fatalities are limited – and indeed commercial airline companies reported none for the last two years – the reports to NASA disclose thousands of repeated errors that point to potentially serious flaws in the safeguards set up for the aviation industry
In fact, NASA issued more than 2,600 safety alerts to the industry based on the reports from 2000 through August 2010. That’s roughly one for every workday.
But the reports and alerts do not always lead to changes that will lower or eliminate some errors. Instead, von Thaden says, regulators run a “hot button” system that reacts only to serious and immediate problems rather than leading to investigations and solutions into systemic flaws.
The errors range from the seemingly laughable — a flight attendant on his second day of duty accidentally deployed an emergency-evacuation slide at Portland International Airport in 2007 – to the obviously dangerous, as when in 2008 a commercial airline pilot leaving SeaTac International Airport nearly hit a plane leaving nearby Boeing Field and complained the air traffic controller “cut it way too close.”
Based on the reports, NASA has identified top threats to air safety, which include “pilot pushing”: “Pilots feel pressured to take aircraft when they would not normally do so” –and “company budget pressures, unrealistic scheduling practices” and lack of proper rest.
Indeed, a report by the Investigative Reporting Workshop and Frontline last year found that “the major airlines use the regionals to help keep fares low, leading to low pay and sometimes onerous working conditions for pilots” and noted the role pilot fatigue can play in accidents.
In a report last May, the congressional watchdog Government Accountability Office faulted FAA for failing to use the NASA data and a related system maintained by the airline companies to spot national safety trends. Many of the airlines’ reports are forwarded to NASA through ASRS, but not all of them are.
The system catches “the big stuff – the hazards that prevent the accidents,” maintains Linda Connell of NASA, administrator of Aviation Safety Reporting System.
But the FAA said, in a prepared statement, that while the NASA reporting system “can be useful, it has some inherent limitations.
“The reports represent a subjective opinion or perception about an event and do not always include complete information,” the FAA statement said. “Because the reports are anonymous, the FAA cannot investigate or validate the data. In addition, NASA only publicly releases a small portion of the data, which makes it unreliable for statistical analysis purposes.”
Despite the shortcomings, the GAO echoed air-safety experts and concluded that NASA’s system and the related ones maintained by the airlines represent “the best source of information for hidden risk in the system.”
Of the reports, 62 percent come from commercial pilots, while private pilots file 26 percent and air-traffic controllers file 10 percent.
The pilots’ reports are filled with jargon and acronyms, but also contain casual narratives sometimes alarming in their informality or in the kind of surprisingly simple miscues they reveal.
For example, a regional airline pilot showed how fatigue can lead to absent-minded mistakes. In a 2006 incident, a pilot forgot to retract the aircraft’s landing gear after taking off for Cincinnati from Willard Airport in Champaign, Ill. He could not figure out what was causing the low rumbling noise underneath the plane’s nose and almost doubled back to land before the co-pilot noticed the pilot’s error.
“I was tired,” wrote the pilot, who had started his day early in Omaha. He noted Willard Airport is “a busy little airport” with lots of distractions and it is “easy to get out of your ‘groove.’ ”
Pilot fatigue “is a huge problem and it’s getting worse,” said von Thaden, the Illinois air-safety researcher, a finding reflected in the reports to NASA.
While pilot fatigue has garnered a lot of attention following the 2009 crash of a commuter airplane in upstate New York that killed 50, the NASA data points to a much broader universe of factors that come close to producing tragedy. They include:
- Comunications breakdowns can lead to episodes like this one : The pilot of a small plane landing at Sikorsky Airport in Bridgeport, Conn., noticed a larger air taxi bearing down from behind, less than a mile away and descending fast – a likely recipe for the two planes to collide on the runway. It turned out the air taxi had not been “handed off” as required to a local air traffic controller by a regional controller. To make matters worse, the local tower had no radar to show the large plane, the pilot of the small plane wrote in his report to NASA. Both planes aborted their landings.
- Equipment failures can be a factor: Many of the reports to NASA concern problems on various aircraft, as when a B777 crew had to make an emergency landing in 2009 because a foam liner between the fuel tank and an engine had worn away, causing what appeared to be a fuel leak.
- The relatively low level of training and experience among private pilots compared to commercial pilots. For example, a flight instructor showing a student pilot how to use a cockpit computer near Yakima, Wash., in 2008 was jarred into reality by a call from the local tower – with control-tower crash-avoidance alarms raging in the background – just in time to look out of the cockpit to see another small plane pass a bare 400 feet overhead.
Christopher Willenborg, head of the Massachusetts Department of Transportation’s aeronautics division, said issues such as runway safety and adequate pilot training have required closer scrutiny in the last five years. The NASA data, he said, could be an important avenue to averting tragedy in the future:
“I think as we move forward (these reports) will be something we will look into on a regular basis.”
Contributing to this story were the Connecticut Health Investigative Team, Investigate West, the National Institute for Computer-Assisted Reporting, New England Center for Investigative Reporting, NPR, the Rocky Mountain Investigative News Network, inewsource, and the Wisconsin Center for Investigative Reporting.
All are members of the Investigative News Network which is supported by the Ethics and Excellence in Journalism Foundation, the John S. and James L. Knight Foundation, the McCormick Foundation, the Open Society Foundations and philanthropist Buzz Woolley.