AeroSafety world
EMOTIONAL AWARENESS Essential element for CRM CYPRUS ATC CONFLICT Diplomats seek solutions TARGETED TRAINING Loss of control, dual HGS BIRD STRIKES Mitigating the hazard LIGHT UP THE NIGHT NIGHT VISION LIMITATIONS
The Journal of Flight Safety Foundation August 2010 BASS-Asia Business Aviation Safety Seminar-Asia November 9–11, 2010 Singapore Aviation Academy, Singapore
Endorsed by
Today’s best safety practices for the Asia Pacific region.
The rapid growth of business aviation in the Asia Pacific region represents opportunity for organizations and national economies.
As other regions have discovered, however, expansion is also a safety challenge. Fortunately, business aviation has already developed best practices that BASS-Asia can be applied in Asia Pacific.
is a new safety seminar, sponsored by four leading organizations to transmit practicable knowledge and techniques supporting safe flight.
To register or to see a preliminary agenda, go to
SINGAPORE
AVIATION AsBAA ACADEMY Asian Business Aviation Association EXECUTIVE’sMessage Changing Vision bout once a year I find a need to use a their training, and wasn’t even in a manual they famous quote. This time the urge came to could have read on their day off. me as I read a detailed analysis of the hu- It is not possible for pilots to know everything man factors surrounding Turkish Airlines about their advanced equipment, so instead they AFlight 1951 that, last year in March, crashed short watch the systems do their thing, day in and day of the runway at Amsterdam Schiphol. out, and come up with mental models about how In 1962, U.S. President John F. Kennedy wel- they think the systems work. That solution works comed a large group of Nobel laureates to a din- pretty well until the stick shaker activates early ner at the White House by saying, “This is the and dumps the aircraft in your lap, or the throttles most extraordinary collection of talent, of human retard themselves while you are trying to recover knowledge, that has ever been gathered together from a stall, or an engine goes into climb thrust at the White House, with the possible exception while you are trying to stop on a wet runway with of when Thomas Jefferson dined alone.” one reverser locked out. In the 18th century it was possible for a re- During the last several decades, complexity has markable individual like Thomas Jefferson to increased, and the pressure on the training system have some mastery over most areas of human has increased. Cost pressures force training to be knowledge. One hundred and sixty years later, done in the absolute minimum amount of time. even a room full of geniuses could not make a New training requirements are layered on top of similar claim. I have to wonder if the same thing one another and compete for what little training has happened to our own industry while we were time is available. That doesn’t leave much time to busy cutting costs. develop understanding of complicated systems. I would never claim to have the intellect of We need to consider revising the approach. Tradi- Thomas Jefferson, but when I was a young man tional training focuses on observable tasks and observ- studying for my Boeing 727 flight engineer exam, able outcomes. It is an approach that has its roots in the I felt like I had a pretty comprehensive knowledge time-and-motion studies of the late 1800s. This may not of how those aircraft systems operated. I am not be the way to prepare somebody to deal with a one-in-a- sure that pilots flying the latest generation of so- million event in an automated system. Maybe it is time phisticated aircraft feel the same way. to talk about things like education, understanding and Today, there is so much interconnected tech- insight, and change our perspective. The problems are nology, so many possible modes of operation; different today. It is time to adapt. it just doesn’t seem realistic to suggest that one person can know all that is going on. Yet, a misun- derstanding of how systems interact can be fatal. In the case of TK 1951, the pilots made a reasonable assumption about how the autothrottle system William R. Voss would respond with one radar altimeter out. They President and CEO guessed wrong. The right answer was never in Flight Safety Foundation www.flightsafety.org | AeroSafetyWorld | August 2010 | 1 AeroSafetyWorld
contents August2010 Vol 5 Issue 7 features
11 InSight | U.K. Runway Contamination
14 CoverStory | Night Vision Goggles
14 19 HumanFactors | Emotional Intelligence
24 CausalFactors | Night Spatial Disorientation
24 28 HelicopterSafety | FOQA for HEMS
30 FlightTraining | LOC, HGS and More
36 HumanFactors | Mystery Illness
40 TrafficControl | Cyprus Airspace Concerns
44 ThreatAnalysis | Bird Strike Mitigation departments
1 Executive’sMessage | Changing Vision
5 EditorialPage | Training Footprint 30 7 SafetyCalendar | Industry Events
2 | flight safety foundation | AeroSafetyWorld | August 2010 Istanbul FIR
Turkey
Ankara36 FIR Nicosia FIR
Area of concern Damascus FIR
Ercan Hellas UIR Cyprus Larnaca Paphos Nicosia FIR 40 44
Beirut FIR AeroSafetyWORLD telephone: +1 703.739.6700 Mediterranean8 InBrief Sea | Safety News William R. Voss, publisher, FSF president and CEO [email protected] 48 DataLink | Fewer Runway Excursions in 2009 J.A. Donoghue, editor-in-chief, FSF director of publications [email protected], ext. 116 52 InfoScan | Screen Test Cairo FIR Tel Aviv FIR Mark Lacagnina, senior editor [email protected], ext. 114 57 OnRecord | Landing on the Hudson Wayne Rosenkrans, senior editor [email protected], ext. 115 Linda Werfelman, senior editor [email protected], ext. 122 Rick Darby, associate editor [email protected], ext. 113 Karen K. Ehrlich, webmaster and production coordinator [email protected], ext. 117 Ann L. Mullikin, art director and designer [email protected], ext. 120 About the Cover The Santa Barbara County Fire Department Susan D. Reed, production specialist demonstrates night vision goggles. [email protected], ext. 123 © Chris Sorensen Photography Patricia Setze, librarian [email protected], ext. 103
We Encourage Reprints (For permissions, go to
www.flightsafety.org | AeroSafetyWorld | August 2010 | 3 Serving Aviation Safety Interests for More Than 60 Years
Officers and Staff light Safety Foundation is an international membership organization dedicated to Chairman, Board of Governors Lynn Brubaker the continuous improvement of aviation safety. Nonprofit and independent, the President and CEO William R. Voss FFoundation was launched officially in 1947 in response to the aviation industry’s need Executive Vice President Kevin L. Hiatt for a neutral clearinghouse to disseminate objective safety information, and for a credible General Counsel and knowledgeable body that would identify threats to safety, analyze the problems and and Secretary Kenneth P. Quinn, Esq. recommend practical solutions to them. Since its beginning, the Foundation has acted in the Treasurer David J. Barger public interest to produce positive influence on aviation safety. Today, the Foundation provides leadership to more than 1,040 individuals and member organizations in 128 countries. Financial Chief Financial Officer Penny Young Accountant Misty Holloway MemberGuide Flight Safety Foundation Headquarters: 601 Madison St., Suite 300, Alexandria, VA, 22314-1756 USA Membership tel: +1 703.739.6700 fax: +1 703.739.6708 Director of Membership flightsafety.org and Seminars Kelcey Mitchell Seminar and Exhibit Coordinator Namratha Apparao Membership Services Coordinator Ahlam Wahdan
Business Development Director of Development Susan M. Lausch
Communications Director of Communications Emily McGee Member enrollment ext. 102 Ahlam Wahdan, membership services coordinator [email protected] Seminar registration ext. 101 Technical Namratha Apparao, seminar and exhibit coordinator [email protected] Director of Seminar sponsorships/Exhibitor opportunities ext. 105 Technical Programs James M. Burin Kelcey Mitchell, director of membership and seminars [email protected] Technical Donations/Endowments ext. 112 Programs Specialist Norma Fields Susan M. Lausch, director of development [email protected] Manager of FSF awards programs ext. 105 Aviation Safety Audits Darol V. Holsman Kelcey Mitchell, director of membership and seminars [email protected] Technical Specialist/ Technical product orders ext. 101 Safety Auditor Robert Feeler Namratha Apparao, seminar and exhibit coordinator [email protected] Library services/seminar proceedings ext. 103 International Patricia Setze, librarian [email protected] Regional Director Paul Fox Web site ext. 117 Karen Ehrlich, webmaster and production coordinator [email protected] Past President Stuart Matthews Regional Office: GPO Box 3026 • Melbourne, Victoria 3001 Australia Telephone: +61 1300.557.162 • Fax +61 1300.557.182 Founder Jerome Lederer 1902–2004 Paul Fox, regional director [email protected]
flight safety foundation | AeroSafetyWorld | August 2010 Editorialpage
Training Footprint
ot long ago I would write stories training is one of many business elements operators, we are told, are distancing them- fairly regularly about aviation to account for in the budget cycle. selves from the idea that economic deci- training, probably a couple each Make no mistake, training for mod- sions play any role in developing training year. One of the most dependable ern flight operations is a pricey process, programs. If that were to be true, operators Nrecurring themes I would hear from train- involving travel, rooms and food, expen- would voluntarily boost systems training ing providers was the difficulty they had sive training technology and technicians, in nearly every phase of the process, from in dealing with the additional required and instructors. Plus, crews in training computer-based training to fixed-base training modules regulators continually are not working, so productivity is lost. training devices to the full-motion simula- loaded onto the operators. Some of this Meanwhile, our aircraft are becoming tors. But we won’t see much of this because was back in the old days of tombstone increasingly complex, often in the quest it is expensive, and the economic squeeze regulation when the industry knee-jerked to make flying safer. Perhaps counter- is still on for most operators. to do something, anything, after each ac- intuitively, experience and studies are Airlines and corporate flight depart- cident, but some of it also was the result of showing that the greater the amount of ments don’t want their complete training well-considered responses to developing automation, the more hidden become process mapped out for them, but if the knowledge and technology. dangerous failure modes. training footprint is not enlarged volun- The problem, the providers would The crew of the Turkish Airlines 737 tarily to deal with the complexities of mod- always say, was trying to fit the new that crashed short of the runway in Am- ern cockpit systems we may see regulators material into the existing training “foot- sterdam apparently didn’t realize what move more and more in that direction, print,” the investment in time and money it meant to the automation system as a especially should there be one or two more operators set aside for the process. They whole to have one of its two radar altim- high-profile accidents. With operators in weren’t worried about the validity of eters displaying an altitude below sea level a public state of denial about the role eco- the additional material as much as they (ASW, 6/10, p. 32). This is but one of many nomic considerations take in formulating a were concerned about which part of the examples. In addition, there seems to be training plan, I see a rough road ahead for existing training curriculum would get an increasing number of catastrophic cooperative efforts to boost systems train- compressed or even tossed out in order loss-of-control accidents in which mis- ing without a regulatory mandate. to make room for the new stuff. managed automation plays a role. The fixed training footprint is not the The key thought here is that it seems same for all operators or all types of opera- increasingly obvious that pilots need ad- tions, of course, but in most cases there ditional training in their aircraft’s systems are defined amounts of time and money. and their operation. J.A. Donoghue Managers who don’t make and adhere The point that ties the start of this Editor-in-Chief to a budget are not managers at all, and column to the automation part is that AeroSafety World www.flightsafety.org | AeroSafetyWorld | August 2010 | 5 help us make flying safer.
Your tax-deductible contribution to the Flight Safety Foundation endowment through the Combined Federal Campaign (CFC) supports the work of aviation safety professionals worldwide.
Flight Safety Foundation (flightsafety.org) is the only non-profit organization whose sole purpose is to provide independent, impartial, expert safety guidance and resources for the commercial and business aviation industry. Practices for safe operations are researched, initiated, and actively publicized and distributed by the Flight Safety Foundation.
However, there is always more to be done. This job is never “complete.” Please consider a gift to the Flight Safety Foundation Endowment.
Ask your CFC campaign manager or human resources department how to set up your contribution. The CFC is open to Federal civilian, postal and military employees. Your tax-deductible gifts to the Flight Safety Foundation endowment can be made by check or ongoing payroll deductions during the campaign season, September 1–December 15.
Use your powers of deduction to improve aviation safety.
Combined Federal Campaign Flight Safety Foundation CFC donor code #34228 ➤ safetycalendar
CALL FOR PAPERS ➤ 16th International SEPT. 13 ➤ Airworthiness Surveyor Theory SEPT. 24–25 ➤ A Practical Approach to Symposium on Aviation Psychology. Course. U.K. Civil Aviation Authority International. Safety Management Systems. Beyond Risk Wright State University and Air Force Research London Gatwick. Sandra Rigby,
www.flightsafety.org | AeroSafetyWorld | August 2010 | 7 inBrief Safety News Mexican Downgrade Proposed Penalties
he U.S. Federal Aviation Administration (FAA) has down- “While Mexico has been responsive to the FAA’s findings graded Mexico’s aviation safety rating to Category 2, after and has made significant improvements in recent months, it Tdetermining that its civil aviation authority does not meet was unable to fully comply with all of the international safety safety standards established by the International Civil Aviation standards,” the FAA said. “However … Mexico continues to Organization (ICAO). make progress. The FAA is committed to working closely with The downgrade from FAA Category 1 to Category 2 means the Mexican government and providing technical assistance to that Mexican air carriers may not establish new service to the help Mexico regain its Category 1 rating.” United States; they may continue their existing service, however. A Category 1 rating means that a country’s civil aviation authority meets all ICAO standards. Category 2 means that a country “either lacks laws or regulations necessary to oversee air carriers in accordance with international standards or that its civil aviation authority … is deficient in one or more areas, such as technical expertise, trained personnel, recordkeeping or inspection procedures,” the FAA said. After the FAA announcement, Aeromexico, a Mexican airline that flies to and from U.S. airports, issued a statement noting that the FAA’s action “does not refer to the level of safety of the airlines, nor does it reflect the safety of Aeromexico, which complies with the highest international standards of operational safety.”
Wikimedia
Rudder Pedal Limitations
iting two accidents involving protection from the adverse “potentially hazardous rudder pedal effects of a potential APC Cinputs,” the U.S. National Transpor- after rudder inputs at high tation Safety Board (NTSB) has asked airspeeds.” the European Aviation Safety Agency Both accidents cited by (EASA) to modify its certification speci- the NTSB involved wake fications to limit rudder pedal sensitivity. turbulence encounters during The NTSB recommended that EASA which pilots’ rudder inputs Certification Specifications for Large caused the vertical stabilizer Aeroplanes be modified to “ensure limit loads to be exceeded by safe handling qualities in the yaw axis a large margin. © BriYYZ/Flickr throughout the flight envelope.” The first accident was encountering wake turbulence from a 747 After the new standard has been the Nov. 12, 2001, crash of an American while climbing from Flight Level (FL) established, EASA should “review the Airlines Airbus A300 after takeoff from 360 (approximately 36,000 ft) to FL 370. designs of existing airplanes to deter- John F. Kennedy International Airport in The crew declared an emergency and mine if they meet the standard,” the New York. All 260 people in the airplane diverted to Calgary, Alberta. Three of the NTSB said. “For existing airplane de- were killed, along with five people on 88 people in the airplane were seriously signs that do not meet the standard, … the ground. The investigation revealed injured and 10 received minor injuries. EASA should determine if the airplanes that, during the encounter with the wake A subsequent accident analysis and would be adequately protected from the of a Boeing 747, the first officer “made simulation by Airbus found that the rear adverse effects of a potential aircraft- a series of full alternating rudder pedal vertical stabilizer attachment had expe- pilot coupling (APC) after rudder inputs before the airplane’s vertical stabi- rienced loads 29 percent greater than the inputs at all airspeeds. If adequate lizer and rudder separated in flight.” design limit, primarily as a result of “the protection does not exist, EASA should The second accident involved an flight crew’s series of alternating rudder require modifications, as necessary, to Air Canada A319, which experienced pedal inputs and … not the result of the provide the airplanes with increased an in-flight upset on Jan. 10, 2008, after wake turbulence,” the NTSB said.
8 | flight safety foundation | AeroSafetyWorld | August 2010 inBrief
Proposed Penalties Seat Belt Airbag
he U.S. Federal Avia- U.S. manufacturer of aircraft seat tion Administration restraints has received a supple- T(FAA) has proposed Amental type certificate from the Eu- a $230,000 civil penalty ropean Aviation Safety Agency (EASA) against Continental Airlines to allow the retrofitting of general avia- for allegedly operating a tion aircraft with its seat belt airbags. Boeing 767 on 22 revenue The AmSafe seat belt airbag already flights during a time when is being installed on 80 percent of new the airplane was not in com- single-engine general aviation aircraft pliance with U.S. Federal as standard equipment, the Phoenix- Aviation Regulations. © Steve Cole/iStockphoto.com based company said. It also has been The FAA says the airline installed in commercial aircraft at more replaced the 767’s nosewheel and tire assembly without installing a required axle than 50 airlines around the world. washer. The company describes the The agency has proposed smaller civil penalties against 11 companies for seat belt airbag as a “self-contained, allegedly violating U.S. Department of Transportation Hazardous Materials modular restraint system designed to Regulations. improve occupant protection from seri- The largest of the proposed penalties — $91,000 each — were against ous head-impact injury and to enhance Boston Scientific Corp. of Natick, Massachusetts, U.S., and Cardinal Health of one’s ability to exit the aircraft following Madison, Mississippi, for allegedly offering fiberboard boxes containing flam- an otherwise survivable accident.” mable liquids to DHL. In each case, DHL employees discovered the leaking The seatbelt airbag is deployed when package, the FAA said. the system’s sensors detect a crash. Am- Lesser penalties were proposed against nine other companies accused by the Safe says the device has saved more than FAA of similar violations. 17 lives since it was first installed in 2001.
SMS Recommendations
n aviation rule-making committee (ARC) has recom- The ARC characterized SMS as “the next step in the mended that the U.S. Federal Aviation Administration evolution of safety in aviation, based on processes and tools to A(FAA) issue regulations and guidance on the implementa- systematically identify hazards and mitigate the risk associ- tion of safety management systems (SMS). ated with those hazards.” The ARC, which developed its recommendations after It also noted that its recommendation is the first step in reviewing public comments that were submitted on an FAA what will be a lengthy rule-making process, “and it is clear the advance notice of proposed rulemaking (ANPRM), said that FAA has a lot of work to do before a proposal can be initiated.” regulations would be desirable, even though the FAA already The process will include development of a cost-benefit has issued advisory information on SMS development and analysis and an evaluation of alternative methods for small implementation. businesses that are subject to any new regulations.
In developing the regulations, the FAA should, among © Xiphias/Dreamstime.com other things, address methods of protecting SMS safety infor- mation and proprietary data against disclosure and inappropri- ate use, the ARC said. “Protecting safety information from use in litigation (dis- covery), Freedom of Information Act (FoIA) requests and FAA enforcement action is necessary to ensure the availability of this information, which is essential to SMS,” the ARC said in its rec- ommendations to the FAA. “The ARC believes that this issue can only be adequately addressed through legislation in the case of discovery, subpoena and FoIA requests. This protective legis- lation must be in place prior to promulgation of an SMS rule.”
www.flightsafety.org | AeroSafetyWorld | August 2010 | 9 inBrief
Human Factors Coordination Fatigue Risk Management
he two U.S. agencies responsible for he U.S. Federal Aviation human factors research relating to Administration (FAA) has set Timplementation of the Next Genera- Tan Oct. 31 deadline for U.S. tion Air Transportation System (NextGen) air carriers to submit a fatigue — the modernization of U.S. airspace risk management plan (FRMP) — have failed to establish a “cross-agency “outlining policies and procedures human factors plan,” the Government for reducing the risks of flight Accountability Office (GAO) says. crewmember fatigue and improv- The GAO praised the Federal Avia- ing flight crewmember alertness.” tion Administration (FAA) and National FAA Information for Opera- © bojan fatur/iStockphoto.com Aeronautics and Space Administration tors (InFO) bulletin 10013, issued for “coordinating their NextGen human in early August, said the FRMP would be required of air carriers operating under factors research efforts” but said that U.S. Federal Aviation Regulations Part 121. the absence of a coordination plan has The FAA said that it would issue another InFO and FAA Order 8900.1, Flight prevented the agencies from designating Standards Information Management System (FSIMS), to provide guidance on the the areas that should be the subject of development and implementation of an FRMP. upcoming research. In addition, the FAA has issued Advisory Circular (AC) 120-103, Fatigue Risk In a report to Congress, the GAO Management Systems for Aviation Safety, and amended versions of several related recommended that the FAA develop a ACs to discuss the development of an overall fatigue risk management system coordination plan and give the people (FRMS). in key positions the authority to set AC 120-103, which is not mandatory, describes the basic concepts of an FRMS priorities for human factors research. and its role in aviation operations, along with implementation guidelines.
Finances Linked to Safety Risks In Other News …
ir operators in Australia have factor cited by nearly 20 percent of those he Civil Aviation Safety Authority identified a shaky economy as the questioned. Other areas identified as of Australia (CASA) is updating its Agreatest risk to aviation safety, ac- presenting safety risks included issues Tmaintenance rules and expects cording to a survey by the Civil Aviation relating to operational personnel, the new package of regulations to take Safety Authority of Australia (CASA). “aircraft characteristics,” airport issues, effect in June 2011. The revisions are CASA surveyed 789 operators — a airspace issues and “lack of understand- intended to “provide Australian avia- number that excludes the 12 largest ing of safety management systems.” tion with clearer, more concise and regular public transport operators. Despite their concerns, 56 percent internationally harmonized mainte- Similar surveys are conducted annually of the operators said that they believe nance rules,” CASA said. … Data from to gather operational and safety data. aviation is “extremely safe” or “very safe,” the European Aviation Safety Agency About 30 percent of the operators CASA said. Two percent said that they be- (EASA) show that in 2009, there was questioned said that economic condi- lieve Australian aviation is “not very safe.” one fatal accident involving a com- tions and profitability represented the Most of the air operators respond- mercial air carrier operated by a com- greatest “potential risk to safety.” Second ing to the survey flew their aircraft fewer pany from an EASA member state. The on the list was “unsafe operators being than 1,000 hours per year. Twenty-two June 1 crash of an Air France Airbus allowed to continue operating” — a percent operated one aircraft, and 20 A330 over the South Atlantic killed 228 percent operated two aircraft. Fifty-eight people. … The International Associa- percent said that their operations in- tion of Flight Training Professionals volved carrying passengers. The survey is being formed, with plans to begin also found that half of these operators’ developing a database of global pilot fixed-wing aircraft were built before training best practices. More informa- 1980; most of their helicopters, however, tion is available from Robert Barnes at were manufactured after 1990.
Compiled and edited by Linda Werfelman.
10 | flight safety foundation | AeroSafetyWorld | August 2010 insight
ast winter’s unusually heavy International Civil Aviation Organiza- Uncertainty snowfall caused major disrup- tion (ICAO) recommendations that So, what information can pilots rely on tions at most U.K. airports. Many operations on contaminated runways when making decisions about operating scheduled airlines were obliged should be the exception and not the on runways that are not dry? Currently, Lto cancel services, while charter airlines norm. U.K. airports have a “back to U.K. Civil Aviation Publication (CAP) continued to fly, albeit with substantial black” policy, which means that con- 493, Manual of Air Traffic Services, delays. The financial implications for taminated runways must be cleared and states that braking action reports must the airlines and airport operators are then treated with deicing/anti-icing be issued in plain language for com- still difficult to gauge. However, with fluid to prevent further contamina- pacted snow and ice — for example, the benefit of hindsight, could the cur- tion. However, this may not always be as “good,” “medium” or “poor.” This is rent U.K. practices regarding opera- practical; tactical decisions on runway derived from a matrix based on friction tions with contaminated runways be closure are not taken lightly and are measuring devices first developed in improved? difficult to predict. Traditionally, our 1959 by the Nordic countries and later The U.K. Civil Aviation Author- benign winters and maritime airflow adopted by ICAO (Table 1, p. 12).1 CAP ity (CAA) currently complies with have rarely put this policy to the test. 493 also says that friction measuring
Contaminated BY DAVID THOMAS and Closed Will U.K. airports be ready for another winter onslaught? © Frank Robitaille/Airliners.net © Frank
www.flightsafety.org | AeroSafetyWorld | August 2010 | 11 insight
devices can produce inaccurate read- those readings shall be passed to pilots.” a braking action report is produced for ings in conditions of slush and thin Does this matter if the airport always pilots. This has not solved the problem deposits of wet snow — a phenomenon clears the runway surface? The answer completely; Norway still has runway ex- highlighted by an operators’ bulletin is yes. There can be a period of uncer- cursions. However, Avinor continues to issued by the U.K. CAA in 2006.2 tainty from the time the runway begins develop tools to deal with this complex It has been known for some years to become contaminated to the time subject, the most recent being the Inte- that readings by friction measuring the airport decides to close it. Likewise, grated Runway Information System, a devices do not necessarily reflect the when the runway is reopened, it probably computer program that will aid airport braking performance of a modern will be wet with deicing/anti-icing fluid, personnel in assessing the runway state airliner and that the devices can produce which should equate to a braking action and braking action, based on automatic differing results. The Norwegian Ac- of “good.” However, under certain narrow meteorological measurements. cident Investigation Board has found temperature-dew point splits at or below that measurements can vary by 0.10 with freezing, ice can form when the deicing/ Across the Pond dry contaminants and by 0.20 with wet anti-icing fluid starts to break down, On the other side of the ocean, the contaminants. These issues are currently which may reduce the braking action philosophy with regard to braking ac- being addressed by research committees to “poor.” In these scenarios, the crew tion reports differs between the FAA formed by ICAO, the European Aviation will have to make an assessment of the and Transport Canada (TC). The FAA Safety Agency, the U.S. Federal Aviation likely runway braking action without any recognizes the difficulty of assessing Administration (FAA) and others. meaningful data. Snow notice to airmen the surface condition of contaminated So, where do we stand with regard to (SNOWTAM) code “9” and European runways and reporting the information to braking action reports if the runway is aviation routine weather report (METAR) pilots. It also acknowledges that the data contaminated with something other than codes “//” and “99” indicate that runway provided by friction measuring devices compacted snow or ice? The answer lies friction measurements are “unreliable.” do not necessarily represent aircraft in CAP 493: “In conditions of slush or Consequently, should we ask the braking performance. Consequently, the regulator to rewrite CAP 493 FAA recently recommended that airport Runway Friction Measurements to allow braking reports to operators no longer provide Mu readings be passed to pilots under all (measured friction coefficients) to pilots. Estimated MOTNE Measured or Calculated Braking METAR conditions? The Norwegian It believes that pilot weather reports Coefficient of Friction Action Code CAA already has done this by (PIREPs) are an invaluable source of 0.40 and above Good 95 adapting the ICAO recom- information for pilots and should be used 0.39–0.36 Medium/ 94 mendations to the Norwe- in support of runway condition reports. good gian winter climate. This has After the Chicago Midway runway excur- 0.35–0.30 Medium 93 enabled the Norwegian airport sion in 2005 (ASW, 2/08, p. 28), the FAA 0.29–0.26 Medium/ 92 poor operator Avinor to develop a set up a workshop on runway condition 0.25 and below Poor 91 reporting matrix for its own reporting. Participants developed a table If for any reason the — 99 environmental conditions. Air- that correlates braking action reports reading is considered port personnel are trained to with estimated runway surface conditions unreliable make an assessment based on a (Table 2). The table has been provided MOTNE = Meteorological Operational Telecommunication visual inspection of the runway to pilots by Boeing and is now used by a Network Europe; METAR = aviation routine weather report to measure the contaminant, number of U.K. airlines. Source: U.K. Civil Aviation Publication 493 friction measurements (which TC has eliminated some of the Table 1 cannot be solely relied upon), issues caused by conflicting read- current weather conditions and ings from friction measuring devices thin deposits of wet snow, friction mea- runway maintenance activities such as by using only decelerometers. The suring devices can produce inaccurate treatment with deicing/anti-icing fluid, measurements conform to Canadian readings. [Therefore,] no plain language sand, etc. After the results of the assess- Runway Friction Index (CRFI) values estimates of braking action derived from ment are interpreted using the matrix, comprising mostly fractions from 0 to
12 | flight safety foundation | AeroSafetyWorld | August 2010 insight
Braking Action Correlations*
Braking Action Estimated Correlations ICAO Term Definition Runway Surface Condition Code Mu Good Braking deceleration is normal for the wheel braking Water depth of 1/8 in or less 5 40 and effort applied. Directional control is normal. Dry snow less than 3/4 in depth above Compacted snow with OAT at or below –15º C Good to Medium — 4 39–36 Medium (Fair) Braking deceleration is noticeably reduced for the Dry snow 3/4 in or greater in depth 3 35 –30 wheel braking effort applied. Directional control Sanded snow may be slightly reduced. Sanded ice Compacted snow with OAT above –15º C Medium to Poor — 2 29–26 Poor Braking deceleration is significantly reduced for Wet snow 1 25–21 the wheel braking effort applied. Potential for Slush hydroplaning exists. Directional control may be significantly reduced. Water depth more than 1/8 in Ice (not melting) Nil Braking deceleration is minimal to non-existent Ice (melting ) — 20 and for the wheel braking effort applied. Directional Wet ice below control may be uncertain. Note: Taxi, takeoff and landing operations in nil conditions are prohibited.
ICAO = International Civil Aviation Organization; OAT = outside air temperature * The correlations are estimates, only. Mu values — reported runway friction coefficients — can vary significantly.
Source: Boeing Commercial Airplanes
Table 2
1, with 1 being theoretically equivalent codes // or 99, crews should be pro- Notes to maximum friction on a dry runway. vided with a similar level of safety 1. Friction measuring devices include continu- Although TC has considerable confi- from the airport operator as would be ous friction measuring equipment and spot dence in this system, some contami- expected under normal conditions. measuring equipment (decelerometers). nants, including slush and loose snow, This is something British crews are 2. U.K. CAA. Flight Operations Division remain outside the system’s capabilities. likely to receive when operating at air- Communication (FODCOM) 19/2006, The Transportation Safety Board of fields with traditionally harsher win- Winter Operations. Oct. 30, 2006. Canada (TSB) forwarded an aviation ters. Unless the regulator changes its 3. TSB Aviation Investigation Report safety advisory to TC after a runway policy on when braking action reports A02A0038. Runway Excursion: Air Canada excursion in 2002.3 As a result of the can be issued, airport operators are Regional Airlines (Jazz) Fokker F-26 MK- 1000, C-FCRK, Saint John, New Brunswick, recommendations made in the adviso- unlikely to invest in new tools to help 27 March 2002. ry, TC now highlights the limitations of assess braking action. The easy option runway surface condition reports and is to continue with the status quo and InSight is a forum for expressing personal opinions CRFI reports, particularly when ambi- hope last winter was one in a million. about issues of importance to aviation safety ent temperatures are near freezing. However, if it was not and next winter and for stimulating constructive discussion, pro Contaminated runway opera- we have a serious runway excursion, and con, about the expressed opinions. Send tions will always be the exception in who will be accountable? your comments to J.A. Donoghue, director of the United Kingdom due to our David Thomas is a captain for Thomas Cook publications, Flight Safety Foundation, 601 climate, and clearing should be the Airlines. This article originally was published Madison St., Suite 300, Alexandria VA 22314- first option. However, when operating by the British Airline Pilots Association in its 1756 USA or [email protected]. under SNOWTAM code 9 or METAR bimonthly journal, “The Log.” www.flightsafety.org | AeroSafetyWorld | August 2010 | 13 Coverstory Lighting Up the Night
Along with their advantages, NVGs also have limitations, and pilots and operators must be thoroughly informed about both.
BY CLARENCE E. RASH Aviation Specialties Unlimited (ASU) Specialties Unlimited Aviation
sers of night vision goggles Brighter Nights that should be, but often are not, ap- (NVGs) are acutely aware of the The brighter scene provided by NVGs plied to NVG flight: detection range advantages these devices offer — which makes it possible for pilots to and recognition range. Detection in improving a pilot’s ability to see objects not otherwise visible — in- range is the distance at which the pres- seeU in darkness and enhancing safety creases situational awareness, enhances ence of an object can be discovered; during night flight. They may be less safety and improves flight capability. recognition range is the distance at cognizant, however, of some of the However, as impressive as these which a detected object can be identi- limitations that NVGs impose on night devices are at increasing the ability to see fied as belonging to a category, such as visual performance. and fly at night, the technology, as many wires, buildings, vehicles or people. As NVG use in helicopter opera- researchers say, “does not turn night tions increases, pilots and operators into day.” Unfortunately, too many pilots Limitations must be educated about the capabilities carry a mental model of daytime flight The consequences of specific limita- and limitations imposed by NVGs, and into their night operations, not being tions of NVGs (Table 1, p. 16) can be the often-misleading effects of NVG aware that, even with NVGs, their visual significant, and their cumulative effects imagery on visual perception. performance is compromised. in degrading night visual performance Since 1999, when the U.S. Federal Perhaps the most common mistake provide pilots with a challenging flight Aviation Administration (FAA) issued by pilots flying with NVGs is “overfly- environment.3 the first supplemental type certificate ing” the aircraft — flying too fast to al- For example, because pilots rely (STC) to permit use of NVGs by a low for adequate reaction to the sudden on the quantity and quality of visual civilian helicopter emergency medical detection of an obstacle. information available to them to make services (HEMS) operator in the United Such a scenario raises the need for decisions that are integral to maintain- States, NVG use has steadily grown.1,2 two additional operational metrics ing safe flight conditions, the NVG’s
14 | flight safety foundation | AeroSafetyWorld | August 2010 coverStory
Although NVGs reduced field of view (FOV) and resolution are any flight, it is not uncommon for available enable pilots the most significant limitations. resolution to be as good as 20/25 and as poor as to see objects The NVG’s 40-degree circular FOV is a completely washed out image. in the dark that smaller than the normal human binocular In addition to reduced FOV and resolution, are not visible to visual field of 120 degrees (vertical) by 200 NVGs have additional limitations that include the naked eye, degrees (horizontal). Pilots describe their im- reduced depth perception, loss of color informa- they do not “turn pression of viewing the outside world through tion and the presence of image noise — which night into day.” NVGs as “looking through a soda straw.” To looks sparkly and obscures fine detail — and compensate for this reduced FOV, pilots must other defects. continuously scan from side to side, as well as up and down. This is fatiguing, and on long External vs. Internal Light flights, pilots may fail to maintain the scan. Al- Perhaps the most overlooked limitation of though unaided side and “look-under” vision NVGs is their inability to discriminate between is important to scan instruments and identify light originating from the external world and colors of lights outside the cockpit, unaided light originating inside the cockpit. side vision also is important in detecting other NVGs have an “automatic gain control (AGC),” A new generation aircraft outside the NVG FOV. which reacts to the ambient light level by increas- of cockpit lighting NVG resolution, which describes the ing the multiplication factor when the ambient design will allow for amount of detail in a scene, has greatly im- light level decreases and decreasing the multiplica- internal viewing of proved since the earliest NVGs were manufac- tion factor when the ambient light level increases. instruments but will tured. Those early devices gave pilots visual As a result, if the lower light levels in the cockpit not artificially lower acuity of approximately 20/50 — or the metric can be “seen” within the FOV of the NVGs, then NVG performance. equivalent, 6/15.4 the AGC reduces the system gain. This results in a system gain that is not optimized for the external illumination level, possibly reducing the pilot’s night vision capability. This dilemma has driven a new generation of cockpit lighting design, one that will allow for internal viewing of instruments but will not artificially lower NVG performance.5
Weighing the Advantages Operators must weigh the advantages and disad- vantages before deciding whether to implement an NVG system. CareFlight, an Australian HEMS operator, recommends in its NVG Implementation Guide that any operator contemplating implementa- tion of an NVG system first conduct a detailed analysis and formulate a business plan.6 Aviation Specialties Unlimited (ASU) “Operators considering the NVG technol- ogy often have to justify a significant invest- Modern systems provide resolution equiva- ment and expenditure without having … any lent to 20/25 (6/8) visual acuity. However, clear way of determining the suitability and obtaining this high resolution requires optimal benefits to their particular operation, let alone a environmental conditions, including high il- method to determine the implementation costs,” lumination and contrast, clear weather, and an the guide says. “If after the analysis phase, it absence of fog, dust and glare sources. During is decided that NVG technology is justified in
www.flightsafety.org | AeroSafetyWorld | August 2010 | 15 Coverstory
the particular organization, operators are then piece of equipment and a fundamentally new confronted with trying to implement both a new cultural shift.” It outlines a 13-step plan that begins with an NVG Limitations and Their Effects on Performance analysis of night mission data — “night scene landing conditions, search requirements, terrain Visual Limitation Effects on Performance encountered and frequency of night operations” Reduced field of view (FOV) 40 degrees circular (normal FOV is 120 degrees by 200 degrees) — to help determine whether NVGs would ben- Reduced resolution (visual acuity) Early systems (20/40); newer systems (20/25), efit the operation. Other steps include determin- but greatly dependent on ambient lighting1 ing the availability and cost of NVG technology; Loss of color information Typically shades of green (or white) against assessing client perceptions and expectations; black background deciding what cockpit modifications are required, Degraded standard night vision Reduced light adaptation resulting from NVG their costs and how they should be implemented; imagery in eyes and outlining NVG training requirements. Presence of halos Although halo sizes have been reduced in newer systems, bright lights appear The guide recommends training one or two surrounded by a glow (halo)2 crews, which then fly for at least three months Distortion Reduced binocular depth perception; using NVGs before the operator evaluates their problematic only in older devices experiences and determines whether changes Reduced depth perception Decreased ability to judge distances; can are needed before other crews undergo training. induce visual illusions “Remember, NVGs will only be of benefit if Presence of image “noise” Obscures fine details; problem increases as ambient light level decreases; appears as [their] implementation and ongoing manage- sparkles or scintillations ment are properly resourced and structured,” the Image defects Can cause distractions; obscures fine detail; document says. defined as various cosmetic blemishes in the NVG imagery resulting from dirt or debris trapped in the system during the manufacturing Education and Training process (e.g., black spots and white spots) Once a decision has been made to implement an NVG = night vision goggles NVG system, operators and pilots must be thor- Notes oughly familiar with the advantages and disadvan- 1. 20/40 vision (metric equivalent 6/12) refers to a person’s ability to see clearly from a distance of 20 ft (6 m) what someone with normal vision sees from 40 ft (12 m). The metric tages of flight using NVGs. This can be achieved equivalent of 20/25 vision is 6/8. through educational courses covering the essen- 2. The primary concern involves halos that are significantly different in size between right and left NVG tubes and within the FOV. The images do not fuse and appear out of focus. tials of night vision technologies, offered by the
Source: Clarence E. Rash operator or an outside training company. These courses should consist of material that describes Table 1 the basic principle, design, operation, and care and maintenance of NVGs. FAA requirements also call for instruction in relevant aeromedical factors such as depth perception, range estimation and vi- sual illusions; scene and terrain interpretation; and abnormal operational characteristics of NVGs. An educational program is not necessarily limited to classroom lectures but may also include use of an eye lane — in which a pilot stands at one side of a dark room and looks through NVGs at an eye chart on the opposite wall to learn to focus the goggles; a terrain board — a miniature layout of the type of terrain where the pilot will operate; or a simulator, as well as computer-based Aviation Specialties Unlimited (ASU) Specialties Unlimited Aviation or Web-based training.7
16 | flight safety foundation | AeroSafetyWorld | August 2010 coverStory
Following an initial classroom introduction to the principles and limitations of NVGs, the next step is to allow pilots to experience these limitations firsthand via operational flights. The chief goal of an effective flight training program is to expose pilots to the perceptual differences in NVG-aided night flight, compared with Aviation Specialties Unlimited (ASU) Specialties Unlimited Aviation unaided day flight, to dispel any misconception that NVGs can turn night into day. Flight training should be conducted by a qualified NVG pilot and should include both ba- sic and mission-specific tasks and maneuvers, in- cluding NVG operational checks and the impact of internal/external lighting systems on NVG performance; airspace surveillance and obstacle avoidance; departures and approaches, with and As with any electro-optical system, sus- without NVGs; NVG malfunction procedures; tained proper operation requires regular recovery from inadvertent entry into instrument inspections. On a nightly basis, pilots should meteorological conditions; and transitioning conduct a brief preflight operational inspec- between NVG-aided flight and unaided flight.8 tion. First, NVGs should be checked for Research and experience show that pilots functionality — checking battery installation need early and continued exposure to the night and tube luminance balance — and for obvious environment across a broad range of op- How NVGs Work erational conditions and environments to he night vision goggles (NVGs) used in improvements in I² tube performance — re- develop good night civil aviation rely on image intensification ferred to as generations (GEN). The current flying skills and (I2) technology to convert both visual light commonly available version, which was fielded 9 T practices. — which can be seen by the naked eye — and in the 1980s, is referred to as GEN III. GEN III+ near-infrared (IR) light — which cannot — to was developed in 2001 and was intended to Hardware electrons, which are multiplied (amplified), and be designated as GEN IV (filmless), but was converted back into visible light. changed back to a thin film design. Technical NVG hardware All aviation NVGs that are in common use are characteristics of GEN III+ NVGs include a fully considerations fall binocular, helmet-mounted systems with two I² overlapped, binocular 40-degree circular field- into three catego- tubes and a dual compartment power pack that of-view, and a resolution designed to result in ries: procurement, gives the pilot immediate backup power. The visual acuity of 20/25 (6/8).1 inspection, and power pack uses AA alkaline batteries. User adjustments are provided for fore-aft maintenance and NVG tubes have three basic components positioning, vertical height, tilt, interpupillary — a photocathode, a microchannel plate and distance, and both objective and eyepiece repair. Available a phosphor screen. All three are sandwiched focus. If one or more of these adjustments is systems may include between two sets of optical elements — in- incorrect, NVG imagery can be degraded. the earliest genera- put optics that focus the incoming photons — Clarence E. Rash tion (GEN) of NVGs, onto the photocathode and an eyepiece that Note or they may be the focuses the outgoing photons into the eye. Since their introduction into military avia- 1. 20/25 vision (metric equivalent 6/8) refers to a newest — GEN III+ person’s ability to see clearly from a distance of tion in the 1970s and their integration into civil intensifier tubes; 20 ft (6 m) what someone with normal vision aviation in the 1990s, NVGs have undergone sees from 25 ft (8 m). Normal vision typically is they are priced ac- several design changes — mostly based on considered to be 20/20 (6/6). cordingly (see “How NVGs Work,”). www.flightsafety.org | AeroSafetyWorld | August 2010 | 17 Coverstory
damage, such as a loose mounting Guidelines and Regulations 3. Rash, C.E.; Manning, S.D. “On the Flight bracket, broken knobs/levers or loose The FAA and other civil aviation Deck, Lighting Must Satisfy a Variety wires. Second, all user adjustments regulatory agencies around the world of Needs.” Human Factors & Aviation Medicine Volume 50 (September– should be verified as functional and have recognized the advantages of October 2003). then optimized by the pilot for his or using NVG devices in civil aviation to her requirements. Most important of enhance situational awareness dur- 4. 20/50 vision (metric equivalent 6/15) re- these is the focus setting. If a com- ing night operations. Standardized fers to a person’s ability to see clearly from a distance of 20 ft (6 m) what someone mercial NVG focusing apparatus is terminology, policies and practices are with normal vision sees from 50 ft (15 m). available, it will produce the most pre- essential for the efficient and effec- Normal vision typically is considered to be cise focus setting; otherwise, focusing tive incorporation of NVGs into civil 20/20 (6/6). should be performed according to the aviation — and this is only possible 5. Rash, Manning. manufacturer’s instructions. Finally, through government regulation. cockpit lighting should be viewed Progress in developing comprehen- 6. CareFlight. NVG Implementation Guide. through the NVGs to ensure compat- sive regulations and guidelines has been Northmead, New South Wales, Australia. ible instrument lighting, dimming as slow, spanning the nearly two decades 7. Ruffner, J.W.; Antonio, J.C.; Joralmon, required. since NVGs began appearing in civilian D.Q.; Martin, E. “Night Vision Goggle The FAA and NVG manufacturers helicopters. Training Technologies and Situational recommend that NVGs be inspected Nonetheless, over this period, the Awareness.” Proceedings of Advanced Technology Electronic Defense System every 180 days. FAA has been soliciting and incorpo- Conference/Tactical Situational Awareness Routine care and maintenance by us- rating recommendations from vari- Symposium. San Diego. 2004. ers is necessary to reduce problems dur- ous aviation organizations. In 1999, ing regular use. NVGs should be handled as a collaborative effort involving the 8. New requirement per August 2009 modifi- cation to U.S. Federal Aviation Regulations like any device that has delicate optical FAA and RTCA,10 along with EU- Part 61.31(k), “Additional training re- components in which optical alignment ROCAE, the European Organisation quired for night vision goggle operations.” is essential to proper operation. for Civil Aviation Equipment, special When NVGs are not in use, the lens committees were formed to develop 9. Ruffner et al. caps should be in place, and the device guidance for introducing NVGs into 10. RTCA — organized in 1935 as the Radio should be stored in its case to reduce civil aviation. This effort produced Technical Commission for Aeronautics the possibility of shock and damage. three guidance documents.11 In ad- but now known only by the abbreviation Batteries should be removed if the dition, in September 2004, the FAA — is a nonprofit corporation that develops consensus-based recommendations device will not be used for an extended published a technical standard order regarding communications, navigation, period. Regular care should include that discussed minimum performance surveillance, and air traffic management cleaning lenses with high quality lens standards.12 system issues. RTCA functions as a federal advisory committee, and its recommenda- cleaning supplies and wiping the exte- Clarence E. Rash is a research physicist with 30 tions are used by the FAA as a basis for rior with a soft cloth to remove dirt. years experience in military aviation research policy, program and regulatory decisions. If a system is suspected of being and development. He has authored more than defective, repairs should be performed 200 papers on aviation display, human factors 11. The documents are DO-268, Concept only by certified repair personnel. Us- and aircraft protection topics. of Operations, Night Vision Imaging System for Civil Operators, published in ers should never attempt to disassemble Notes 2001; DO-275, Minimum Operational NVGs. 1. Salazar, G.J.; Nakagawara, V.B. “Night Performance Standards for Integrated A logbook should be used for each Vision Goggles in Civilian Aviation.” FAA Night Vision Imaging System Equipment, set of NVGs and should contain a record Aviation News Volume 40 (May–June published in 2001; and DO-295, Civil of hours of usage, reported problems, 2001): 21–22. Operators’ Training Guidelines for inspection and calibration dates, and re- Integrated Night Vision Imaging System 2. FAA. Fact Sheet: Helicopter Emergency Equipment, published in 2004. pairs. A record of battery use will ensure Medical Service Safety. June 8, 2010. that extremely fatigued batteries will not
18 | flight safety foundation | AeroSafetyWorld | August 2010 Enabled Emotionally sky, another 50people, killing while aircraft good out toperfectly fall of the into aBuffalo, New York, U.S., suburb. heard of Colgan Air Flight 3407 crashing listened in horror a month later when we a safe landing in the Hudson River, and W an entire crew’s mentalstate, notjustyourown. ‘Emotional Intelligence’meansbeingaware of www.flightsafety.org Among factors the that caused one US Airways Flight 1549 to piloted in early 2009skillfully when Chesley Sullenberger e watched in astonishment |
orld World AeroSafety |
A ugust 2010 ugust aircraft incidents and accidents is is accidents and incidents aircraft emotional and mental states. clearly, aware being while of crew’s the trol his own emotions inorder to think to linked captain’s the be ability to con- success or failure to alarge degree can obviously play amajor role. However, minor injuries, skills flying technical landing that resulted inonly afew crippled aircraftvery made asafe water When role the pilots play in By ShariFrisinger flight experience? flight of hours total Howmany type? aircraft the in have they did hours flight many How training? recurrent last their was abilitiestechnical of pilots: the When (NTSB) and many analysts is on the Board Safety Transportation National U.S. the of focus initial the considered, desirable traits acaptain should have. only the not are skills technical that But some ago time it was realized 1 HUMANF actors |
19 © Silense/iStockphoto humanfactors
The word “emotion” may conjure up nega- tive elements that tend to degrade safety: an- ger, fear, crying, shouting and other unhelpful behaviors, but everyone every day experi- ences more subtle varieties of emotion.2 In the cockpit this might include satisfaction for di Filippo/iStockphoto © Luca having achieved a smooth landing, pride in maneuvering around turbulence, excitement in getting desirable days off, irritation when plans don’t work out, and sometimes annoy- ance with others. Regardless of the situation, there always exists some degree of emotional response, and emotions are simply another type of informa- tion that must be considered in making effective decisions, especially in a team environment. A major factor in Many years ago, airlines implemented cockpit A high degree of situational awareness re- resource management (CRM) techniques to lies on a person being attentive to the environ- maintaining the enhance crew coordination. This new concept ment. Internal situational awareness consists of was partially based on a U.S. National Aero- understanding one’s own emotions and emo- safety of the crew nautics and Space Administration investiga- tional triggers. External situational awareness and passengers is tion that discovered a common theme in many involves insights into team members’ moods accidents — failure of leadership and ineffective and unspoken communication, and appropri- the combination of crew interaction. ately addressing them. CRM focused on how the crew interacted The cornerstones of emotional intelligence the leader’s objective in the cockpit, not necessarily on acceptable (EI) are consciousness of one’s thoughts and thought process and or appropriate cockpit behaviors. During the moods, of how the behaviors resulting from first decade of CRM use, it morphed into crew those impact and influence others, and of the his or her emotional resource management, to include helping all moods and behaviors of others.3 People with a crewmembers work more effectively as a team, high level of EI recognize and control their own awareness. improving situational awareness and providing emotional outbursts, step back from the heat of techniques to break the error chain. any situation, analyze it objectively and take the CRM has become a training mainstay. appropriate action that produces the most desir- To date, CRM has included only the techni- able results. cal skills and thinking abilities — analytical, A person’s perception of reality shapes emo- conceptual and problem solving. However, tions and feelings, and these drive thoughts and research beginning in the 1980s demonstrated behaviors. Status quo is maintained until new that emotions greatly influence a person’s strong feelings are experienced. Simply being cognitive abilities. unhappy in a job is usually not enough to war- To be effective, the next level of CRM rant a change. Getting passed over for a promo- needs to include more of the “people” side tion, accompanied by the belief that the decision — self-confidence, teamwork, cooperation, was wrong, usually sparks anger and an active empathy and flexibility in thoughts and ac- job pursuit. tions. A major factor in maintaining the safety The amygdala is the part of the brain that of the crew and passengers is the combination controls a person’s level of emotional reactiv- of the leader’s objective thought process and ity. It never matures, and, if left unchecked, his or her emotional awareness. it can bring chaos to a life. To compound the
20 | flight safety foundation | AeroSafetyWorld | August 2010 HUMANFactors
problem, the human brain instinctively can- that argument, was to defend yourself. Success is not distinguish between a real threat and an more assured when this emotionally downward- imagined one. spiraling thinking is halted and the problem is Sitting in a theater, watching a panoramic addressed more creatively. or 3-D movie, the sudden loud sound of an The captain in the Colgan Air 3407 ac- airplane approaching will make most people cident chose the “flight” reaction; he chose to reflexively duck. Intellectually, they know the avoid a developing situation.4 When the first airplane is not real, but the emotional brain officer brought up the icing conditions — “I’ve hears the loud sound and tells the body it needs never seen icing conditions. I’ve never deiced. to avoid getting hit. When a situation changes, I’ve never seen any, … I’ve never experienced the emotional brain determines if the stimu- any of that” — the captain’s response was, lus causing the change is a threat. If a threat “Yeah, uh, I spent the first three months in, is sensed, awareness becomes heightened and uh, Charleston, West Virginia and, uh, flew physiological changes take place to cope with but I — first couple of times I saw the amount this new danger. Adrenaline is released to pump of ice that that Saab would pick up and keep the heart faster and prime the muscles for ac- on truckin’ … I’m a Florida man … .” Then he tion. If the situation is later deemed to not be a added, “There wasn’t — we never had to make The next level of threat, logic and objectivity take over again, but decisions that I wouldn’t have been able to it takes four hours for the adrenaline to dissipate make but ... now I’m more comfortable.” The CRM needs to from the body. captain was still unaware of what was rapidly include more of Today’s fears, threats and dangers are developing around him, chatting while the not unlike those of prehistoric man. A flight aircraft’s airspeed rapidly decayed. His failure the ‘people’ side department manager who needs to justify the to quiet his instinctive emotions narrowed his expenses of his department can experience the perception to the point that airspeed, one of — self-confidence, same “fight or flight” reaction that the caveman the most basic elements of flying an airplane, teamwork, did when faced with a saber-toothed tiger. A no longer had his attention. similar reaction occurs when people feel their There were few instances when the captain cooperation, reputation or credibility is threatened. Fear and referred to the first officer’s health. He did not stress envelop thinking and people ask how she felt about her ability to empathy and over-focus on a narrow selection perform her flight duties, of solutions, disregarding alterna- even though she flexibility. tive approaches. When people allow their stressed brains to overtake thoughts, the perspective narrows and the main focus becomes escaping from the situation. Unable to think of alterna- tives, they don’t see the “big picture” or question assumptions. At this level of thought, perception sneezed of the complexity of the situation becomes para- twice and six lyzing, and the focus is on current limitations. minutes later, she Remember the last time you became angry dur- mentioned her ears. ing an argument? It probably wasn’t until later, Basic understanding of after you could see the situation without emo- CRM and crew performance tion, that you thought of several obvious points should have tipped off the captain that could have helped your case. These become that the first officer was not feeling apparent because your rational mind was back well that day and her performance could
© David Marchal/iStockphoto© David in control. Your primary focus, in the midst of be negatively impacted. A person with higher
www.flightsafety.org | AeroSafetyWorld | August 2010 | 21 humanfactors
EI could have recognized that, and probably monitor properly the aircraft’s fuel state and to would have been empathic to her condition properly respond to the low fuel state and the and her inability to actively participate as a crewmembers’ advisories regarding fuel state. viable crewmember. … His inattention resulted from preoccupation The captain told stories for most of the flight. with a landing gear malfunction and prepara- At one point, he rambled for over three minutes tions for a possible landing emergency.” while the first officer only said 34 words, most of This accident was one of the key events driv- which were “yeah” and “uh-huh.” Research on how ing the adoption of CRM in airline training. the mind processes information has revealed that Contrast the reactions and situational aware- people can only consciously execute one task at a ness of the Colgan and United crews to those of time, and unconsciously perform one additional the captain of the US Airways A320 that landed task. When driving in heavy traffic or merging in the Hudson River. Sullenberger kept his emo- onto a freeway, are you able to continue your tions under control and remained focused on conversation? Your mind moves from the doing his job — to safely land the plane. conversation you were having to The captain’s words “my airplane” when looking at traffic, calcu- he took over the controls after the bird strike lating vehicle could have been trigger words, words to focus © Silense/iStockphoto speeds on, snapping his rational brain into action and putting him into a safety frame of mind. He repeated the commands from the first officer, indicating that during those critical seconds there was no room for any misunderstanding. This flight crew’s emotional intelligence was as good as it gets, which enabled their processing information quickly and using every resource available to them at the time. The captain of United Airlines Flight 232, Captains infected and analyzing the best opportunity to speed up a McDonnell Douglas DC-10 that in 1989 and merge. Your automatic mind does not have the attempted to land in Sioux City, Iowa, U.S., with ‘captainitis’ are ability to safely handle non-routine driving tasks. with catastrophic hydraulic and flight control A classic example is United Airlines Flight systems failures, could have reacted to his so absorbed in their 173, a McDonnell Douglas DC-8, which in challenges by becoming indecisive, shutting own world that they 1978 was destroyed when it crashed during an out the crew or dictating orders to them.6 If approach to Portland (Oregon, U.S.) Interna- he had responded in any of these ways, the lose their situational tional Airport.5 The captain’s intense preoc- captain would have reflected the emotional cupation with arranging for a safe emergency pressures he was experiencing, and, as a awareness. landing prohibited him from considering result, his crew would have had his pressures other anomalies. His concentration was so added to their own. Instead, he worked as focused on the emergency landing check- part of the crew, alternating between giving list that he did not modify his plans when direction and explaining his actions and tak- the first officer and flight engineer twice ing input from anyone in the cockpit, includ- warned him about their airplane’s dwindling ing a training pilot. Emotions are contagious, fuel supply. Ten people were killed when the and the strongest expressed emotion will be aircraft crashed into a wooded area due to fuel felt unconsciously by others and mimicked. exhaustion. In this case, the captain’s calm demeanor was The NTSB said, “The probable cause of mirrored by the crew and they were able to the accident was the failure of the captain to contain their emotional reactivity.
22 | flight safety foundation | AeroSafetyWorld | August 2010 HUMANFactors
Aviation history is overflowing There are several techniques that can chances are slim that the discus- with accidents due to pilot error. Many raise your level of EI: sion will end well. of them could have been avoided if the Leaders need a considerable amount of crews were more aware of their own • Be aware of the thoughts going cognition.7 The ability of the leader to emotional reactivity and those of the through your mind. Are they broaden his or her focus from technical others. Captains infected with “cap- stuck in the past and wallowing and task-related activities to include an tainitis” are so absorbed in their own in problems, or are they focused awareness of the moods of the crew is world that they lose their situational on the future and actively looking critical to success. It would benefit all awareness. The captain in Colgan Air for solutions? Once we choose parties to know which skills in specific 3407 was self-absorbed, talking about negative thoughts, they can very circumstances are most appropriate. himself for nearly 20 minutes of the easily spiral downward, the cycle A leader’s behaviors directly affect the last 40 minutes of the flight, missing a descending into hopelessness. team’s disposition, and the team’s dis- number of clues that eventually led to • Acknowledge your emotions. position drives performance. When the the crash; on the other hand, the cap- Remember they are neither leader can analyze and manage his or tain of US Airways 1549 maintained good nor bad, they are what her own emotional reactivity, the team his composure throughout his short they are. Next, identify these members can more easily manage their flight and focused on every element of emotions: Angry? Irritated? De- own emotions. How well the leader the emergency. fensive? Disappointed? Guilty? performs this can have a direct effect Why is EI relevant? The Center Frantic? Miserable? Naming on the safety and morale of the crew. for Creative Leadership found that your emotions makes them less the leading causes of failure among Shari Frisinger, president of CornerStone abstract and helps release their business executives are inadequate Strategies,
eparting from Douala, Cam- Confusion and spatial disorientation repeatedly by instructors and examin- eroon, on a dark night that prevailed when a bank angle warning ers at the airline. was beset by thunderstorms, a sounded. The captain responded with Both pilots were Kenyans. The cap- Kenya Airways Boeing 737-800 erratic flight control inputs that ag- tain, 52, held type ratings in several 737 Dentered a slow right roll that continued gravated the situation and precipitated models, as well as the Airbus A310-300. for nearly a minute without the flight a spiral dive. The pilots were wrestling He had 8,682 flight hours, including 823 crew or the autopilot engaged. The the controls when the 737 disintegrated hours as a 737 pilot-in-command. The captain — the pilot flying — was preoc- in a mangrove swamp, killing all 114 final report by the technical commission cupied with the weather and had lost people aboard. of inquiry said that after he received his situational awareness. The first officer, A technical commission of inquiry initial 737 type rating in 1997, “recurring who had been left out of the loop of the convened by the Republic of Camer- shortcomings” were cited by instructors captain’s planning, was not effectively oon found that the events leading to and examiners. They included defi- monitoring what was going on, and the May 5, 2007, accident were fraught ciencies in crew resource management he did not notice that the autopilot with deficiencies in pilot perfor- (CRM), knowledge of airplane sys- had not been engaged, as intended. mance that had been brought to light tems, adherence to standard operating
BY MARK LACAGNINA Beyond Redemption Spatial disorientation turned a minor upset into a major accident.
24 | flight safety foundation | AeroSafetyWorld | August 2010 Causalfactors
that he tended to be “overbearing.” A 2006 line check found the captain’s performance Wikimedia below standard and required that he undergo another line check; the training manager told the examiner to determine if “complacency or incompetence is the issue.” The captain passed the second line check, and “there was no evidence of any retraining or punitive action taken against him,” the report said. An examin- Ivory Coast er’s report on a proficiency check three months Douala before the accident contained no comments. Abidjan Cameroon Kenya Kenya Airways evaluated pilot performance Nairobi as “not acceptable,” “acceptable,” “standard” (average) or “above standard.” The captain’s per- formance consistently was judged as acceptable. The first officer, 23, had 831 flight hours, including 170 hours in type. Performance assess- ments issued during his training as a 737 first The accident officer cited requirements for improvement in airplane (left) was situational awareness, radio communications, on a scheduled monitoring the pilot flying and calling out devia- flight from the tions. “However, his overall performance during Ivory Coast to training and flight checks was judged to be satis- Kenya, with an factory,” the report said. He earned a 737 second- en route stop in-command type rating in September 2006. in Cameroon. The report said that Kenya Airways should procedures (SOPs), cockpit scanning, situational have avoided pairing the pilots for flight duty be- awareness, planning and decision making. The cause of the deficiencies observed during training pilot’s performance was found to be unsatisfac- and evaluation, and because of known psycho- tory during some proficiency checks, and he was logical traits. The captain was described as having required to receive extra training before another a strong character and a heightened ego, and was check was administered. known to be authoritative and domineering with A proficiency check in 2002 was converted subordinates. “He seems to have been affected by to a training flight after the captain demon- the slow progress of his career and the fact that he strated inadequate knowledge of systems and had remained on the 737,” the report said. His at- procedures. During recurrent training in 2003, titude toward the first officer during the accident an instructor urged him to be more attentive flight was described as “paternalistic.” to checklists and aircraft limitations, be more In contrast, the first officer was known to systematic in responding to system failures, be reserved and nonassertive. The report said provide more consistent briefings and adhere that during the accident flight, he appeared to be to SOPs. A 2004 training session resulted in “intimidated by the meteorological situation” and recommendations that he take time to analyze “subdued by the strong personality of the captain.” system failures and to discuss them with the first officer. A 2005 line proficiency check cited ‘Disquieting Conditions’ deficiencies in the captain’s command ability The accident occurred during a scheduled flight, and teamwork, and his familiarity with airplane KQA 507, from Abidjan, Ivory Coast, to Nairobi,
© Montague Smith/Airliners.net systems and SOPs; the examiner also noted Kenya, with a one-hour stop at Douala, on the
www.flightsafety.org | AeroSafetyWorld | August 2010 | 25 Causalfactors
conditions beyond the runway. The departure clearance required that they maintain runway
heading during the Orozco/Airliners.net Tena © Luis initial climb, and they agreed that a devia- tion south of the as- signed course would avoid the largest cells. The captain told the controller, “After departure, we would like to maintain slightly left of run- way heading due to weather ahead.” The first officer corrected the captain, who then Neither pilot noticed fifth day of the crew’s five-day pairing. The 737 radioed, “Sorry, slightly right.” The controller that the autopilot did landed in Douala on schedule at 2201 local time, approved the request. not engage when the but the crew was told by the airport traffic con- The captain, the pilot flying, then initiated “CMD” (command) troller to taxi to a terminal gate that was differ- the takeoff although neither he nor the first offi- push button, on ent from the one at which the ground crew was cer had requested or received clearance from the the right side of the waiting. That miscue, plus the captain’s decision controller. The flight directors and autothrottle mode control panel to postpone engine start until heavy rain abated, were engaged in the takeoff/go-around mode, at the top this photo, delayed the departure about one hour. with a selected heading of 118 degrees and a se- was pressed. The report said that before leaving the gate lected climb speed of 150 kt. The airplane lifted shortly after midnight, the captain did not con- off at 0006. duct a departure briefing, which was required especially in light of the “rather disquieting me- Tendency to Roll Right teorological conditions.” When he called for the The report said that the airplane had a ten- “Before Taxi” checklist, the first officer began dency to roll right likely because of a slight conducting the “Before Takeoff” checklist; the mispositioning of the rudder trim control captain did not correct him. and/or because of an inherent trim asym- Weather conditions were influenced by metry that resulted during its construction in thunderstorm activity that had begun in the 2006.1 The captain used his control wheel to afternoon and was forecast to last until morn- counter the roll tendency until the selected ing. A storm had passed over the airport while heading was changed from 118 degrees to the 737 was parked at the gate, and there were 139 degrees as the airplane climbed through storms in the vicinity of the airport. When the 1,000 ft. He told the first officer, “I will keep crew taxied onto Runway 12 for takeoff, vis- somewhere around here.” About this time, all ibility was 800 m (1/2 mi) in rain, and surface action on the flight controls ceased, the report winds were from 050 at 10 kt, gusting to 20 kt. said, noting that the crew’s attention likely While holding for takeoff, the pilots used was focused on using the weather radar to the on-board weather radar to analyze the avoid thunderstorms.
26 | flight safety foundation | AeroSafetyWorld | August 2010 Causalfactors
The airplane was banked 11 degrees through 2,600 ft when the first offi- when it entered a rapid spiral dive that right, turning through a heading of 127 cer asked, “I continue with heading?” was precipitated by the captain’s rud- degrees and climbing through 1,600 ft The captain did not respond, but the der inputs. The first officer apparently when climb power was selected and the selected heading was changed from 120 recognized what needed to be done to captain said, “OK, command.” This was degrees to 165 degrees. recover but mistakenly called, “Right, a reference to engagement of the autopi- captain.” He then corrected himself, lot, which, according to company SOPs, ‘We Are Crashing’ exclaiming, “Left, left, left … correc- is accomplished and called out by the At 0007:19, the captain uttered an ex- tion, left.” pilot flying and cross-checked by the pi- pression of surprise when the enhanced At this point, the pilots made con- lot monitoring. However, the first officer ground-proximity warning system gen- flicting control movements. The first said nothing in response to the captain’s erated an aural warning: “bank angle.” officer tried to roll left and lower the callout, and neither pilot noticed that the A flight simulation later conducted by nose while the captain held full right flight mode annunciator indicated that Boeing showed that the airplane easily roll and nose-up pitch control. “The the autopilot was not engaged. could have been returned to wings-level first officer’s action was corrective, Recorded flight data showed that flight without excessive control inputs while the captain’s action was aggra- the “CMD” (command) push button at this point — if the pilot flying was vating, but the situation was already on the mode control panel had indeed not spatially disoriented. However, the beyond redemption,” the report said. been pressed to engage the autopilot, report said that the captain responded The 737 was in a 60-degree right but the report said that the autopilot with control inputs that were “erratic.” bank and a 48-degree nose-down likely failed to engage because forward He moved the control wheel 22 degrees pitch attitude, and descending at pressure was being applied manually to right, 20 degrees left, 45 degrees right 14,000 fpm with an airspeed of 287 kt, a control column at the same time. The and then 11 degrees left. when it struck terrain at 0007:42. The report said that the captain’s subsequent The airplane was banked 50 degrees emergency locator transmitter was behavior indicated that he believed the right and was climbing through 2,770 damaged on impact and did not emit autopilot was engaged and that trim ft at 0007:23, when the captain again a usable signal; the wreckage was not inputs effected by the autothrottle in attempted to engage the autopilot by found until 1730. response to excessive airspeed changes pressing the “CMD” button on the Among the recommendations might have reinforced that impression. mode control panel. However, because generated by the accident investigation The airplane was climbing through the flight director roll-command bars was that all pilots should receive formal 2,400 ft when the air traffic control- were more than one-half scale from upset recovery training. ler issued a new altimeter setting. center, the autopilot engaged in the This article is based on the technical com- “The two pilots executed the change control wheel steering mode. Thus, mission of inquiry’s report, “Technical of altimeter setting without noticing the autopilot did not respond to the Investigation Into the Accident of the B737-800 or interpreting the deteriorating flight selected flight modes; its sole input was Registration 5Y-KYA Operated by Kenya parameters that were clearly visible on to reduce the bank angle to 30 degrees. Airways That Occurred on the 5th of May 2007 in Douala.” The full report is avail- the EADI [electronic attitude director The report said that the captain able at
Data BY LINDA WERFELMAN Development Air Methods
A Foundation-backed project will use FOQA data to fight HEMS safety risks.
light Safety Foundation and Air of Manny’s Sonoma Aviation, a full- accidents, and 29 fatalities, according Methods, the world’s largest air service fixed base operation in Santa to data developed by Ira Blumen, M.D., ambulance operator, have begun Rosa, California, U.S. — is designed program and medical director of the a two-year project designed to to identify safety risks in HEMS University of Chicago Aeromedical Fuse flight operational quality assurance operations and develop procedures to Network.1 More recently, a published (FOQA) — sometimes known as flight eliminate them. report cited 11 crashes of EMS helicop- data monitoring — to gather safety The HEMS industry in the United ters in the United States since Septem- information on helicopter emergency States has experienced a surge in ac- ber 2009; those accidents resulted in 19 medical services (HEMS) operations. cidents in recent years, including 146 fatalities.2 The project — financed through HEMS crashes between 1998 and 2008 In response, government and pri- a grant to the Foundation from the — 50 of them fatal. In 2008 alone, there vate agencies have conducted studies estate of Manuel S. Maciel, founder were 13 accidents, including nine fatal and issued recommendations. Flight
28 | flight safety foundation | AeroSafetyWorld | August 2010 HelicopterSafety
Safety Foundation has been among them, with that the concepts developed by the IHST also the release in 2009 of the Industry Risk Profile will ultimately help individual HEMS opera- developed by Aerosafe Risk Management, which tors to implement a safety management system criticized the widely varying standards and con- (SMS) — typically described as a predictive flicting practices throughout the industry.3 mode of managing safety in which data collec- The risk profile was especially critical of tion and analysis enable risks to be identified the absence of a “publicly visible accountability and addressed before they cause an accident or structure for the industry,” the variations in serious incident. standards for HEMS professionals and the “lack Air Methods already has implemented an of confidence by the stakeholders that effective SMS, and FOQA will be a major component of health care can be effectively delivered.” that initiative, Lugger said. The new study by the Foundation and Air The U.S. Federal Aviation Administration Methods, along with companies in other seg- (FAA) estimates that there are about 840 EMS ments of the aviation industry — including helicopters in operation nationwide. The FAA Aerobytes, Appareo Systems, L-3 Communica- has calculated the HEMS fatal accident rate at tions and the Office and Professional Employ- 1.18 per 100,000 flight hours, compared with ees International Union, AFL-CIO — will use 1.13 per 100,000 flight hours for all general FOQA to develop a better picture of the proce- aviation and air taxi flights, 1.0 for turboshaft dures being used by HEMS operators and pilots. helicopters and 1.94 for all piston helicopters. “We don’t know exactly how these pilots However, the FAA said, “the number of HEMS are flying their approaches to the landing sites,” accidents nearly doubled between the mid- said Robert Vandel, a retired FSF executive 1990s and the HEMS industry’s rapid growth vice president who is now a Foundation fellow period from 2000 to 2004.” working on the study. “We can’t begin to figure The U.S. National Transportation Safety out how to solve this problem because we don’t Board (NTSB), which issues an annual “Most know exactly what’s going on, considering the Wanted” list of transportation safety im- divergence of standard operating procedures. provements, added a new category in 2008: We need to get enough data to analyze and then improving the safety of EMS flights. The assemble experts — safety personnel, pilots and NTSB’s specific recommendations call for manufacturers — to identify the best practices stricter regulations for EMS flights conducted and improve what we can.” with medical personnel aboard; implement- Eric Lugger, Air Methods corporate safety ing flight-risk evaluation programs for EMS manager, said that plans call for the company operations; requiring formalized dispatch and to use 10 of its helicopters in the study, retrofit- flight-following procedures, including current ting all 10 with quick access recorders. After the weather information; and installing terrain recorders have been installed, a one-year data- awareness and warning systems (TAWS) on Quick access collection period will begin, followed by a pe- EMS aircraft. recorders, like the riod of data analysis and writing a report on the Notes one pictured here, study, he said. The goal, Lugger said, is “quality are being installed improvement in the operation of the aircraft.” 1. Werfelman, Linda. “Closing the Loop.” AeroSafety in Air Methods Vandel said that the study should be com- World Volume 4 (March 2009): 14–18. helicopters as part plete in mid-2012 and that, “by then, we’ll have 2. Levin, Alan. “Medevac Industry Opposing Upgrades of a FOQA data data to build on.” Wanted by NTSB.” USA Today, Aug. 19, 2010. development project. The study’s findings will be used to help the 3. Aerosafe Risk Management. Helicopter Emergency International Helicopter Safety Team (IHST) in Medical Services (HEMS) Industry Risk Profile. April developing recommendations to improve the 2009. L-3 Communications L-3 overall safety of HEMS operations. Vandel said Risk_Profile.pdf>. www.flightsafety.org | AeroSafetyWorld | August 2010 | 29 FlightTRAINING © Embraer By Wayne Rosenkrans | From Orlando special purposes Dual HGS, practical stall scenarios and unreliable-airspeed rehearsals help pilots manage risk of airplane loss of control. efenses against loss of con- During the World Aviation Training pilot understanding of aerodynamics trol in flight (LOC-I) figured Conference and Trade Show (WATS and the adoption of updated stall recov- prominently in recent confer- 2010), presenters from two early- ery guidance from Airbus and Boeing ence presentations to pilot adopter airlines also outlined a related Commercial Airplanes. Dtraining specialists from major and rationale for equipping fleets with dual Several presenters concurred that regional airlines. A recurrent theme head-up guidance systems (HGS) and despite measurable improvements since was how to apply lessons learned mentioned plans for HGS-qualified first the 1980s, the LOC-I trend of the last from transport airplane accidents officers to land aircraft after approaches five years should ring alarms (Table 1, that happened in unremarkable flight to Category III minimums. Others at and Figure 1, p. 32). “If we are going conditions with properly functioning the April gathering in Orlando, Florida, to lower the overall accident rate, we autoflight systems. U.S., called for stronger emphasis on have to address loss of control in flight 30 | flight safety foundation | AeroSafetyWorld | August 2010 FlightTRAINING because it stands alone as the largest But in his opinion, only about 10 percent let the airplane slow down, [wait for] threat,” said John Cox, president and of the pilots he polled in 2009 and 2010 stick shaker [to activate], add power CEO of Safety Operating Systems. demonstrated an adequate knowledge of and do not lose a foot of altitude,” Cox aerodynamics, the limitations of LOC-I explained. “Those were the criteria; Regaining Control training in flight simulation training people are still teaching that flawed Avoidance, recognition and recovery devices (FSTDs) and recent changes in approach. It’s flawed because if pilots remain essential elements of the LOC-I the response to stall indications recom- don’t reduce angle-of-attack and don’t solution, Cox said. “The first and most mended for large commercial jets. accept some altitude loss to quickly get critical skill is avoidance — how we “One of the things the industry has flow reattachment to the wing, they are teach crews to not put the airplane in taught [inappropriately in FSTDs] is not maximizing the aerodynamic per- a state that has the potential for upset ‘power out’ recoveries,” he said. “We need formance and they are decreasing the and how to recognize when it is near an to rethink this because there are parts likelihood of a successful outcome.” upset condition,” he explained. “[The in- of the flight envelope — particularly in The new aspect is the commitment dustry must teach] not only the incom- high altitude and high drag conditions to accept some altitude loss as a matter ing, next-generation pilots but equally — where pilots do not have excess thrust, of survival. “We should make sure that — or even more importantly — the and the airplane will not accelerate out of the pilots in flight decks today — and cadre of pilots on the flight deck today.” a stall. There is a high drag coefficient at the incoming generation — learn this A “flawed impression” may exist the critical angle-of-attack near stall, and well,” Cox added. “In April 2010, the in the airline industry of the baseline powering out may not always be possible. U.K. Civil Aviation Authority issued understanding of aerodynamics among ‘Powering out’ certainly was the way I a flight crew training notice in which today’s average pilots, he added, citing was taught to fly Boeing 737s in 1981, so they said ‘reduce angle-of-attack; it is answers he heard to questions about we may have been teaching the wrong the primary stall recovery step.’ aerodynamics that he posed infor- recovery for a long time.” “In recent weeks, Airbus and Boeing mally to professional pilots from various In those days of 737 training, the have changed their stall-recovery backgrounds. “Up to a point, they hardest maneuver was to set up the procedures, and I commend both understood the potential consequences airplane for what he used to call a organizations. [The updated procedure] of high–angle-of-attack flight,” Cox said. “precision stall.” “Trim it carefully, is to reduce angle-of-attack, lower the Fatal Accidents: Loss of Control in Flight vs. CFIT Worldwide Commercial Jets, Selected Periods, 1988–2009 Period 1988–1993 1991–1995 1992–2001 1993–2002 1994–2003 1997–2006 1998–2007 1999–2008 2000–2009 Percentage of all fatal accidents in period LOC-I 34.23 27.1 27.7 25.7 30.5 21.3 24.4 24.2 22.5 CFIT 36.8 28.8 24.1 22.9 22.9 22.5 20.0 18.7 18.0 Percentage of all fatalities in period (onboard and external if reported) LOC-I 25.5 39.2 34.3 31.5 39.3 30.4 36.7 40.1 35.2 CFIT 53.6 32.1 31.1 29.6 25.0 30.7 21.0 19.3 18.3 Rate of fatal accidents in period (per million departures; multiply value shown by 10–7) LOC-I 3.29 2.27 1.90 1.69 1.90 1.06 1.21 1.18 1.06 CFIT 3.54 2.41 1.66 1.51 1.42 1.12 0.99 0.91 0.84 CFIT = Controlled flight into or toward terrain; LOC-I = Loss of control – in flight Note: Periods are examples selected from 1988–2009 editions of the Boeing Commercial Airplanes Statistical Summary of Commercial Jet Aircraft Accidents. Editions vary in the number of years covered, accident-category definitions and onboard/external fatalities counted. Source: Adapted from Boeing Commercial Airplanes Table 1 www.flightsafety.org | AeroSafetyWorld | August 2010 | 31 FlightTRAINING nose, level the wings beforehand that unreliable airspeed could occur Fatal Accidents: and increase thrust, any time during the simulated flight from Prague Loss of Control in Flight vs. CFIT understanding that to Moscow. Elements included the auxiliary 25 with engines mounted power unit inoperative per provisions of the min- under the wing, the imum equipment list (MEL) and assignment of a 20 pilot may be adding to departure runway with a tail wind. “We wanted 15 the [resulting] nose- them to come to our recurrent training already 10 up pitch. That may prepared,” Hurych said. “Our target was to show have to be countered, pilots the behavior of the aircraft and let them 5 Loss of control–in ight so pilots may want practice solving this very difficult situation. They Percentage of accidents Percentage CFIT 0 to be a bit judicious were advised to use all airplane documentation 1997–2006 1998–2007 1999–2008 in how rapidly they … including an Airbus presentation on unreliable Running average apply thrust with very speed, which is of great value.” CFIT = controlled flight into or toward terrain high bypass–ratio fan Early in this simulator exercise, the instruc- Note: Data are percentages of total fatal accidents involving engines. It is a con- tor inserts a frozen standby pitot tube condition worldwide commercial jets, based on data published by Boeing Commercial Airplanes. sideration. [Then the and thunderstorms on the weather radar display. Source: John M. Cox pilot should] reduce Later, in cruise flight, the instructor inserts speed brakes or retract simultaneous faults on both airspeed channels Figure 1 them, and return to along with an air data reference–frozen fault. normal flight. … There is a caveat: If the airplane “If this appeared shortly before reaching the manufacturer has specific [actions for known] [assigned] flight level, the pilot flying still had flight characteristics, follow them first.” the speed indication, but unfortunately it was wrong,” Hurych said. Unreliable Airspeed “Shortly after, the crew lost all the [air- Early avenues of inquiry into the Air France speed] indications, had to start with the memo- Flight 447 accident investigation — the June 1, ry items and then had to revert to the paper 2009, crash of an Airbus A330 in the Atlantic checklists for unreliable speed indication. The Ocean — prompted Czech Airlines to recon- scenario’s intent was for the flight crew to bring sider how it trains pilots to recognize and safely the aircraft back to Prague. The emergency was handle situations involving unreliable airspeed, declared, and while using the paper checklists, according to Roman Hurych, a captain and chief the crew began their descent in preparation for flight instructor for the company. Recurrent approach and landing.” line-oriented flight training (LOFT) with an Czech Airlines found that the advance brief- unreliable speed indication began in September ing and pre-exercise preparation made all crews 2009, and ironically around the time the train- hyperattentive to any airspeed fluctuation as a ing was introduced, an actual event occurred. possible anomaly, and some began troubleshoot- “We reacted very quickly [to Flight 447],” he ing suspected unreliable-airspeed indications said. “We realized it could happen to anybody at caused by normal turbulence encounters during any time. … We also had to admit that the last climb. “They knew what was to happen, but time the [typical] pilot operated the aircraft with they didn’t know when,” he said. “We saw crews unreliable speed was during his or her type rating comparing indications that they almost never course. So that was the main reason, to give all of compare and monitor during normal line flying. our pilots the chance to practice again how to fly Generally, all crews came very well prepared for the aircraft without the speed indication and, at the session and coped very proficiently with all the same time, to fly manually at high altitude.” [aspects] of the scenario.” The airline designed the FSTD scenario The real incident in late 2009 also occurred so flight crews would be briefed a few days during a flight from Prague to Moscow. During 32 | flight safety foundation | AeroSafetyWorld | August 2010 FlightTRAINING climb on autopilot, the crew noticed their altitude stall as a surprise … if at all possible. We will use modes disappear, and then the autopilot discon- any [tactic] necessary for distraction.” nected. Airspeed on one side showed 170 kt while A U.S. Federal Aviation Administration the other side showed 210 kt. “An instructor in (FAA)–industry working group that studied stall the right seat took over the controls and contin- training was concerned that misconceptions ued to climb out using the initial pitch and thrust about the practical implications of the agency’s as per memory items,” Hurych said. “At about practical test standards could amount to negative thrust-reduction altitude, they were in clean con- training. “The practical test standards say that the figuration because they had retracted flaps before applicant for a pilot certificate ‘recovers to a refer- recognizing the speed discrepancy.” ence airspeed, altitude and heading with a mini- Effective crew resource management helped mal loss of altitude,’” Kolisch said. “We do our the crew to maintain control, compute pitch approach-to-stalls [in airplanes] at practical high and thrust values for level off, perform actions and low altitudes, including at 400 ft.” Mesaba’s on the paper checklist, declare an emergency, turn back to Prague and complete an unevent- 39000 ful landing, he said. The cause of this unreliable HDG airspeed was still under investigation as of April. M.683 ALT The basic procedure being taught in recurrent 200 240 AP 20 training, however, worked as advertised. FD1 100 Approach to Stall 220 10 G In just a five-week period in 2009, three fatal S 000 airline accidents involving stalls occurred while 0 39 flight crews were flying approaches to land with 200 the autopilot engaged, said Paul Kolisch, manager, 900 flight operations training, Mesaba Airlines. “My T 10 contention is that these pilots were not trained for 180 800 these events,” he said. “I don’t know any pilot in the airline business, or operating sophisticated corpo- 240 rate airplanes, who has arrived at an inadvertent 29.92IN stall while hand flying the airplane. … Traditional [FSTD] stall training has shifted to an artificial Source: Paul Kolisch, Mesaba Airlines; Illustration: Susan Reed choreography where the pilot stops trimming in order to keep good control of the airplane during training strongly emphasizes the “recovers to a Exposure to high- the recovery, sits there and waits until [the stick reference altitude” and de-emphasizes “minimal,” altitude stalls in an shaker activates], then recovers. Not one of the which it considers difficult to define. No train- FSTD will lack some 2009 accidents happened that way.” ing injuries or fatalities have resulted despite the fidelity but reinforce Mesaba has adopted what it calls “practi- intentional distractions that startle pilots, he said. awareness of the risk cal training for approach to stalls” from a One concern of FAA-industry committee factors in line flying. conviction that unrealistic traditional training members was the fidelity gap between the stall generates unsafe expectations of what actually characteristics of typical FSTDs and aircraft will occur. “We do the training primarily in a performance in stalls, he said. “I am opposed classroom or briefing room prior to going into to trusting computer ‘speculation’ when we fly the simulator,” Kolisch said. “When we get into the simulators — we just don’t know how the the simulator, we don’t do a ‘stall series.’ Those airplane would behave,” Kolisch said. two words don’t occur together in our syllabus. “If we don’t take pilots up high in these jet [Instead,] at some point, the pilot encounters the airplane simulators, they won’t understand www.flightsafety.org | AeroSafetyWorld | August 2010 | 33 FlightTRAINING this [gap]. … If the first time they experience a and aircraft speed; an acceleration carat that high-altitude stall event is in an airplane, they’re transforms to an energy symbol; and tail-strike going to be in big trouble.” Based on review of advisory information. “Every time the pilots Mesaba’s videos, the opinion of some airplane look out of the [forward windshield through the upset specialists was that some stall recoveries combiner (i.e., the HUD screen)], they see the that were successful in an FSTD would have energy status for precise energy management of been an airplane upset in reality, he added. the aircraft with the flight path,” Kemény said. “After landing, we have the same capability as Head Up Constantly an instantaneous indication of the aircraft brake JetBlue since 2007 has deployed dual Rockwell performance. This is an immediate decision- Collins HGS-5600 HGSs on its Embraer 190 making tool; after the touchdown, the pilots regional jets. From the last quarter of 2009 have the picture of deceleration values.” through the first quarter of 2010, Lufthansa If unsafe deceleration on a contaminated CityLine partnered with the company to do the runway or brake problems occur, the flight crew same for its 190/195 fleet after more than three sees the remaining runway and the point at and a half years of preparation, said Christof which the aircraft will stop — rather than rely- Kemény, a captain with Lufthansa CityLine, in a ing on imprecise sensations that something is joint presentation with Mark Maskiell, a captain going wrong after landing, he added. with JetBlue. The systems are now used rou- During HGS training, Lufthansa CityLine tinely by all pilots in all weather conditions, and had to encourage pilots “to be patient with safety enhancement remains high on their lists themselves” as they advanced through four of objectives, they said. levels of proficiency (Table 2), learning the new Lufthansa had analyzed advantages and skills and presentation of the world outside the disadvantages of HGS compared with autoland aircraft. About six months typically elapse from systems and envisioned how HGS could be used the first day in an HGS FSTD to line flying at in the context of air traffic management trans- level 4, in which the pilot is fully “proficient us- formations imminent in Europe, the United ing HUD as another flight deck tool,” he said. “We see States and elsewhere. The decision to replace autoland with dual much less Kemény cited findings of the most recent HGS worked out as expected, Kemény said. Flight Safety Foundation study of safety benefits “Analysis of results showed that we have in- deviation on from HGS technology (ASW, 5/10, p. 38) as a creased situational awareness for both pilots,” reinforcement of Lufthansa CityLine’s conclu- he explained. “The conformal flight path vector the glideslope sions that the technology could deliver signifi- of the HUD is comparable to what pilots would and localizer, cant safety advantages. Analysis of company see head down, and we have real-time aircraft Bombardier CRJ landings with crews using HGS energy monitoring and improved assessment of and in speeds also demonstrated unprecedented, consistent deviations. … After six months of operating, the touchdown-zone accuracy compared with land- data are proving that with the Embraer 190, we during the ings by crews flying non-equipped CRJs. see much less deviation on the glideslope and “By its design and certification, autoland is localizer, and in speeds during the final portion final portion not capable of any advantage for required navi- of the approach. This means increased landing of the gation performance approaches or nonprecision accuracy. There is good reason to believe that we approaches, whereas a head-up display [HUD] also have a reduced risk of hard landings and tail approach.” can be used from taxi before takeoff until after strikes. The visual indication of our brake perfor- landing,” he said. mance after landing is something no other system Among HGS capabilities most relevant provides while looking out of the window.” to safety are the speed error tape, graphically The present and future role of HGS as an aid depicting the offset between the selected speed in unusual attitude recovery has been especially 34 | flight safety foundation | AeroSafetyWorld | August 2010 FlightTRAINING gratifying, Kemény said. “We have intuitive guid- Four HGS Proficiency Levels in Pilot Training ance during abnormal situations such as unusual attitude recoveries, engine failures and traffic-alert Level 1 Initial introduction and collision avoidance system resolution adviso- Tunneling, fixation and adaptation can be reduced using an FTD ries.” For further enhancement of aid to unusual attitude recovery, the platform can be configured Level 2 Secondary awareness to display g-loads in the combiner, he said. Prioritizing the information acquisition (FFS phase) At the beginning of 2011, Lufthansa CityLine will qualify all first officers to conduct Category Level 3 A world beyond the combiner III approaches. Steep approaches using HGS in Integration of HGS into conformal world and combining other the 190 already have been approved by European cockpit information authorities along with constant descents on all Level 4 The HUD as another flight deck tool and the HGS as the primary nonprecision approaches, he added. flight display reference JetBlue’s Maskiell said that as of April 2010, Final stage of proficiency; symbology becomes second nature 40 of the airline’s 190s have dual HGS. Acquiring Pilot becomes more aware of air mass effect and performance of aircraft the capability — and the requisite FSTDs — was on the agenda from the company’s founding. On HGS = head-up guidance system; FTD = flight training device; FFS = full flight simulator; HUD = head-up display the training side, all of the company’s FSTDs have Source: Christof Kemény, Lufthansa CityLine dual HGS. “More than a handful of pilots have shared with me that maybe the most challenging Table 2 event they’d had was conducting a flight without dual HGS when the system was [inoperative per only the standard operating procedures and the provisions of] the MEL for some reason. … They flight crew operating manual (FCOM) — and become very reliant on that device — not to the genuinely understanding systems. point of being unsafe [without it] but definitely to “Nowadays, understanding systems is of the point of knowing that there is a difference. … much more value than knowing information by In four years, there has not been a single [HGS- heart,” Blokzijl said. “I have pilots in my airline induced] safety event noted.” who can recite the Airbus FCOM backwards and forwards without a mistake … but they are Human Factors unable to transfer that knowledge into practi- LOC-I also has been linked with concerns about cal [application] in the aircraft. If we can make how best to instill safety attitudes and a positive them understand why things are happening and culture of professionalism from one generation the influence of certain failures in the system on of pilots to the next, said Cor Blokzijl, direc- the rest [of the system], a ‘light goes on’ and the tor flight operations, Mandala Airlines. Unease pilots are able to do what’s required.” about pilot professionalism (ASW, 6/10, p. 24) Continual transfer of expertise to less expe- has been increasing in some parts of the world rienced pilots ought to bridge the gap between — especially the perception that within today’s practicing narrowly focused tasks during recur- generation of pilots new to their airline careers, rent training in FSTDs and truly enhancing cog- some lack self-motivation or are too distracted nitive skills. Understanding of systems, system by other pursuits to study beyond minimum interfaces and dynamics of system failure — “if requirements or to read aviation safety media. this is failing, what else?” — have become a key “This affects their in-flight situation recogni- factor today in successful threat management, tion,” Blokzijl said. he added. Preparation to manage automation and LOC-I risks also requires a distinction between recitation To read an enhanced version of this story, go to of rote facts about airplane systems — knowing © Alex Beltyukov/Airliners.net Investigators can’t determine why five crewmembers and a number of passengers on an A319 became ill. he Irish Air Accident Investigation Unit inadequate air supply in the cabin or on the (AAIU) says its investigators were unable to flight deck. There also was no indication of pinpoint the cause of a May 2008 incident depressurization. in which five crewmembers and an un- The incident occurred around 1245 local Tdetermined number of passengers on an Airbus time on May 27, 2008, about 12 minutes after A319-132 became ill, with symptoms ranging from departure from Dublin for a flight to Cologne, drowsiness to a loss of sensation in their limbs. Germany.1 Six crewmembers and 119 passen- In its final report on the incident, the gers were aboard. AAIU ruled out one hypothetical cause The first indication of a problem occurred after another, saying there was no evidence as the airplane climbed through 10,000 ft, the of air contamination, poor air quality or report said. 36 | flight safety foundation | AeroSafetyWorld | August 2010 HUMANFactors “The purser called the cockpit on the inter- After the flight crew donned oxygen masks, com and reported that something was wrong the captain’s symptoms disappeared. — that almost all the passengers had fallen asleep Airport emergency personnel met the and that the [cabin crewmember] near her ap- airplane, the Irish national police were notified, peared unresponsive,” the report added. She later and “a decision was made to hold the aircraft at told investigators that it was unusual for so many a remote ramp position,” the report said. After passengers to be sleeping so early in the flight. about one hour, the A319 was towed to a termi- As the flight crew discussed the purser’s nal stand (gate), and at 1308, AAIU investigators comments, the captain said that he felt “some- arrived at the scene. what unwell.” The report said that later, he Police and airport authorities talked with “recalled a tingling sensation in his right arm, the crew and agreed to allow the passengers to comparable with the arm ‘falling asleep.’” leave the airplane. After disembarking, every- The flight crew at first suspected a pressur- one who had been in the airplane was offered ization problem, but the electronic centralized medical attention, but no one accepted the aircraft monitor (ECAM) showed no warnings offer, the report said. or failures and the cabin altitude indication was At the time of the incident, there was no medi- 1,700 ft. Nevertheless, the flight crew donned cal practitioner at the airport, and the only medical oxygen masks, declared an emergency and services available were from first-response medical began a descent, telling air traffic control (ATC) personnel, the report said. that they wanted to return to Dublin. They declared an emergency with Dublin ATC at Crew Interviews 1243, and the captain directed the cabin crew to Investigators interviewed crewmembers, includ- use portable oxygen cylinders. The airplane was ing the purser, who said that she had begun to landed in Dublin at 1257. feel “unwell” after takeoff. © Chris Photography Sorensen www.flightsafety.org | AeroSafetyWorld | August 2010 | 37 humanfactors The three other cabin crewmembers The next day, after the airplane the end of the flight, the pilot monitor- said that after takeoff, they felt tired. was moved to a maintenance facil- ing started to feel slightly unwell and One said that he might have been “un- ity, a series of tests were conducted went on oxygen briefly. His feeling able to perform his cabin service task,” to determine whether oil from the of being unwell disappeared. …. The another complained of dizziness, and engines or the auxiliary power unit AAIU inspector, who was not on oxy- the third — who had only three weeks (APU) had gotten into the cabin air gen, reported no ill effects.” of experience on the job — said she was supply. The tests involved running the Six days of tests at Toulouse revealed tired and “somewhat unwell,” the report engines, the APU and other equip- no anomalies and found only traces of said. The report did not discuss the first ment, and taking swabs from several toluene, ethylbenzene and styrene that officer’s condition. locations on the airplane, especially typically are found in aircraft exhaust, Investigators also talked with pas- near air outlets. No evidence was and traces of volatile organic com- sengers seated in different parts of the found to indicate any problem with pounds — all well below the exposure cabin and found that some passengers, oil contamination. limits. The report said the most toxic most of whom were described as “at “At one point during these tests, two substance found in the analysis was the older end of the age spectrum,” felt members of the inspecting team, which nicotine, also in very low concentrations. drowsy. Others said that they “had not In addition, a laboratory analysis of the noticed anything unusual or any feeling cabin air failed to find any trace of the of drowsiness or lack of well-being,” the oil used in the engines and the APU. The report said the most report said. “Many said their first indica- Authorities agreed to return the air- tion of anything unusual was when they toxic substance found craft to service. In the ensuing months, noticed that the aircraft was descending.” there was no recurrence of the problem, The cockpit voice recorder (CVR) in the analysis was the report said. recorded the entire event, as well as part of the previous flight, and showed nicotine, also in very ‘An Ongoing Issue’ that the flight deck environment dur- The continuing investigation deter- low concentrations. ing both flights was “very relaxed and mined that passenger baggage was the jovial before the purser expressed her only cargo on the aircraft and that, concerns” — so jovial, in fact, that “in particular, dry ice (a solid form of investigators initially were concerned numbered up to 15 people, reported a carbon dioxide) was not carried on the that the pilots might have been “af- strong smell in the cabin,” the report aircraft,” the report said. fected by a contaminated atmosphere.” said. “However, the other members of All six crewmembers had started After the flight crew declared the the team reported nothing unusual.” their duty day between 0315 and 0345, emergency, however, they were “fo- The tests were repeated with special and all had worked the previous day cused completely on the task in hand,” test equipment, and again, no problems until between 1300 and 1700. Fatigue- the report said. were found. monitoring software used by the “After three days of testing, it was operator showed that the flight crew Inconclusive Tests decided, in consultation with the op- had no “exposure to excessive fatigue” A series of tests failed to determine the erator and the aircraft manufacturer, during the month before the incident; cause of the event. that the aircraft be flown to the manu- two cabin crewmembers had “slightly An air composition check in the flight facturer’s facility at Toulouse [France] elevated fatigue exposure” for one day deck, cabin and baggage hold found no for further tests,” the report said. “The about two weeks earlier. unusual levels of methane, hydrogen flight to Toulouse was flown, unpres- The AAIU said that poor cabin air sulfide, carbon monoxide or oxygen. surized, at 10,000 ft, operated by quality “has been an ongoing issue in Examination of baggage in the two pilots from the operator and an commercial air transport operations. A baggage hold revealed nothing suspi- AAIU inspector in the jump seat. The number of recurring faults and defects cious, nor did a thorough examina- handling pilot remained on portable have [been] found to be the cause in tion of the cabin. oxygen for the entire flight. Toward many cases.” 38 | flight safety foundation | AeroSafetyWorld | August 2010 HUMANFactors Air quality typically has not been a “In the environment of a modern Recommendations problem in A319s or related airplanes, passenger aircraft, the list of potential The AAIU recommended that the Irish however, the report said. cabin air contaminants is large … Aviation Authority (IAA) review its In this case, investigators ruled [and] the task of detecting possible licensing requirements for the coun- out the “faults and defects” that have contaminants is daunting,” the report try’s major airports to ensure that they typically been cited in the past. There said. “Volatile organic compounds comply with International Civil Avia- was no evidence of a leak in the engine (VOCs) are [particularly] challeng- tion Organization (ICAO) guidelines, or APU oil seals, a gaseous discharge ing contaminants from the point of specifically with guidelines calling for from the air conditioning system or view of detection. These compounds major airports to maintain adequate from a fire extinguisher, or smoldering evaporate and are pumped overboard medical services. electrical wires, the report said. The by the aircraft’s air conditioning sys- The IAA Manual of Aerodrome galley ovens were not being used, and tem and so disappear, without leaving Licensing says that airports should be the aircraft did not undergo preflight a trace, in a relatively short period of equipped with medical supplies ap- maintenance or heavy cleaning. In ad- time. The investigation of this event propriate to the size of the airport, but dition, no problems were reported on demonstrated the difficulties of find- it does not require that they be staffed the previous flight. ing evidence of contamination after a with medical personnel. “During the exhaustive and pro- reported event, in spite of significant ICAO says, however, that a medi- longed tests undergone by this aircraft, resources available to, and utilized by, cal clinic should be established at any no re-occurrence of the problem was the investigation.” airport with at least 1,000 employees. found,” the report said. “Furthermore, The report praised the purser ICAO also calls for airports to be the aircraft has subsequently returned for being “positive and proactive” in staffed during their busiest hours with to service for an extended period, and promptly notifying the flight crew “at least one person trained to deal no re-occurrence of the problem has of the conditions she had observed with … basic measures for treatment been reported during this time. in the cabin, enabling their quick and protection of spills or leaks of “The investigation noted that, in response. radioactive materials, toxic or poison- the months that followed this event, On the ground, however, the delay ous substances.” The AAIU report said three further cabin air quality events, in deciding how to handle the situa- that this event “could have been dealt relating to fumes in the cabin, cabin tion prolonged the time spent by the with more effectively if such expertise crew reporting feeling unwell, un- passengers and crew in the airplane was available at the scene.” usual smells, etc., were reported to the — and prolonged their exposure to the The AAIU also recommended that AAIU. In each case, different operators potentially contaminated air supply, the the Dublin Airport Authority review and aircraft types built by different report said. the medical services that it provides manufacturers were involved. In none The report added that investiga- at the airport, as well as the “response of these cases was a definite source of tors could not explain what caused procedures to ensure that passengers the problem identified.” the problem or “why only some of the and crew are not unduly detained in a passengers complained of any symp- potentially toxic environment following Possible Contaminants toms and [why] the symptoms of the cabin air quality events.” The report said that, in most cases, affected passengers were limited to This article is based on AAIU accident report poor cabin air quality results from drowsiness (i.e., no passengers report- 2010-008, published July 15, 2010. contamination associated either with ed feeling unwell). The fact that those solid or liquid particles such as oil and who reported the symptoms recovered Note fuel, which produce gaseous byprod- rapidly after landing would indicate ucts with residues that can be detected, the absence of any toxic contaminant. 1. The accident report did not name the operator but identified the airplane by or with highly volatile substances that The failure to detect any abnormal using its registration number, which is that are dispersed through an airplane’s air residues within the aircraft after the of an A319 operated by Germanwings, a conditioning system and outflow valve event would also suggest the absence low-cost carrier based in Cologne/Bonn, and leave little, if any, residue. of toxic contaminants.” Germany. www.flightsafety.org | AeroSafetyWorld | August 2010 | 39 TRAFFIcControl Amid negotiations for the reunification of Cyprus, risks of pilot confusion persist in the Nicosia FIR. Uncivil Aviation By Wayne Rosenkrans embers of airline pilot global aviation organizations on how to Cypriots and Turkey call the Turkish unions worldwide have handle communications. The recom- Republic of Northern Cyprus (TRNC), received the third safety bul- mendations address the fundamental a political entity unrecognized by the letin since 2001 spotlighting issue of controllers from the Greek United Nations (U.N.) that was estab- theM protracted political dispute that has Cypriot ACC and the Turkish Cypriot lished in 1983 after the 1974 occupa- created uncoordinated and potentially ACC providing instructions to aircraft tion of about 36 percent of the island conflicting air traffic services (ATS) in crews although these ACCs do not com- by Turkish military forces. The U.N. part of the Nicosia Flight Information municate directly with each other. Peacekeeping Force in Cyprus controls Region (FIR).1 The airspace of concern The International Civil Aviation Or- a military buffer zone, called the Green extends northward to Turkey from the ganization (ICAO) recognizes only the Line, between the two parts of the island, Turkish Cypriot–governed northern Republic of Cyprus, the southern part of and the U.N. Good Offices Mission to part of Cyprus, an island nation in the the island governed by Greek Cypriots, Cyprus assists in negotiations now un- eastern Mediterranean Sea (Figure 1). and its Nicosia ACC as responsible for der way for reunification of the island. The bulletin from the International ATS in the Nicosia FIR. The Turkish A June safety briefing to IFALPA’s Federation of Air Line Pilots’ Associa- Cypriot community, however, since 1977 Air Traffic Services Committee by a tions (IFALPA) reminds flight crews also has asserted authority over its air representative of the Nicosia ACC about unique operational risks in this transportation system and ATS, which prompted IFALPA’s bulletin. The rep- part of the Nicosia FIR, especially the primarily has evolved into flights on resentative discussed 2006–2009 ATS potential for confusion about which ATS Route A-28 between Ercan Interna- events in the area of concern and their ATS area control center (ACC) has au- tional Airport and Turkey. significance (Table 1, p. 42), said Carlos thority. IFALPA urged pilots to be aware This airport is located east of Nico- Limon, president of IFALPA and cap- of consensus recommendations from sia in the part of the island that Turkish tain for Mexicana Airlines. The events 40 | flight safety foundation | AeroSafetyWorld | August 2010 TRAFFICControl were categorized as pilot deviations from Nico- with Ercan, which is not recognized by ICAO. sia ACC controller clearances, unauthorized The ATS communication issues — such as penetrations of Nicosia FIR airspace — meaning who designates levels, etc. — can become quite entry of an aircraft without traffic coordination complicated, particularly to flight crews that have at least 10 minutes in advance — and devia- not operated before in that airspace. Sometimes tions from the published in the past, IFALPA had received safety reports, air traffic management but to be fair, IFALPA has not directly received procedures, he said. reports concerning this particular airspace for a “The issues between couple of years.” the northern part of the Nicosia FIR and The intent of the bulletin was to highlight parts of the Ankara FIR have been known to us that the problem continues, and to remind for a long time,” Limon said. “IFALPA classifies flight crews of IFALPA’s recommendations, he that particular part of the airspace as ‘critically said. They are a brief subset of comprehensive deficient’ in particular because of the problems measures endorsed by the United Nations and ATC Deviation Issues in Nicosia FIR Istanbul FIR ATS Route Turkey A/UA-28 Ankara FIR Nicosia FIR Area of concern Damascus FIR Ercan Hellas UIR Cyprus Larnaca Paphos Nicosia FIR Beirut FIR Mediterranean Sea Cairo FIR Tel Aviv FIR ATS = air traffic service; FIR = flight information region; UIR = upper information region Note: The Ercan International Airport, Ercan Terminal Control Area and Ercan Advisory Area are not officially recognized by the Republic of Cyprus, which has responsibility for all Nicosia FIR airspace by decision of the International Civil Aviation Organization. Source: Department of Civil Aviation of the Republic of Cyprus and Flight Safety Foundation © Felix Gottwald/Airliners.net © Felix Figure 1 www.flightsafety.org | AeroSafetyWorld | August 2010 | 41 TRAFFIcControl published as Jeppesen Briefing Bulletin FRA Northern Cyprus as an occupied territory — 99-A, “Nicosia FIR/UIR IATA Communica- not as a sovereign state — so Ercan TMA and tions/Control Procedures (extract from IATA Ercan Advisory Airspace do not exist officially. Information Bulletin) .” Nevertheless, aircraft crews are being switched Finding an ATS solution acceptable to from Ankara ACC to Ercan ACC on a routine all concerned has been “incredibly difficult” basis. That is causing quite a bit of confusion, as despite technical initiatives and negotiations indicated by the reports we have seen. This is facilitated by ICAO and European organiza- one of the last of a handful of disputed pieces of tions, Limon added. “The problem appears to airspace left in the world, but Nicosia FIR sees be political, and it needs to be resolved because quite a bit of traffic out of Beirut and Damas- clearly there are flight safety issues. cus, and major Middle East air carriers have to transit this airspace.” The existence of contested airspace itself Nicosia ACC Reports of Failures to Coordinate creates problems incompatible with regional Flight Plans and Deviations From ATC Instructions harmonization of ATS risk reduction ef- Year forts. Eurocontrol’s Single European Sky ATM Type 2006 2007 2008 2009 Research (SESAR) project and the advent of Deviations from published 396 443 497 458 ATS safety management systems hopefully will ATM procedures encourage resolution, Voss said. “There’s just Deviations from ATC clearance 55 32 49 100 no way for alternative ATS and airspace, which Unauthorized penetration 373 429 450 390 of airspace (Nicosia FIR) circumvent what ICAO has allocated, to align with the SESAR goals,” he said. ACC = area control center; ATM = air traffic management; ATC = air traffic control; FIR = flight information region European organizations have tried unsuccess- Note: Data were furnished in June 2010 to IFALPA by the Department of Civil Aviation, fully so far to develop a technical workaround Ministry of Communications and Works, Republic of Cyprus. that would reduce risk on an interim basis given Source: International Federation of Air Line Pilots’ Associations (IFALPA) the practical realities. “With the type of telecom- Table 1 munications technology available, there are means available to mitigate the risk in Nicosia “We’re just trying to make sure that people FIR without upsetting the underlying political who have to operate in this airspace can operate issue — if the politicians are willing to yield and safely. We certainly would not want to appear to allow safety to become a priority,” Voss said. “For judge the issue, we just want it resolved.” example, a solution that would make a lot of sense would be to simply make sure that all the FSF Perspectives controllers have all the aircraft on their displays.” The consensus guidance from IFALPA and other A past obstacle to this workaround was lack aviation organizations advises flight crews to of an official ICAO airport identifier code for comply with ICAO expectations by monitor- Ercan International Airport, impeding the ex- ing but politely disregarding communication change of flight plan data via the networks used from Ercan ACC controllers, said William Voss, in air navigation planning throughout the world. president and CEO of Flight Safety Foundation. “To directly connect Ercan to another location, Flight crews that operate to and from Ercan In- ICAO would have had to assign this identifier, ternational Airport and Ercan Terminal Control which basically would have legitimized the exis- Area (TMA) do so outside of the ATS system tence of an unrecognized facility,” he added. recognized by ICAO. Other proposals would have set up sharing of “ICAO has said that Nicosia FIR is the flight plan data through third parties, Voss said. only legitimate FIR for Cypriot airspace,” Voss “This might not be possible directly between said. “The United Nations only recognizes Ercan and the rest of the world, but via Ankara 42 | flight safety foundation | AeroSafetyWorld | August 2010 TRAFFICControl ACC so that the exchange would stay and swift navigation of the aircraft land- and telling them to ignore Ercan Ad- within the Turkish government,” he ing to or taking off from the Turkish visory Area’s guidance, which is very said. “Data simultaneously would be Cypriot airports as well as using TRNC important for the safety of the flights,” retrievable by authorized third parties. airspace,” he said. The position taken he said. “Under these circumstances, A datalink between Ercan ACC and is that the legal basis for exercising air the refusal of the Nicosia ACC to Ankara ACC would enable sharing data traffic control of such flights on ATS cooperate with the Ercan ACC hinders with Eurocontrol — something Ankara Route A-28 is an agreement between the our efforts to increase flight safety in does today with its own data — then Civil Aviation Department of Northern this area and discourages some flights, Eurocontrol could give real-time access Cyprus and the Directorate General of which take a detour.” to Nicosia ACC and so on. Civil Aviation of Turkey, he said. This Examples of issues that Greek “By routing the communications agreement specifies that for safety pur- Cypriot and Turkish Cypriot aviation that way, there would be no need for poses, Northern Cyprus controllers have safety professionals might be able to ICAO or any other party to formally responsibility for air traffic within the address cooperatively include sharing recognize the disputed airport. That’s southern part of the Ankara [FIR] and of ATS safety event data and recordings still two sets of controllers working the on ATS Route A-28, he added. of related pilot-controller communica- same airspace, but at least they would “Implementation of our Systematic tion; lack of coordination when Nicosia see each other’s traffic in addition to Modernization of Air Traffic Manage- ACC routes aircraft off published monitoring the communications.” ment Project in 2008 enabled us to airways; and differences in how Nico- integrate [two new radar surveillance sia ACC controllers and Ercan ACC Northern Cyprus Perspective sensors in Northern Cyprus] with controllers handle aircraft departing Hasan Topaloglu, director of the Civil all the radar systems of Ankara ACC, to the south from Antalya, Turkey, in Aviation Department of the TRNC, told which increased security and safety, the Ankara FIR under their respective AeroSafety World that Turkish Cypriots and enabled the Ankara ACC to extend agreements with Ankara ACC, he said. decided to operate Ercan Interna- assistance where we deem necessary,” “These problems can only be solved tional Airport and the associated ATS Topaloglu said. “We have also increased with good will,” Topaloglu said. “As infrastructure to counteract a historic the number of our controllers, and Turkish Cypriots, we have underlined policy of isolating them. “Perhaps one we are constantly working in close on many occasions that this is a techni- of the most important restrictions [by cooperation with the Ankara ACC in cal matter, not a political matter, that the Greek Cypriot side] was one against order to ensure the safe conduct of all requires close cooperation of both sides freedom of travel, preventing direct flights in the region. We also monitor on the island. Since the negotiations flights to and from the Turkish Cypriot the Nicosia ACC controllers’ contacts for a comprehensive settlement are side with the [sole] exception of [flights to make sure safety rules are applied being conducted under the auspices to and from] the Republic of Turkey,” properly.” Total investment in Ercan of the United Nations, we believe that he said. “The Ercan ACC and Ercan In- ACC, control tower buildings and radar the technical expertise of ICAO, as an ternational Airport, established as a ne- sensor replacement was approximately expert agency of the United Nations, cessity due to the rejection of [requests $25.4 million, he said. could facilitate the establishment of to] the Greek Cypriots to provide “Approximately 600 aircraft per such [cooperation] on the island. … service to the north of the Green Line, day pass through the Ercan Advisory The Turkish Cypriot side is ready to have been in service for over 30 years Airspace,” he said. “In 2009, 1.8 million find mutually acceptable technical and and are technologically up-to-date and passengers used Ercan International operational arrangements, without effective to ensure flight safety.” Airport. Our prediction is that the traf- prejudice to the political and legal posi- Having made that fundamental fic will continue to increase, and these tions of the parties.” policy decision, domestic law of North- figures will be multiplied by two or ern Cyprus was amended to obligate the three in the next decade.” Note Civil Aviation Department to implement From the perspective of Ercan 1. IFALPA. “ATC Deviation Issues in the ICAO standards and recommended controllers, “Nicosia ACC causes Nicosia FIR.” Air Traffic Services Briefing practices to accomplish “safe, regular problems by contacting flight crews Leaflet no. 11ATSBL01, August 2010. www.flightsafety.org | AeroSafetyWorld | August 2010 | 43 tThreahreataAnnalysis © Jason J. Rainsford/AirTeamImages.com © Jason J. BY PAUL ESCHENFELDER AND RUSS DEFUSCO Bird Strike Mitigation beyond the airport Pilots must be prepared for bird strike avoidance and damage control. 44 | flight safety foundation | AeroSafetyWorld | August 2010 ThreatAnalysis etween November 2007 and January Turboprops are likewise at risk, but for differ- 2009, U.S. civil aviation experienced ent reasons. Propellers with composite material four major accidents caused by bird tend to shatter when struck. A de Havilland DHC- strikes. The accidents demonstrated the 8, on landing at Toronto City Airport, struck rangeB of aircraft categories and types affected geese just at touchdown. Both propellers lost large by this threat, and served as a reminder that chunks of the blades and vibrated so severely that the entire aviation community is challenged. A the crew had to shut down the engines on the run- Piper Seneca, a transport helicopter, a Cessna way. The airport management had been tolerating Citation business jet and an Airbus A320 were the geese on the field until this incident. all destroyed, and 17 people died.1–4 While general aviation airplanes typically do Three months prior to the US Airways not have the same engine ingestion concern as A320 bird strike accident, a similar accident oc- transport category jets, their overall design and curred at Rome Ciampino Airport. A Ryanair certification make them much less able to resist Boeing 737-800 encountered a large flock of damage from bird strikes. Mid-size to large birds starlings during its approach. The flight crew can penetrate the windshields and can cause pilot attempted a go-around, but birds were ingested incapacitation or disorientation, resulting in loss into both engines, and both lost thrust. The of control. The drag caused by the loss of the crew landed the aircraft on the runway, but the windshield has also resulted in accidents because left main landing gear collapsed. Although no enough thrust is not always available to overcome one was killed, there were 10 injuries and the the huge drag increase. Likewise, collision-caused airplane was damaged beyond repair. deformation of wing or tail surfaces can increase Before the Ryanair accident, an A320 oper- stall speed considerably and affect handling ated by Balkan Holidays encountered a flock qualities, especially at slower speeds. Damage to left-engine of gulls while departing the seaside resort of Other aspects of the problem have received fan blades of a Boeing Bourgas, Bulgaria. Both engines were dam- concentrated attention and reduced hazards on 767-432-ER following aged by bird ingestion and lost thrust. The crew airports. While not always properly implement- the ingestion of gulls had pre-briefed an immediate return plan and ed, well-developed and documented standards after takeoff at Rome successfully executed their plan. The airplane exist for airport habitat management, means for Fiumicino Airport, was landed safely, but a total of 32 fan blades on deterring wildlife from entering airfields, active July 2007. both engines had to be changed. dispersal of birds and other wildlife, and even lethal methods when population control must be employed. Such efforts must continue and be constantly monitored, but these strategies will not solve the problems of off-airport hazards, commu- nication failures, inadequate pilot training and procedures, or lack of operational guidelines by aircraft owners and regulators that led to the primary causes of the accidents cited. Agenzia Nazionale per la Sicurezza del Volo (ANSV)Volo del Nazionale per la Sicurezza Agenzia What is missing is a comprehensive, inte- grated plan that involves all parties: airports, aircraft operators, air traffic controllers, aircraft and engine manufacturers, regulators and others. What would an effective bird strike miti- gation policy look like? In the US Airways accident, the New York area airports were well known for the large bird populations affecting www.flightsafety.org | AeroSafetyWorld | August 2010 | 45 ThreatAnalysis them. La Guardia Airport has had a problem allowed thrust, the crew placed the engine at with resident Canada geese for some time. John risk of a high-energy collision, almost guaran- F. Kennedy International Airport is located teeing damage. across the fence line from a U.S. government A better technique based on current guide- wildlife refuge with a very large gull colony, lines for confronting large flocks of birds close protected by federal law. to the airport is to fly through the flock at low The U.S. Air Force Bird Avoidance Model engine rotation speed, allowing the engine to (BAM) had shown the risk of high bird concen- bypass the bird remains around the engine trations in the New York area during the A320 core without cascading damage to the com- accident period.5 The presence of large numbers pressor blades.6,7 of birds in the area should have been cause for But the crew had no training on the cur- action by aircraft operators, but was not. rent technique. Nor is training required by any No aviation hazard today is successfully regulator. Nor is any training available. mitigated without effective policy guidance for In another serious event in 2007 in Rome, the flight crews and adherence to that policy. a Delta Air Lines 767-400 was taxiing for In the Ryanair 737 accident, the crew departure. The crew observed a large number response was incorrect in our view. In many of gulls on the runway and in their departure low-altitude scenarios, the commonly used path. The crew discussed the situation but did response is to increase thrust and climb to not report the gulls, ask for bird dispersal prior avoid the hazard. But the problem with this to takeoff or delay takeoff waiting for the birds technique in connection with bird encounters to move. Instead, they took off into the birds is that it increases the kinetic energy of impact, and ingested gulls into both engines, the im- which equals one-half of the mass times veloc- pact causing serious vibrations and significant ity squared. In this case, velocity is determined loss of thrust in both engines. The aircraft was by engine rotation. By selecting maximum returned safely, but both engines were dam- aged beyond repair. Fast forward to February 2010 and another Bird Dispersal Goes Digital Delta flight conducting a departure from Tampa, Florida, U.S. Warned that large birds were in their departure path by the airport irport bird dispersal is becoming, if not an exact science, at least an organized and highly sophisticated one. traffic controller and by the crew of the Airbus A One example of a high-tech tool is the Ultima, a tablet that preceded them, the Delta crew took off, touch-screen personal computer offered by Scarecrow Bio-Acoustic and bird strikes damaged their aircraft. Delta Systems of Uckfield, East Sussex, England. Combined with an airfield Air Lines reportedly had no policy for its crews vehicle–mounted processor and loudspeakers, the system emits re- to mitigate this hazard. corded distress calls of as many as 20 species to drive birds away, while Hazard avoidance is superior to application logging all actions and GPS locations in real time. The system creates a database featuring date, time, location, system operator, species, flock of emergency procedures. Avoidance can take size and dispersal direction, all of which can be used for data analysis a number of forms, many of them simple and and to store records for program documentation and auditing. cost-free. If birds are in the takeoff path, the The Ultima includes a report-generation function that allows sort- pilot should notify the airport operator and ing by combined factors such as dates, species, location and operator delay departure until the birds move or are name. Printouts are available in spreadsheet or graphical formats. scared away. Another alternative is to depart Ultima has been installed at airports in Pittsburgh; London Luton; Belfast, Northern Ireland; Cancun, Mexico; and Christchurch, New via another runway that is free of hazard. Zealand. The company reports that it has sold more than 70 units since Likewise, for landing, flight crews should use the product’s introduction in 2008. a different runway if birds are reported on the — Rick Darby landing runway. Or go around and wait for the birds to leave. 46 | flight safety foundation | AeroSafetyWorld | August 2010 ThreatAnalysis Another important area where Bundesstelle für Flugunfalluntersuchung (BFU) study and action are needed is the lack of adequate aircraft design specifica- tions. This problem is complex, because many interrelated systems are involved: aircraft design and operation, engine design and operation, airport mitiga- tion, bird population control, airport habitat, training, warning systems, policy, etc. It is complicated, because there is no one answer but, as with all aviation hazards, an interdisciplinary approach is required. An Air Berlin Boeing 737-700 windshield after encountering a flock of white-fronted The majority of bird strikes occur geese at about 2,150 ft and 226 kt indicated airspeed. below 3,000 ft. If departing from an airport in a high-bird-threat environ- airframes and windows. While modern 2. A Cessna Citation was climbing through ment, jets should use International heated windows should resist a gull or 3,000 ft after departing from Wiley Post Airport, Oklahoma City, Oklahoma, U.S., Civil Aviation Organization Noise duck, larger birds may penetrate them or in March 2008. It struck a flock of migrat- 8 Abatement Procedure 1. This rapid shower the pilots with glass as the inner ing white pelicans, causing right-engine climb to above 3,000 ft above ground pane of the window spalls or shatters. failure and wing damage. Loss of control level would, in all likelihood, have Likewise, the small bird that bounces off followed, with the ensuing crash killing all prevented the US Airways accident. like a tennis ball when struck at slower five occupants. General aviation aircraft should depart speed suddenly becomes a bowling ball 3. A Sikorsky S-76 helicopter, flying at low al- at best angle-of-climb speed. Those when struck at high speed. Below 10,000 titude in January 2009, encountered a large techniques enable the aircraft to clear ft, limit aircraft speed to 250 kt indi- bird that penetrated the front canopy. Ei- the hazard zone below 3,000 ft faster cated airspeed or less. ther the crew or the controls were disabled by the collision, and the helicopter crashed, and climb at a lower speed, which can Aviation operations successfully killing eight of the nine occupants. lessen the severity of impact. When mitigate a variety of hazards every day. 4. In January 2009, a US Airways Airbus landing in an area of high bird activity, The industry has built strong defenses A320 ingested Canada geese in both the aircraft should remain at 3,000 ft against them. We can do the same with engines, necessitating a ditching on the or above if possible until necessary to the birds. Hudson River. No occupants were killed, three sustained serious injuries and the descend for landing. Capt. Paul Eschenfelder is the lead instructor for aircraft was destroyed. If birds are encountered en route, Embry-Riddle Aeronautical University’s Airport on climb or descent, the flight crew Wildlife Training Seminar, the only such course 5. The BAM is an interactive risk calculation should pull up — consistent with good approved by the FAA for full compliance with tool, accessible on the Internet at BY RICK DARBY Fewer Runway Excursions in 2009 Fatal accident rate for 2000–2009 also shows improvement. he 2009 picture of fatal accidents released data from Boeing Commercial One of the 2009 excursion accidents in worldwide commercial avia- Airplanes.1 involved fatalities, compared with three tion showed no overall improve- Twelve of the 62 total accidents, or in 2008. Some of the latest excursions ment from 2008, but in one 19 percent, were overruns or veer-offs, involved equipment failures rather than Timportant category, there was good both classified as runway excursions faulty takeoffs or approaches. One excur- news. Runway excursion accidents — (Table 1).2 Of the 53 accidents in sion resulted from the inability to fully targeted by the Flight Safety Founda- 2008, 16 — 30 percent — were runway extend the left main landing gear; anoth- tion Runway Safety Initiative and the excursions. In 2007, the Boeing data er, right main landing gear failure and Runway Excursion Risk Reduction Tool- included 10 excursions, 26 percent of collapse; a third, engine thrust-reverser kit — were fewer, according to newly the 38 total accidents. failure and uncoordinated thrust. 2009 Airplane Accidents, Worldwide Jet Fleet Onboard Fatalities Phase Damage (External Major Date Airline Model Accident Location of Flight Description Category Fatalities) Accident? Jan. 6 China Southern Airlines 777-200 Pacific Ocean Cruise Flight attendant broken ankle Jan. 15 US Airways A320 New Jersey, U.S. Climb Multiple bird strikes, ditching Destroyed Jan. 17 Iran Air F-100 Yazd, Iran Parked Fuselage struck by ambulift Substantial Jan. 19 Iran Air F-100 Tehran, Iran Landing Veered off runway Substantial Feb. 9 Air Méditerranée A321 Paris Landing Overshot runway turnoff Substantial Feb. 13 BA CityFlyer RJ-100 London Landing Nose landing gear collapse Substantial Feb. 16 Air Algérie 737-400 In Aménas, Algeria Landing Overran runway Substantial Feb. 19 Atlasjet Airlines A320 Istanbul, Turkey Tow Towbar failure Feb. 23 Royal Air Maroc 737-800 Medina, Saudi Arabia Takeoff Tail strike Substantial Feb. 23 Lion Air MD-90 Batam, Indonesia Landing Gear-up landing Substantial Feb. 25 Turkish Airlines 737-800 Amsterdam, Netherlands Landing Crash during approach Destroyed 9 (0) March 2 CityJet RJ-85 Dublin, Ireland Tow Aircraft struck tug Substantial March 9 Lion Air MD-90 Jakarta, Indonesia Landing Veer-off Destroyed March 20 Emirates A340 Melbourne, Australia Takeoff Tail strike Substantial March 23 FedEx MD-11 Tokyo Landing Hard landing Destroyed 2 (0) April 4 Air China A321 Beijing Landing Tail strike Substantial April 9 Aviastar Mandiri BAe 146 Wamena, Indonesia Approach Struck hill Destroyed 6 (0) April 12 Wizz Air A320 Timisoara, Romania Landing Hard landing Substantial April 16 Jade Cargo International 747-400 Incheon, South Korea Landing Veer-off Substantial April 20 Royal Air Maroc 767-300 New York. Landing Hard landing Substantial April 27 Magnicharters 737-200 Guadalajara, Mexico Landing Gear-up landing Substantial Table 1 (continued next page) 48 | flight safety foundation | AeroSafetyWorld | August 2010 DataLink 2009 Airplane Accidents, Worldwide Jet Fleet Onboard Fatalities Phase Damage (External Major Date Airline Model Accident Location of Flight Description Category Fatalities) Accident? April 29 Bako Air 737-200 Massamba, DR Congo Cruise Crashed en route Destroyed 7 (0) May 4 Northwest Airlines A320 Denver Landing Tail strike Substantial May 6 World Airways DC-10 Baltimore Landing Hard landing Substantial May 7 NAS Air A320 Alexandria, Egypt Landing Hard landing Substantial May 8 Saudi Arabian Airlines MD-90 Riyadh, Saudi Arabia Taxi Veer-off Substantial May 8 Asiana Airlines 747-400 Frankfurt, Germany Approach Flap departed, punctured Substantial fuselage May 19 American Airlines 777-200 Miami Parked Employee fall (1) June 1 Air France A330 Atlantic Ocean Cruise Crashed into Atlantic Ocean Destroyed 228 (0) June 3 China Cargo MD-11 Urumqi, China Landing Hard landing Substantial June 3 Aeroflot-Nord 737-500 Moscow Cruise Heavy hail encounter Substantial 6 June Myanma Airways F-28 Akyab, Myanmar Landing Departed runway Destroyed June 8 United Airlines 777-200 Pacific Ocean Cruise Flight attendant broke ankle June 9 Saudi Arabian Airlines MD-11 Khartoum, Sudan Landing Hard landing Substantial June 27 US Airways 737-400 Tampa, Florida, U.S. Landing Hard landing Substantial June 30 Yemenia A310 Indian Ocean Approach Crashed into Indian Ocean Destroyed 152 (0) July 7 Rossiya Russian Airlines A320 St. Petersburg, Russia Landing Tail strike Substantial July 17 Transaero Airlines 737-400 Moscow Landing Tail strike Substantial July 21 Aeromexico 737-700 San Francisco Tow Landing gear collapse Substantial Aug. 3 Saha Air 707-300 Ahwaz, Iran Initial Climb Uncontained engine failure Substantial Aug. 4 Sata Internacional A320 Ponta Delgada, Portugal Landing Hard landing Substantial Aug. 10 All Nippon Airways 737-800 Tokyo Landing Tail strike Substantial Sept. 4 Air India 747-400 Mumbai Taxi Fuel leak, fire Substantial Sept. 13 Lufthansa Cargo MD-11 Mexico City Landing Hard landing Substantial Sept. 14 Contact Air Flugdienst F-100 Stuttgart, Germany Landing Gear-up landing Substantial 1 Oct. Wind Jet A319 Catania, Italy Cruise Turbulence, hail Substantial Oct. 2 Malaysia Airlines 737-400 Kuching, Malaysia Tow Landing gear collapse Substantial Oct. 6 Boliviana de Aviación 737-300 Cochabamba, Bolivia Cruise Hail encounter Substantial Oct. 20 Centurion Air Cargo MD-11 Montevideo, Uruguay Landing Hard landing Substantial Oct. 21 Sudan Airways 707-300 Sharjah, United Arab Initial Climb Struck terrain Destroyed 6 (0) Emirates Oct. 30 Pegasus Airlines 737-800 Malatya, Turkey Taxi Wing stuck light pole Substantial Nov. 2 Delta Air Lines MD-90 Phoenix Climb Bird strike Substantial Nov. 18 Iran Air F-100 Isfahan, Iran Landing Landing gear collapse Substantial Nov. 19 Compagnie Africaine MD-82 Goma, DR Congo Landing Overrun Destroyed d’Aviation Nov. 28 Avient Aviation MD-11 Shanghai Takeoff Overrun Destroyed 3 (0) Dec. 1 TAF Linhas Aéreas 727-200 São Paulo Taxi Struck maintenance stand Substantial Dec. 2 Merpati Nusantara F-100 Kupang, Indonesia Landing Veer-off Substantial Airlines Dec. 17 TAF Linhas Aéreas 727-200 Manaus, Brazil Approach Wind shear Substantial Dec. 21 Merpati Nusantara 737-300 Makassar, Indonesia Landing Hard landing Substantial Airlines Dec. 21 Canadian North 737-200 Calgary, Canada Parked De-icer fall (1) Dec. 22 American Airlines 737-800 Kingston, Jamaica Landing Overrun Destroyed Dec. 29 Wizz Air A320 Boryspil, Ukraine Landing Veer-off Substantial Total accidents (62) 413 (2) 13 Source: Boeing Commercial Airplanes Table 1 www.flightsafety.org | AeroSafetyWorld | August 2010 | 49 DataLink Four of the 2009 excursions, one- Accidents, Worldwide Commercial Jet Fleet, by Type of Operation third, were classified as major acci- dents, a category that partially overlaps On-board Fatalities All Accidents Fatal Accidents (External Fatalities)* with the fatal accident category.3 Six of Type of operation 1959–2009 2000–2009 1959–2009 2000–2009 1959–2009 2000–2009 the excursions in 2008 — 38 percent — Passenger 1,344 301 475 72 27,833 (778) 4,942 (171) were major accidents. Scheduled 1,235 280 430 69 23,719 4,938 The 37 approach and landing ac- Charter 109 21 45 3 4,114 4 cidents (ALAs) accounted for 60 percent Cargo 224 81 73 14 255 (329) 42 (73) of the 2009 total, compared with 31 — 58 Maintenance test, ferry, 116 11 44 3 208 (66) 17 (0) percent — in 2008 and 23 — 61 percent positioning, training and demonstration — in 2007. Although ALAs as a percent- Totals 1,704 393 592 89 28,296 (1,173) 5,001 (244) age of the total number of accidents have U.S. and Canadian 530 77 176 14 6,153 (381) 355 (15) not changed by more than three percent- operators age points in the past three years, their Rest of the world 1,174 316 416 75 22,143 (792) 4,646 (229) consequences were less severe in 2009. Totals 1,704 393 592 89 28,296 (1,173) 5,001 (244) Four of the ALAs in 2009, or 11 percent, *External fatalities include ground fatalities and fatalities on other aircraft involved, such as helicopters involved fatalities. The corresponding or small general aviation airplanes, that are excluded. percentages for 2008 and 2007 were 19 Source: Boeing Commercial Airplanes percent and 22 percent, respectively. Table 2 Most of Boeing’s data in its annual accident summaries concern the most 69 fatal accidents in scheduled service for 1999–2008 were 97 percent and 86 recent year plus the previous nine years, compared with 74 in the previous percent, respectively. thus offering a chance to compare suc- period, a 7 percent improvement. The Boeing tabulated fatalities accord- cessive 10-year periods. For example, the number of fatal charter accidents held ing to the U.S. Commercial Aviation 2008 report included 1999 through 2008 steady at three. Safety Team/International Civil Aviation numbers; the 2009 report comprises There were 4,942 on-board fatalities Organization (CAST/ICAO) standard 2000 through 2009 (Table 2). from 2000 to 2009, compared with 4,670 taxonomy (Figure 2).5 “Loss of control In 2000–2009, the fatal accident from 1999 to 2008, a 6 percent differ- in flight” (LOC-I) and “controlled flight rate for scheduled commercial pas- ence. That included an increase in fatali- into terrain” (CFIT) continued to be senger operations was 0.42 per million ties during scheduled operations from involved in the greatest number of fatali- departures, compared with 0.45 in 4,666 to 4,938. There were four fatalities ties in the most recent 10 years. How- 1999–2008 and 0.50 in 1998–2007. in charter operations in both periods. ever, loss of control on-board fatalities The 10-year period beginning in 2000 Among all the accidents in 2000– were 1.8 to 2.0 times higher than CFIT included 301 accidents in passenger op- 2009, 23 percent involved fatalities fatalities in the most recent 10-year erations, a 6 percent increase over the 283 (Figure 1). The corresponding ratio periods. In 2000–2009, there were 1,759 in 1999–2008 and a 5 percent increase for 1999–2008 and 1998–2007 was 25 LOC-I on-board fatalities and 961 CFIT above the 286 in the 1998–2007 stretch. percent. The fatal-accident proportion fatalities. For 1999–2008, the numbers The increases in the latest 10 years from 1959–2009, comprising most of were 1,926 and 961, respectively; for included both scheduled operations and the years of passenger-jet service, was 1998–2007, 1,984 and 1,137 respectively. charter flights, 6 percent and 17 percent 35 percent. The combined category “runway respectively. Accidents in cargo opera- Among the 304 nonfatal accidents excursion — landing” plus “abnormal tions increased in the most recent period in the latest 10-year period, 292 — 96 runway contact” plus “undershoot/over- from 79 to 81, or by 3 percent. percent — involved either hull loss shoot” resulted in 606 on-board fatalities Fatal accidents in 2000–2009 num- or substantial damage.4 A smaller in the 2000–2009 period, an increase of bered 72, 5 percent fewer than the 76 proportion — 85 percent — of fatal 49 percent over the 408 in 1999–2008. in the previous period and 8 percent accidents involved hull loss or substan- In the “system component failure — fewer than in the 78 in 1998–2007. The tial damage. The comparable numbers non‑powerplant” category, the latest 50 | flight safety foundation | AeroSafetyWorld | August 2010 DataLink 10-year period had 314 on-board fatali- Accidents, by Injury and Damage, Worldwide Commercial Jet Fleet ties, or 26 percent fewer than the 426 in 1959 through 2009 the previous 10-year period. There was Total 1,704 592 fatal accidents 1,112 non-fatal accidents (35% of total) (65% of total) a surge of “unknown or undetermined” 479 accidents with hull loss 417 hull loss accidents accident on-board fatalities, from 120 in 25 accidents with 1999–2008 to 504 in 2000–2009. substantial damage 643 substantial damage 88 accidents without 52 accidents without substantial substantial damage damage (but with serious injuries) Notes 0 100 200 300 400 500 600 700 800 900 1,000 1,100 1,200 1,300 1,400 1,500 1,600 1,700 1,800 1. Boeing Commercial Airplanes. Statistical Number of accidents Summary of Airplane Accidents: Worldwide 1999 through 2009 Total 393 Operations, 1959–2009. Available via 89 fatal accidents 304 non-fatal accidents (23% of Total) (77% of Total) the Internet at Fatalities 800 606|21 test flights or resulting from hostile action 600 504|0 400 314|0 such as sabotage or hijacking are excluded. 156|69 193|9 154|38 200 122|3 110|4 96|1 23|0 1|8 2|2 3. Boeing defines a major accident as one in 0 LOC-I CFIT RE- UNK SCF-NP MAC RI-VAP RE- OTHR F-NI WSTRW FUEL RAMP SCF-PP which any of three conditions is met: the Landing + Takeo airplane was destroyed; there were multiple ARC + USOS Number of fatal accidents fatalities; or there was one fatality and the (89 total) 20 16 16 4 3 2 3 5 4 2 1 1 9 3 airplane was substantially damaged. Flight Safety Foundation supports the use of this CAST = U.S. Commercial Aviation Safety Team; ICAO = International Civil Aviation Organization; ARC = abnormal runway contact; CFIT = controlled flight into terrain; F-NI = fire/smoke (non-impact); FUEL = term to designate the most severe accident fuel related; LOC-I = loss of control – in flight; MAC = midair/near midair collision; OTHR = other; RAMP = category in preference to the traditional ground handling; RE = runway excursion; RI-VAP = runway incursion – vehicle, aircraft or person; term hull loss, which the Foundation believes SCF-NP = system/component failure or malfunction (non-powerplant); SCF-PP = system/component failure or malfunction (powerplant); UNK = unknown or undetermined; USOS = undershoot/overshoot; is more significant for insurance actuarial WSTRW = wind shear or thunderstorm. purposes than as a measure of risk. No accidents were noted in the following principal categories: aerodrome, abrupt maneuver, air traffic 4. Substantial damage is “damage or failure management/communications, navigation, surveillance, cabin safety events, evacuation, fire/smoke (post-impact), ground collision, icing, low altitude operations, runway incursion – animal, security which adversely affects the structural related or turbulence encounter. strength, performance or flight charac- Note: Principal categories are as assigned by CAST. Airplanes manufactured in the Russian Federation or teristics of the airplane, and which would the Soviet Union are excluded because of lack of operational data. Commercial airplanes used in military normally require major repair or replace- service are also excluded. ment of the affected component.” Source: Boeing Commercial Airplanes 5. The taxonomy is described at Screen Test Some light-airplane pilots are not fully trained for electronic displays. REPORTS Administration] aircraft registry and an analysis of GAATAA [General Aviation and Air Taxi No Measurable Improvement Activity and Avionics] Survey data for the years Introduction of Glass Cockpit Avionics Into Light Aircraft 2006 and 2007.” Accident data came from the U.S. National Transportation Safety Board (NTSB). Safety Study NTSB/ NTSB Aviation Accident Database. SS-01/10; PB2010-917001. March 9, 2010. 87 pp. Figures, tables, Of the 8,364 airplanes included in the study, appendix. 2,848 had conventional cockpit displays and hanks to trickle-down technology, new 5,516 had glass cockpits. Variables selected for light aircraft have undergone a transition analysis included accident severity, weather, Tfrom analog flight instruments to computer time of day and the purpose of the flight. Pilot screens similar to those on modern transport information, such as age, highest certificate category aircraft. The “glass cockpit,” in which level, possession of an instrument rating and the electronic displays integrate aircraft control, flight hours, also was analyzed. autopilot, communication, navigation and sys- The researchers identified 266 accidents tems monitoring, represents a significant change between 2002 and 2008, 62 of them fatal. The for general aviation. report says, “The cohorts [groups of subjects with But does the glass cockpit make flying light a defining characteristic] selected had similar air- aircraft — defined here as having a maximum frames, numbers of engines and engine types but gross weight of 12,500 lb/5,700 kg — safer? differed principally in their type of primary flight So far, no. “The introduction of glass cock- instrumentation,” glass versus conventional. pits has not resulted in a measurable improve- “The percentage of accidents resulting in ment in safety when compared to similar aircraft fatality was about twice as high for the glass with conventional instruments,” the study says. cockpit cohort as for the conventional cohort,” The report says that the accident data analy- the report says. sis of conventional versus glass light airplanes For another viewing angle, researchers included “(1) a comparison of specified aircraft looked at accident rates based on flight hours. models manufactured during the five years “Those results indicate that the total accident from 2002 through 2006, the years that spanned rate per 100,000 flight hours was higher for the the transition of the fleet from conventional to glass cockpit cohort in 2006 but higher for the glass cockpit displays, (2) statistical compari- conventional cohort in 2007,” the report says, sons of retrospective accident data for the years noting that the cohort rates for the combined 2002 through 2008 by display type, and (3) a years were roughly the same. The rates for fatal comparison of aircraft and flight activity data accidents in both years were higher for glass obtained from the FAA [U.S. Federal Aviation cockpit aircraft, although the report cautions 52 | flight safety foundation | AEROSafetyWorld | August 2010 InfoScan that the rates were based on small numbers and digitally based displays, which might have been therefore subject to a large standard error. expected to be helpful in those conditions. Accident rates per 100,000 flight hours for “The NTSB reviewed FAA and manufacturer 2006 and 2007 combined were greater for con- training materials and programs applicable to ventional aircraft than for glass cockpit aircraft glass cockpit aircraft and visited aircraft manu- during the daytime, with the relative positions facturers to observe factory training available reversed at nighttime. Fatal accidents occurred to general aviation pilots transitioning to glass at a higher rate for glass cockpit aircraft during cockpit avionics,” the report says. both daytime and nighttime. Accident rates in Prospective pilots must pass a knowledge visual meteorological conditions were similar test and obtain an instructor’s endorsement to be for both groups but higher for glass cockpit eligible to take the practical test for a certificate aircraft in instrument meteorological conditions or rating. Questions on the test do not assess the (IMC). The glass cockpit cohort had higher candidate’s knowledge of electronic flight instru- fatal accident rates in both meteorological ments. However, the report says, “there are gener- conditions. al requirements for all pilots to be knowledgeable “Both cohorts experienced equally low fatal about the operation and limitations of the aircraft accident rates for instructional flights, but the they fly — including all aircraft systems — and glass cockpit cohort experienced a higher fatal to be proficient in the use of those systems.” But accident rate during personal/business flights,” aside from the general requirements, there is no Fatal accidents the report says. FAA mandate for equipment-specific training. In 255 accidents for which sufficient infor- “With the exception of training provided by occurred at a higher mation was available from NTSB records, the airframe manufacturers with the purchase of a only statistically significant difference in acci- new aircraft, pilots must currently seek out and rate for glass cockpit dent event categories was a higher percentage of obtain equipment-specific glass cockpit training aircraft during collision with terrain for the glass cockpit cohort on their own,” the report says. — 16 percent versus 8 percent. Some of the larger avionics manufacturers both daytime As a group, “accident pilots of glass cockpit– provide software for personal computers that equipped aircraft were older, held higher levels allows pilots to interact with the display. “These and nighttime. of pilot certification, were more likely to hold software simulators are not intended to replicate an instrument rating and had more flight hours the functionality of an approved flight simulator than those flying aircraft with conventional or training device, but rather to serve as interac- instruments,” the report says. tive procedural trainers that allow pilots to prac- Despite those pilots’ relative maturity, could tice using glass cockpit avionics and experience inexperience with glass cockpits have been a various display system malfunctions and failures factor? The evidence was ambiguous. Distribu- that may not be easily or safely replicated in the tions of flight time in type were not significantly aircraft,” the report says. different between pilot cohorts, the report says. Insurance companies often require training re- It adds, however, that “data concerning flight quirements for pilots transitioning to glass cockpits experience in aircraft make and model made as a condition for coverage. But, the report says, no distinction in cockpit design, so some pilots those requirements are tailored to individual pilots may have been experienced in the aircraft type and vary among insurance companies. while having little experience with the particular “The lack of equipment-specific training cockpit display in the aircraft.” requirements from the FAA and the variability In sum, glass cockpit aircraft had lower total of insurance company requirements result in a accident rates. But accident and fatal accident wide range of initial and recurrent training ex- rates were higher for the glass cockpit cohort periences among pilots of glass cockpit aircraft,” in IMC and at night, despite the capabilities of the report says. www.flightsafety.org | AEROSafetyWorld | August 2010 | 53 InfoScan Researchers considered several accident case training materials and aeronautical knowledge studies that raised issues about the functional requirements for all pilots.” differences between electronic and conventional — Rick Darby displays. “The wide variety of complex glass cockpit BOOKS equipment designs, and their proprietary tech- Reasons Unknown? nology, demands that any discussion of these displays be system-specific,” the report says. The Crash of TWA Flight 260 “Consequently, as electronic systems replace Williams, Charles M. Albuquerque, New Mexico, U.S.: University of New Mexico Press, 2010. 268 pp. Figures, end notes. analog gauges, the expectation that average general aviation pilots will understand the in- f TWA 260 sounds unfamiliar, it is not a sign ner workings of their cockpit instruments is no that you are losing your memory. You may longer realistic. This problem is compounded by Inot have been born yet when the accident oc- the fact that, unlike analog gauges, the function- curred, on Feb. 19, 1955. ality and capability of electronic display systems The Martin 404 was engaged in a short can continue to evolve after they are installed passenger flight from Albuquerque to Santa Fe, because of subsequent software revisions.” about 60 mi (97 km) to the northeast. To avoid Glass cockpit displays may fail differently the Sandia Range that looms over northern and than conventional displays. The report describes western Albuquerque — and was hidden by one accident in which a blocked pitot tube, storm clouds at the time — the flight was routed which would have affected only the airspeed indirectly. Something went wrong. indicator of a conventional cockpit display, As Williams tells it: “At 7:12 [a.m. local time, resulted in the loss of airspeed plus altitude and Capt. Ivan] Spong was in the act of changing rate-of-climb information in a glass cockpit radio frequencies when the terrain-warning bell display. “The information provided to the pilot suddenly sounded its alarm. Instinctively, both indicated only that the air data computer had pilots looked out the window. Nothing but gray failed, with no indication of why it had failed or cloud, but then, flashing through a weak spot whether the situation could be safely corrected in the cloud just beyond the right wing tip, they in flight,” the report says. “The NTSB concludes saw the sheer cliff side of Sandia Crest — an ap- that pilots are not always provided all of the palling shock, for they should have been 10 mi information necessary to adequately understand [16 km] from the mountain. the unique operational and functional details of “Reacting instantly, they rolled the airplane the primary flight displays in their airplanes.” steeply to the left and pulled its nose up. The The NTSB made six recommendations to heading indicator spun rapidly. When it was the FAA, including these: indicating a westerly heading, they started to “Revise airman knowledge tests to include level the wings. This was their final act. Hidden questions regarding electronic flight and naviga- by the dense cloud, another cliff side lay directly tion displays … ; ahead. When they struck it, they were still in a “Require all manufacturers of certified left bank, nose high. The airplane exploded.” electronic primary flight displays to include Bad weather hindered search parties trying information in their approved aircraft flight to reach the wreckage, of which little was visible manual and pilot’s operating handbook supple- except the airplane’s tail, but that did not matter ments regarding abnormal equipment operation much. According to a report by a helicopter or malfunction due to subsystem and input pilot, it was definitely the accident aircraft and malfunctions … ; [and,] “there was no possibility any survived.” “Incorporate training elements regarding The author, a member of a mountaineer- electronic primary flight displays into your ing club who had been hiking in the Sandia 54 | flight safety foundation | AEROSafetyWorld | August 2010 InfoScan Range the day before, volunteered to join one known the pilot, Ivan Spong, knew that he was of the search parties. He was among those who not a happy-go-lucky fly guy recovering from a reached the nearly inaccessible accident site. rough night on the town,” Williams says. “His Why return after 55 years to this story, peers regarded him as a serious and highly involving an airliner type that has long since competent professional who adamantly refused been retired from service and flight technology to deviate in the slightest degree from flying that is primitive by today’s standards? Williams regulations.” believes that the accident investigation by the Another TWA Martin 404 pilot, Larry De- U.S. Civil Aeronautics Board (CAB), a prede- Celles, “had recently succeeded in discovering cessor of the National Transportation Safety both the nature of and the correction for some The controversy over Board, reached conclusions that were seriously fluxgate compass errors aboard TWA aircraft,” the probable cause, mistaken. Additionally, the controversy over Williams says. the probable cause, which continued for years Williams explains that the fluxgate compass which continued and resulted in the Air Line Pilots Association “sensed the direction of the horizontal compo- (ALPA) dissenting from the CAB findings, of- nent of the Earth’s magnetic field and generated a for years, offers fers the drama of a detective story. small electrical voltage that was compatible with The CAB accident report released in Octo- the commonly available electric meters already the drama of a ber 1955 said, “The weather was such that the in use on the instrument panel. To maintain detective story. visibility all along the airway was good for many accuracy during the twists and turns of normal miles ahead to the north. … Even if all naviga- flight, the compass was kept as level as possible by tional aids and instruments had failed, all the mounting it on a gyroscope-and-gimbal system. captain had to do was look outside to determine “Ironically, the Achilles heel of the system that he was not following the airway. was the very stabilizing gyroscope that was the “Therefore, from all available evidence, and key to its reliability. Steep turns could induce the lack of any evidence to the contrary, the torques that surreptitiously caused … erroneous Board can conclude only that the direct course readouts until it eventually realigned itself — a taken by the flight was intentional.” process usually requiring several minutes of Williams says, “ALPA pointed out that straight and level flight.” another airline pilot had taken off minutes after The CAB amended its report in 1957, delet- Spong and had testified that the mountains and ing the word “intentional.” Nevertheless, says the Rio Grande valley were completely obscured Williams, “a lengthy paragraph had been in- by a snowstorm, but it fell on deaf ears. The serted that said the same thing in a roundabout weather was good — the report said so! way. … ALPA’s arguments for an instrument “When ALPA asked CAB whether they re- malfunction were dismissed — almost con- ally believed — as their report had seemed to temptuously. They did not, said the Board, even imply — that the TWA pilots had entered into ‘warrant serious considerations [sic].’ They were some sort of a suicide pact, they replied, ‘No. not a possible contributing factor.” No. We meant no such implication. We believe ALPA sent the CAB a critique of its amended they were taking a shortcut.” report in 1958, with a copy to TWA pilots. The The Board’s determination of probable cause report and ALPA’s responses are reprinted in the was “a lack of conformity with prescribed en book. Most of the CAB text and ALPA replies are route procedures and the deviation from airways too long to quote, but here is one brief sample: at an altitude too low to clear obstructions ahead.” The CAB report said, “It is difficult to un- TWA pilots and those from other airlines derstand why the flight took the heading it did were overwhelmingly skeptical that the captain from the airport to Sandia Mountain.” had knowingly changed course from the ap- ALPA responded, “It is not difficult to proved flight plan. “Those who had personally understand this. The flight experienced a www.flightsafety.org | AEROSafetyWorld | August 2010 | 55 InfoScan malfunction of the fluxgate compass system management, safety regulations and relevant which was providing heading data to each pilot’s technical topics. The ESP library contains RMI [radio magnetic indicator]. No other con- downloadable reports, workshop presentations clusion is reasonable.” and “just culture” materials. Additional docu- In 1960, the CAB issued a supplement to its ments, CDs, DVDs and reports, such as Airspace revised report, saying that the probable cause Infringement Risk Analysis, are available in the was “a deviation from the prescribed flight path ESP Portfolio Literature section. for reasons unknown.” Safety Alerts Board. Proactive messages for the — Rick Darby ATM community identify safety concerns and best practices. Safety alerts from 2004 to the WEB SITES present are available. Readers may also subscribe for electronic delivery. European ATM Safety Human Factors. This section discusses human Eurocontrol Safety, 56 | flight safety foundation | AEROSafetyWorld | August 2010 OnRecord Landing on the Hudson Ditching the A320 on the river was the only viable option after a bird strike. BY MARK LACAGNINA The following information provides an aware- the first officer began conducting the quick ness of problems in the hope that they can be reference handbook checklist for a dual engine avoided in the future. The information is based failure. The departure controller asked the on final reports by official investigative authori- captain if he wanted to return to La Guardia or ties on aircraft accidents and incidents. try to reach New Jersey’s Teterboro Airport. The captain later told investigators that he decided JETS the airplane was “too far away, too low and too slow” to reach either airport, and that the Both Engines Lost Power only viable option, the river, was “long enough, Airbus A320-214. Substantial damage. Five serious injuries, smooth enough and wide enough.” Thus, he told 95 minor injuries. the controller, “We’re going to be in the Hud- s the airplane climbed out from New York’s son.” He made a public address announcement, La Guardia Airport the afternoon of Jan. instructing the passengers and cabin crew to A15, 2009, the captain briefly paused from “brace for impact.” flight deck tasks to comment, “What a view of The first officer initially attempted to relight the Hudson today.” After a bird strike moments the left engine, which was producing slightly later, he realized that the airplane likely would more power than the right engine. The report end up in the river. noted that the checklist was designed for a dual The A320 had encountered a flock of Can- engine failure that occurs above 20,000 ft. Thus, ada geese 2,818 ft above ground level (AGL), the A320 crew had time only to conduct a por- and each engine had ingested at least two of the tion of the checklist and did not reach the final big birds, said the report by the U.S. National items, which pertain to a ditching. Also, the pi- Transportation Safety Board (NTSB). Both lots did not know that because of core damage, engines had been operating with fan speeds of neither engine could be relighted. Each engine 82 percent. After the bird strike, left-engine fan core had ingested a goose weighing 8 lb (4 kg), speed decreased to 35 percent and right-engine which is more than three times the weight that fan speed decreased to 15 percent. The flight current certification standards require an engine crew immediately activated the engine ignition to withstand during bird-ingestion tests. systems and the auxiliary power unit. About three minutes after the bird strike, the The captain took control of the airplane first officer told the captain, “Two hundred fifty and also handled radio communications while feet in the air. Hundred and seventy knots. … www.flightsafety.org | AEROSafetyWorld | August 2010 | 57 OnRecord Try the other one?” The captain agreed that he aircraft operators to provide pilot training on should attempt to relight the right engine. The the procedures. first officer then advised that airspeed was 150 kt and that he had extended the flaps to position Control Input Causes Hard Landing two. “You want more?” Boeing 717-200. Substantial damage. No injuries. “No, let’s stay at two,” the captain said. He isual meteorological conditions (VMC) told investigators that he chose the flaps 2 set- with intermittent rain showers prevailed ting because he wanted to have enough energy Vat Australia’s Darwin Airport the night of to flare the airplane and reduce the descent rate Feb. 7, 2008, and the 717 flight crew was cleared sufficiently before touchdown; flaps 3 would to conduct a visual approach to Runway 29. Per have increased drag but would not have lowered company procedure, the crew used the instru- the stall speed significantly. ment landing system (ILS) for guidance during As the airplane neared the river, the captain the approach, said the report by the Australian asked, “You got any ideas?” Transport Safety Bureau (ATSB). “Actually [I do] not,” the first officer said. The crew spotted the runway when the The report noted that the A320 was certi- aircraft was at 3,100 ft and about 9 nm (17 km) fied for ditching under standards that assumed from the threshold. The copilot, the pilot fly- in part that engine power is available and ing, used the autopilot’s vertical speed mode to that the descent rate is 3.5 fps. Performance initiate a descent to capture the ILS glideslope. calculations indicated that the actual descent The report said that the descent rate increased rate was 12.5 fps. Recorded flight data showed to over 1,000 fpm, reaching a maximum of 1,600 that calibrated airspeed was 125 kt — nearly 20 fpm, while airspeed varied between 209 kt and kt below the airspeed specified in the ditching 211 kt. portion of the dual engine failure checklist — The air traffic controller told the crew that when the airplane contacted the calm water the pilot of a preceding aircraft had reported a Damage to the lower with a pitch angle of 9.5 degrees and a right rain shower on the approach and that the run- roll angle of 0.4 degrees. Damage to the lower way was wet. fuselage skin allowed fuselage skin allowed near-freezing water to The 717 was at 1,893 ft and descending enter the airplane. at 1,900 fpm when the copilot disengaged the near-freezing water The 150 passengers and five crewmembers autopilot while keeping the autothrottle engaged. to enter the airplane. evacuated the airplane through the forward The aircraft was in landing configuration but still and overwing exits (ASW, 7/10, p. 24). NTSB slightly above the glideslope at 1,379 ft and de- attributed the survival of all aboard to the per- scending at about 700 fpm when it crossed the ILS formance and professionalism of the flight crew outer marker. The descent rate increased again and and cabin crew, the ready availability and rapid was 1,840 fpm when the aircraft intercepted the response of rescuers, and “the fortuitous use” for glideslope at 1,159 ft; airspeed was 153 kt. the domestic flight of an airplane equipped with “The aircraft was then flown slightly below slide/rafts for extended overwater flight. the glideslope,” the report said. The copilot hand Post-accident tests conducted in a flight flew the 717 using the ILS, runway lighting and simulator showed that even if the airplane had the precision approach path indicator (PAPI) as been turned toward La Guardia or Teterboro references. The pilot-in-command (PIC) acti- immediately after the bird strike, it would not vated the windshield wipers when a rain shower have reached either airport. was encountered at 700 ft. “They could see the Based on the investigation, NTSB issued runway lighting and the PAPI, and continued 33 recommendations, including requirements the approach,” the report said. for aircraft manufacturers to develop checklists Airspeed was on target at 136 kt and de- for dual engine failures at low altitude and for scent rate was about 700 fpm until the aircraft 58 | flight safety foundation | AEROSafetyWorld | August 2010 OnRecord descended through a radio altitude of 213 ft 15 148 kt,” or 20 kt above the reference landing seconds before touchdown. The descent rate speed (VREF), said the report issued in July 2010 increased to 1,168 fpm, and the PIC called, by the German Federal Bureau of Aircraft Ac- “Sink rate.” cident Investigation. Company policy for stabilized approaches The aircraft encountered heavy rain and requires a go-around if descent rate exceeds moderate turbulence during the approach. “It was Investigators found 1,000 fpm below 400 ft in VMC. “The PIC re- apparently quite difficult for the crew to keep the ported that he allowed the approach to continue aircraft on the three-degree glideslope,” the report patches of melted because the high rate of descent was considered said. “There was deviation from the glideslope to be momentary and the copilot had taken cor- both above and below.” However, conditions rubber on the main rective action” by increasing the pitch attitude, improved as the Embraer neared the runway. The the report said. approach controller told the crew that the surface landing gear tires. The autothrottle reduced thrust to idle below wind was from 360 degrees at 14 kt. a radio altitude of 30 ft, and the copilot made The aerodrome controller told the crew that an abrupt control input to increase the pitch at- a thunderstorm had passed over the airport and titude. “Had the flight crew overridden the auto- had moved east. The controller said that the throttle and increased thrust in response to the runway was wet but that there was no standing high rate of descent … the severity of the hard water on it. Investigators determined, however, landing may have been reduced,” the report said. that the runway actually was covered by 3 mm Descent rate was 1,072 fpm when the aircraft (about 1/8 in) of standing water, with braking touched down on the main landing gear with a action medium to poor, and estimated that the vertical acceleration of 3.6 g. Embraer required a landing distance of 2,312 m The PIC assumed control and taxied the (7,585 ft) under the existing conditions. The 717 to the terminal. The crew reported the hard estimate also assumed that “the crew flew the landing to company engineers. “The damage aircraft in accordance with all the required to the aircraft included several creases to the parameters,” the report said. fuselage skin above the wing area and to the However, the aircraft crossed 54 ft over the underside of the fuselage behind the wing,” the runway threshold at 150 kt and touched down report said. “Several longerons in the rear cargo 981 m (3,219 ft) from the threshold at 128 kt, in area were also damaged.” None of the 88 passen- what was described by the report as a “soft land- gers and six crewmembers was injured. ing.” The ground spoilers deployed automati- cally. The aircraft was not equipped with thrust Aquaplaning Ground-Loop Overrun reversers, and “the crew reported that braking Embraer 145. Minor damage. No injuries. action failed to bring the aircraft to a stop before nbound from Zurich, Switzerland, with 16 the end of the runway,” the report said. passengers and three crewmembers the after- Groundspeed was about 52 kt when the PIC Inoon of July 18, 2005, the aircraft was nearing steered left, toward a taxiway near the end of the the destination, Nuremburg, Germany, when the runway. “The aircraft ground-looped about 200 crew listened to the automatic terminal informa- degrees to the left, leaving the runway tail-first tion service (ATIS) broadcast, which indicated and coming to rest with the main landing gear in part that surface winds were from 290 degrees units on the grass,” the report said. at 28 kt, gusting to 40 kt. Investigators found patches of melted rub- The crew was cleared to conduct the ILS ber on the main landing gear tires — a sign of approach to Nuremburg’s Runway 28, which is reverted-rubber aquaplaning (hydroplaning), 2,700 m (8,859 ft) long and 45 m (148 ft) wide. which occurs when the wheels lock and fric- “In view of the wind conditions, they increased tional heating forms a “steam cushion” between the approach speed (VAPP) commensurately to the tires and the runway, the report said. www.flightsafety.org | AEROSafetyWorld | August 2010 | 59 OnRecord Surprised by Severe Turbulence pushback commenced he believed that the tug Airbus A330-300. Minor damage. Seven minor injuries. driver was only going to push the airplane about he A330 was more than two hours into a 10 ft [3 m], just enough to trigger the aircraft scheduled flight with 206 passengers and 13 communication addressing and reporting sys- Tcrewmembers from Hong Kong to Perth, tem (ACARS) ‘out’ time.” Western Australia, the night of June 22, 2009, When the tug driver pushed the CRJ700 when it encountered severe turbulence. Six pas- beyond 10 ft, the wing walker signaled the sengers and a cabin crewmember, the only people driver to stop. The other ground crewmembers who were not seated with their seat belts fastened, saw the wing walker signaling the driver to stop, sustained minor injuries, the ATSB report said, and one of them ran toward the driver, trying to noting that the seat belt sign was not on. get his attention. “He stated that the tug driver The PIC consulted with medical personnel was focused on the cockpit of the airplane and aboard the aircraft and at the airline’s dispatch was directing the starting of the airplane’s no. 2 support company, and decided to continue the engine,” the report said. flight. The A330 was landed without further The wing walker “continued to attempt to incident at Perth about five hours later. The in- alert the tug driver; however, the tug driver jured people were treated at a local hospital and did not observe the wing walker before the tail discharged the same day. Examination of the section of the CRJ700 struck the tail section of aircraft revealed minor internal damage. the CRJ200,” the report said. “The empennages The aircraft was at Flight Level (FL) 380 (ap- of both airplanes were substantially damaged.” proximately 38,000 ft) when the turbulence was There were no injuries to the 48 people aboard encountered near Kota Kinabalu, Malaysia. “The the CRJ200 or to the 64 people aboard the cloud associated with the convective activity CRJ700. consisted of ice crystals, a form of water that has minimal detectability by aircraft weather radar,” TURBOPROPS the report said. “Consequently, the convective activity itself was not detectable by [the A330’s] Prop Control Linkage Disconnects radar. As the event occurred at night with no CASA 212. Substantial damage. Two minor injuries. moon, there was little opportunity for the crew he flight crew was conducting a cargo to see the weather … and select the seat belt sign flight on Nov. 1, 2008, from Bethel, Alaska, on prior to the onset of the turbulence.” TU.S., to Toksook Bay, where night VMC prevailed. When the first officer, the pilot Communications Breakdown flying, moved the power levers forward while Bombardier CRJ200, CRJ700. Substantial damage. No injuries. turning from base to final at about 600 ft AGL, ecause there was lightning in the vicin- the right engine did not respond, and the air- ity of North Carolina’s Charlotte-Douglas plane yawed right. BInternational Airport the afternoon of The captain took control and moved June 28, 2008, ground crewmembers were not both power levers full forward to initiate a using headsets for communication. A CRJ200 go-around. “The airplane’s yaw to the right in- had been pushed back from a gate, and its flight tensified, and it began to descend rapidly,” the crew was awaiting taxi clearance when another NTSB report said. “[The captain] said that he ground crew began to push a CRJ700, operated applied full left aileron and rudder to correct by the same airline, from another gate. the yaw but was unable to maintain altitude. “A wing walker was stationed at the He observed that the left engine torque meter [CRJ700’s] left wing, in plain sight of the tug was indicating 100 percent torque and the right driver,” the NTSB report said. “The wing engine torque meter was indicating between walker was aware of the CRJ200, and when the zero and 10 percent torque.” 60 | flight safety foundation | AEROSafetyWorld | August 2010 OnRecord The captain said that he was telling the first 12,500 ft. “There were no other reports of radio officer to feather the right propeller when “the transmissions from the Twin Otter, and it did stall warning horn sounded, the stall warning not arrive at Oksibil,” the report said. light illuminated, and I used both hands to pitch A search was launched about the time at the aircraft forward to avoid a stall.” The 212 which the Twin Otter’s fuel supply would have then struck the tundra. been exhausted, and the wreckage was found Examination of the airplane revealed that two days later about 6 nm (11 km) from Oksibil. the mechanical linkage connecting the right The report said that the aircraft was in a climb- power lever to the right propeller pitch control ing left turn when it struck a mountain at 9,300 shaft had disconnected, preventing the pilots ft; the emergency locator transmitter was unser- from controlling thrust. Company maintenance viceable and did not transmit a signal. personnel had disconnected and reconnected “The aircraft had been flown into cloud the linkage when the right Honeywell TPE331 while tracking toward the gap,” the report said. engine was removed for repairs and a leased “The accident was consistent with controlled engine was installed about two months — and flight into terrain [CFIT] while maneuvering in 237 flight hours — before the accident. “Since the vicinity of the gap. The location of the acci- the bolt that connects the propeller pitch control dent was to the northeast of the route normally linkage to the splined shaft was not found, it flown through the gap to Oksibil.” is unknown if the bolt failed or if maintenance personnel failed to properly tighten/torque the Momentary Incapacitation bolt at installation,” the report said. Beech King Air B200T. No damage. No injuries. he pilot and a crewman were conducting an CFIT Near a Mountain Gap infrared fire-mapping reconnaissance flight De Havilland Twin Otter. Destroyed. Fifteen fatalities. Tin southeastern New South Wales, Austra- he flight crew was scheduled to conduct two lia, the morning of Aug. 31, 2009. The pilot was round-trip flights under visual flight rules flying the King Air with the global positioning T(VFR) between Jayapura, Papua, Indone- system (GPS) coupled to the autopilot. sia, and Oksibil the morning of Aug. 2, 2009. While descending from FL 200 to FL 150 to re- Returning to Jayapura on the first trip, the PIC turn to Bankstown, air traffic control (ATC) made radioed company ground crew to ask for a quick several radio transmissions that the pilot did not turnaround because of deteriorating weather acknowledge, the ATSB report said. The crewman, conditions that might result in clouds blocking who was seated in the cabin and completing tasks ‘The pilot a gap in the mountains along the route, said the associated with the reconnaissance, queried the slumped forward, report by the Indonesian National Transport pilot on the intercom but received no reply. Safety Committee. “The crewman turned toward the pilot and unconscious.’ The Twin Otter landed at Jayapura at 0935 observed that the pilot was suffering what ap- local time and departed with 12 passengers, a peared to be a seizure,” the report said. “Shortly company engineer and the two pilots at 1015 thereafter, the pilot slumped forward, uncon- for the second flight to Oksibil. Estimated flight scious. The crewman moved the pilot back time was 50 minutes, and the aircraft had suf- from the aircraft’s flight controls and checked ficient fuel for 2 hours and 50 minutes of flight. the autopilot and instruments to ensure that the About 35 minutes after takeoff, the Twin Ot- aircraft was under control and pressurized.” ter crew discussed weather conditions with the The crewman was not a pilot, but he had crew of an Indonesian air force Lockheed C130 significant experience with airborne fire- that was en route from Oksibil to Jayapura. The mapping operations. He declared an emergency, C130 crew said that the cloud base at Oksibil telling ATC that the pilot was unconscious. was low and the cloud tops over the gap were at “The aircraft continued to track on autopilot via www.flightsafety.org | AEROSafetyWorld | August 2010 | 61 OnRecord preloaded GPS waypoints to overhead Bank- 240 ft below the summit. Witnesses said that the stown at FL 150 while the crewman attended ceiling in the area was about 200 ft and visibility to the pilot and sought advice from the aircraft was 1/4 mile (400 m) or less in rain and fog. operator and ATC.” The pilot slowly regained consciousness but Belt Buckle Prompts Shutdown initially was unresponsive and appeared unaware Beech 58 Baron. Destroyed. No injuries. of his surroundings. After about five minutes, hortly after departing from a private airport however, the pilot began to respond to ATC in Thabazimbi, South Africa, the after- transmissions and fly the aircraft. He landed the Snoon of Jan. 17, 2009, the pilot and the five King Air at Bankstown without further incident. passengers heard a loud banging noise coming He then was taken to a hospital for observation from the right side of the aircraft. “The pilot and tests, and was released that evening. observed the engine indication parameters, and “It was later determined that the pilot had a they were normal,” said the report by the South previously undiscovered medical condition that African Civil Aviation Authority. However, was the likely cause of the in-flight seizure,” the as the aircraft continued to climb, the noise report said. The pilot told investigators that he became louder. “The pilot then switched off the had experienced a brief but very intense head- right-hand engine [and feathered the propel- ache on the way to work that morning. ler] because he thought it was problematic,” the report said. PISTON AIRPLANES As the pilot turned back to the airport, he told the passengers to ensure that their restraints Ceiling Falls on Air Tanker were fastened. “One of the passengers, seated on Lockheed P2V-7 Neptune. Destroyed. Three fatalities. the copilot’s seat, realized that he had not been efore departing from an air tanker base in strapped in [and that] his seat belt and buckle Missoula, Montana, U.S., for a VFR posi- were hanging out of the aircraft and were the Btioning flight to another wildfire-fighting source of the noise,” the report said. base in Alamogordo, New Mexico, the morn- The pilot attempted unsuccessfully to ing of April 25, 2009, the first officer received a restart the right engine. “The aircraft started weather briefing that included areas of instru- yawing to the right … and became uncontrol- ment meteorological conditions (IMC) and lable,” the report said. “The aircraft was turn- mountain obscuration along the route. ing toward the dead engine. The pilot looked The first officer, the pilot flying, selected for a safe landing area but ran out of time, as an initial cruise altitude of 11,500 ft but subse- the aircraft was descending very quickly.” The quently conducted a series of descents to main- Baron was destroyed when it struck terrain, but tain VFR conditions below the clouds. About no one aboard was hurt. two hours into the flight, the airplane was being flown southeast over Utah’s Great Salt Lake at ‘Should Not Have Been Flying’ 6,000 ft — about 1,800 ft above the surface. The Cessna T310R. Destroyed. One fatality. first officer conducted a further descent to 5,800 witness saw the 310 fly low over her house ft after crossing the shoreline. near Latrobe, Pennsylvania, U.S., the He asked the captain if they were high Amorning of Aug. 31, 2008. She said that the enough to clear the upcoming terrain. “The cap- airplane appeared to be descending very quickly tain did not respond, and the first officer did not “with the left wing up and the right wing down,” challenge the captain about the issue,” the NTSB and the engines sounded as if they were running report said. About 10 minutes later, and shortly at full power. She heard a thud after the airplane after inadvertently encountering IMC near descended below the trees and saw a plume of Stockton, Utah, the P2V struck a ridge about smoke. 62 | flight safety foundation | AEROSafetyWorld | August 2010 OnRecord The NTSB report said that the 78-year- northeast, and the pilot unknowingly conducted old pilot had lost control of the airplane after the approach with a tail wind of 11 kt to 21 kt, becoming incapacitated by a cardiovascular said the report by the New Zealand Transport event during a local flight that originated at Accident Investigation Commission. the Latrobe airport. The pilot did not have a “Nearing the landing site, the pilot brought current medical certificate. A coroner’s report the helicopter to an out-of-ground-effect hover, said that the pilot had been hospitalized re- where it started an uncommanded right yaw,” cently for congestive heart failure and that the the report said. “The pilot attempted to correct pilot’s cardiologist “did not know the decedent the yaw, but the helicopter struck a tree and fell was a pilot and was actively flying an air- to the ground.” The pilot and one passenger plane.” The cardiologist told the coroner that sustained serious injuries. the pilot “should not have been flying with his “The investigation determined that the medical condition.” uncommanded yaw and loss of control resulted from the approach being attempted under con- ditions that were noted in the flight manual to HELICOPTERS be conducive to a loss of tail rotor effectiveness,” the report said. Sun Glare, Illusion Cause CFIT Bell 206B. Destroyed. One fatality. No Weather Brief for VFR Flight hile departing from a helibase in Bell 430. Destroyed. Four fatalities. Carmacks, Yukon, Canada, the morn- he flight crew did not receive a weather Wing of Aug. 9, 2008, the pilot lifted the briefing before departing from Hydera- JetRanger into a low hover, facing away from Tbad, India, the morning of Aug. 3, 2008, the Yukon River and the rising sun, conducted for a 225-nm (417-km) charter flight to a 180-degree pedal turn and then departed over Raipur, with an en route refueling stop in the river. “Shortly thereafter, there was a loud Jagdalpur. Low visibilities and ceilings, and impact and splash, and pieces of the wreckage isolated, embedded thunderstorms were fore- drifted down the river,” said the report by the cast for the route. Transportation Safety Board of Canada. The The crew had filed a VFR flight plan with a pilot, who had 23,000 flight hours, drowned. requested cruising altitude of 3,000 ft direct to The report said that the pilot’s vision likely Jagdalpur, but shortly after departure, the PIC was obscured by the bright sunlight and glare told ATC that they were descending to 2,500 ft from the surface of the water, and that he likely because of weather, said the report by the Indian experienced somatogravic illusion when the Directorate General of Civil Aviation. forward acceleration caused him to believe About 27 minutes later, the helicopter was that the helicopter was climbing rather than about 60 nm (111 km) northeast of Hyderabad descending. when ATC lost radio communication with the crew. A search was launched three hours Control Lost in Hover after the flight’s estimated time of arrival at Kawasaki-Hughes 369D. Substantial damage. Two serious injuries, Jagdalpur. The helicopter’s emergency loca- one minor injury. tor transmitter failed to activate. On Nov. 13, he helicopter was heavily loaded but not the wreckage was found on a hill about 140 overweight when it departed from Haast, nm (259 km) northeast of Hyderabad. The TNew Zealand, the morning of Aug. 11, 2008, aircraft had struck the hill at 2,700 ft, about 80 to transport three track-maintenance workers ft below the top. Local villagers said that there to the Maori Saddle. Because of tall trees, the had been heavy rain in the area when the crash destination could be approached only from the occurred. www.flightsafety.org | AEROSafetyWorld | August 2010 | 63 OnRecord Preliminary Reports, June 2010 Date Location Aircraft Type Aircraft Damage Injuries June 1 near Cayos Arcas, Mexico Bell 412EP substantial 11 none The pilot landed the float-equipped helicopter near a platform in the Gulf of Mexico after experiencing a tail rotor problem. June 2 Midlothian, Texas, U.S. Bell 222UT destroyed 2 fatal Witnesses said that the tail boom and main rotor separated during a maintenance test flight. June 2 Spokane, Washington, U.S. Robinson R22 Beta destroyed 1 fatal The solo student pilot was turning base when a main rotor blade apparently struck and severed the tail boom. June 7 Leeds, England Cessna CitationJet substantial 2 none The airplane overran the runway during a takeoff rejected because of an engine fire. June 7 Edenton, North Carolina, U.S. Beech 60 Duke substantial 1 fatal, 1 serious The pilot said that the Duke struck trees after the flight instructor retarded the left throttle on takeoff during an instrument proficiency check. June 10 Nazca, Peru Cessna 208B NA 9 NA The Caravan did not return from a sightseeing flight and is believed to have been hijacked. June 10 near Port O’Connor, Texas, U.S. Bell 206L-3 substantial 2 minor, 1 none The LongRanger was ditched in the Gulf of Mexico after the tail rotor failed en route to an offshore platform. June 13 Felipe Carrillo Puerto, Mexico Cessna 208B destroyed 9 fatal Visual meteorological conditions (VMC) prevailed when the Caravan crashed on takeoff. June 15 Bankstown, New South Wales, Australia Piper Mojave destroyed 2 fatal The aircraft stalled and crashed while returning to the airport after an engine problem occurred on takeoff for an air ambulance flight. June 16 Ottawa, Ontario, Canada Embraer 145LR substantial 36 NA The nose landing gear collapsed when the aircraft overran the wet runway on landing. Several injuries were reported. June 17 Ruidoso, New Mexico, U.S. Cessna T310R destroyed 5 fatal, 2 serious VMC prevailed when the 310 entered a steep descent and struck terrain on final approach. June 18 Buenos Aires, Argentina MBB BO-105CBS destroyed 2 fatal The helicopter was on an aerial photography flight when it crashed in a residential area. June 18 Chiclayo, Peru Dassault Falcon 20 substantial 8 none The pilots landed the Falcon on open ground after both engines lost power on takeoff. June 19 Yangadou, Republic of Congo CASA 212-100 destroyed 11 fatal The aircraft was on a charter flight from Cameroon when it struck a ridge near the destination (ASW, 6/10, p. 11). June 19 Plymouth, Massachusetts, U.S. Cessna 401 destroyed 3 serious The 401 struck trees and crashed after both engines lost power on final approach. June 21 Kinshasa, Democratic Republic of the Congo McDonnell Douglas MD-82 substantial 101 none The flight crew shut down the left engine after it ingested debris from a tire that burst on takeoff. The nose landing gear would not extend during the return to Kinshasa, and the MD-82 veered off the runway on landing. June 23 Québec City, Quebec, Canada Beech King Air A100 destroyed 7 fatal The King Air struck terrain shortly after one of the pilots reported an engine failure on takeoff. June 23 Puerto Barrios, Guatemala Colemill Panther destroyed 2 fatal The modified Piper Chieftain stalled and crashed on takeoff after a touch-and-go landing. June 23 Kotelniki, Russia Kamov 60 substantial 2 serious The helicopter landed hard and rolled over after birds struck the fenestron on approach. June 26 Broomfield, Colorado, U.S. Lockheed P2V-5 Neptune substantial 2 none The air tanker had brake problems on landing, overran the runway and struck a ditch. June 27 Dublin, Ireland Boeing 737-800 none 1 serious While exiting the 737, a passenger sustained leg injuries when the airstairs partially collapsed. June 30 Wiesbaden, Germany Beech King Air 200 substantial 2 minor An unspecified technical problem occurred during approach, and the flight crew conducted an emergency landing short of the runway. NA = not available This information, gathered from various government and media sources, is subject to change as the investigations of the accidents and incidents are completed. 64 | flight safety foundation | AEROSafetyWorld | August 2010 IASS FSF 63r d a n n u a l In t e r n a t i o n a l Ai r Sa f e t y Se m i n a r No v e m b e r 2–5, 2010 Milan, Italy Under the patronage of the President of the Italian Republic With the endorsement of the Prime Minister f o n d a z io n e 8 ott ob re 2 0 0 Hosted by 1 For information, contact Namratha Apparao, +1 703.739.6700, ext. 101, [email protected], or visit our Web site at flightsafety.org. A SAfe OperAtiOn — A ViAble buSineSS Executive Aviation Safety Symposium October 9-10, 2010 University of Southern California Join leaders in Aviation Safety for a 2-Day Symposium on the university of Southern California’s campus in los Angeles, CA. Keynote Speaker Capt. Chesley b. “Sully” Sullenberger iii Guest Speakers Yannis Yortsos, Dean, USC Viterbi School of Engineering and Capt. Sully Sullenberger prof. Warren bennis Dave pruitt William Voss Michael barr the Honorable robert Sumwalt thomas Almy prof. Jody Hoffer Gittell Victor D’Avanzo http://viterbi.usc.edu/aviationexec US 1571 Aviation Mag Ad.indd 1 8/4/10 1:00 PM