Remarks of John Hammerschmidt
Member, National Transportation Safety Board
At the National Convention of the Home Office Life
Underwriters Association
Washington, D.C. , April 28, 1998


Introductory Remarks

Thank you for inviting me to speak to you this morning, and for this opportunity to tell you a little about the National Transportation Safety Board and how our work may be of interest to your work, especially as regards general aviation safety.

First, let me describe who we are. The National Transportation Safety Board, with 396 employees, is one of the federal government’s smallest agencies, but is charged by Congress with investigating every civil aviation accident [ The Safety Board defines and accident as an event that results in substantial damage to an aircraft or serious injury to a person] in the United States and significant accidents in the other modes of transportation---railroad, highway, marine and pipeline---and issuing safety recommendations aimed at preventing future accidents.

On call 24 hours a day, 365 days a year, Safety Board investigators may be required to fly anywhere in the world at a moments notice to investigate significant accidents. In the years I have been at the NTSB, we have investigated many high profile accidents, such as the Space Shuttle Challenger and the Exxon Valdez, and a great many more less notable accidents - such as the one involving the nuclear attack submarine USS Houston.

Since it’s inception in 1967, the Safety Board has investigated more than 100,000 aviation accidents and issued almost 10,000 safety recommendations in all transportation modes and in so doing has become the world’s premier accident investigation agency.

The Safety Board is also responsible for maintaining the government’s data base on civil aviation accidents, for conducting special studies of transportation safety issues of national significance, and for providing investigators to serve as U.S. Accredited Representatives overseas in connection with accidents involving U.S.- registered aircraft, or involving aircraft or major components of U.S. manufacture. Moreover, to ensure that pilots , aviation mechanics, and mariners are treated fairly and objectively, the NTSB and it’s Office of Administrative Law Judges may serve as the "court of appeals" in connection with violations or certificate action taken by the Federal Aviation Administration or the U.S. Coast Guard Commandant, or when civil penalties are assessed by the FAA. This formal review process also affords the Board Members with many detailed snapshots of what is happening in the aviation system, usually involving circumstances and incidents that are not accidents. For example, we often review appeals by airmen of denials by the FAA of the issuance of medical certificates, for one reason or another.

Recent legislation has also given the Safety Board responsibility for accidents involving public use (government-operated ) aircraft, except those operated by the armed forces and intelligence agencies. In addition, the Board now coordinates all Federal assistance to the families of victims of catastrophic transportation accidents .

At an annual cost of less than 15 cents a citizen, the NTSB is one of the best buys in the government.

Investigation of General Aviation Accidents

Whereas the investigation of major airline accidents, such as TWA Flight 800, is conducted by the Board’s major accident division, the investigation of general aviation accidents is conducted by the Safety Board’s Regional Operations and General Aviation Division. This division consists of 56 aviation investigators located in the Board’s Washington, D. C. headquarters and nine regional and field offices in major cities across the United States: Anchorage; Atlanta; Chicago; Denver; Dallas-Fort Worth; Los Angeles; Miami; New York; and Seattle. The Division currently investigates approximately 2,100 accidents and incidents annually, either as field investigations or "limited" investigations. Field investigations are usually conducted by a single investigator who , working with party representatives, e.g., the FAA, airframe and engine manufacturers, gathers all of the detailed on-scene information pertinent to the accident. "Limited" investigations generally are conducted for nonfatal aviation accidents in which the aircraft is substantially damaged or destroyed and there are only minor injuries, and are essentially desk investigations in which the FAA provides us with the factual information.

Analysts located in the Washington, DC headquarters are responsible for reviewing the standardized accident forms and narrative reports prepared by the regional investigators for the public record of the investigation . Computer brief reports concerning relevant facts, findings, and probable cause(s) and factors are prepared for all field and limited investigations , and summary narrative reports are prepared for selected cases. The data from all investigations are maintained to identify trends, assess program effectiveness, provide statistical support for Safety Board studies and safety recommendations, and support other related safety purposes.

Overview of General Aviation Accidents

Speaking of trends, the safety of general aviation has improved steadily over the years, with the overall accident rate diminishing from a high of about 15 accidents per 100,000 flight hours in 1974 to an all-time low ( based on preliminary data ) of 7.51 in 1997. In 1997, the fatal accident rate ( 1.42 ) and the number of fatal accidents ( 350 ) were the lowest in recent history , although the number of fatalities ( 646 ) increased slightly from the record low ( 631 ) established in 1996.

That’s the good news. The bad news is that pilots are still cited as a cause or contributing factor in about 80% of all general aviation accidents. While new technology and design have clearly played an important role in the improved safety record, the potential for continued improvements down the road may be diminishing and we need to focus on human factors elements affecting the safety of flight operations to continue this historical accident reduction trend: factors such as situational awareness, error chain detection and analysis, communication skills, decision-making, stress management, etc.

Of course, the Safety Board investigates all types of general aviation accidents: including those involving fatalities, air traffic control, midair collisions, newly certificated aircraft or engines, in-flight fire, in-flight breakup, hot-air balloons, crop dusters, and those that expose the public to high risk or attract high public interest.

(in order to give a real-world picture of what is happening in the system, mention the Daily Alert Bulletin from yesterday morning -- indicating preliminary reports of four general aviation fatal accidents over this past weekend:

• 2 fatal when VFR Cessna 172 nosed over and crashed ¾-mile short of runway (Ramona, CA airport);

• 1 fatal when VFR Piper PA32 ran off runway 05 on departure and crashed through perimeter fence (New Bedford, MA airport);

• 3 fatal when VFR Beech BE58 crashed while attempting approach, but runway was not lighted (Kerrville, TX). Aircraft given heading and distance to alternate airport;

• and a "local" accident involving a helicopter , 2 fatal when a Hiller FH-1100 crashed under unknown circumstances (7 miles south of Stevensville, MD on Kent Island.)

Speaking of public interest, earlier this month the FAA prominently announced here in Washington its current safety agenda for both commercial aviation and general aviation. In general aviation the agenda will be focussing on pilot decision-making, loss of control, weather, controlled flight into terrain, crash survivability and runway incursions. Of course, these are all important safety areas that dovetail with the Safety Board’s work over the years. I might mention for clarification that the NTSB is an independent government agency, and that the FAA has the right, by law, to be a party to all NTSB aviation investigations. The two agencies work closely, yet independently. This morning I would like to focus on that first safety area that was referenced - pilot decision making - not only because it is exceptionally important, but because it is often neglected in the overall mixture of safety considerations.

Case Studies

In order to validate this point, I would like to give you a few specific details about six accidents that occurred in 1996 - which I hope will provide a real-world picture of what our accident investigations are revealing.

Although most general aviation accidents are not accompanied by media headlines, the human side of an accident --one which we can all readily identify with -- as well as the unfortunate, but all too frequent, accident causal role of the pilot, is sometimes dramatically and tragically emphasized -- as it was two years ago (this month) in the Jessica Dubroff case.

As you may recall, Jessica, the 7 year old trainee, accompanied by her father and by the pilot-in-command, were engaged in a trans-continental record attempt involving 6,660 miles of flying in 8 consecutive days. The pilot in command was 52 years old and was a stockbroker by profession. He held a commercial pilot certificate with instrument rating, and a flight instructor airman certificate. The airplane was a single-engine Cessna 177B. The 1st leg of the trip ( about 8 hours of flying ) had been accomplished the previous day and began/ended with considerable media attention. On the morning of the 2nd day, in Cheyenne, Wyoming, the pilot-in-command and Jessica participated in media interviews, pre-flighted , and loaded the airplane.

Photo of 7-year old pilot near plane.2-minute video (3.4M)

(text description)

This video is very poignant - and shows the reality of how quickly a pilot can get into trouble. The pilot-in-command had received a weather briefing and was advised of moderate icing conditions, turbulence, IFR flight precautions, and a cold front in the area of the departure airport. The airplane was taxied in rain to take off on runway 30. While taxiing, the pilot-in-command acknowledged receiving information that the wind was from 280 degrees at 20 gusting to 30 knots and that a departing Cessna 414 pilot reported moderate low-level windshear of +/- 15 knots. The airplane then departed on runway 30 towards a nearby thunderstorm and began a gradual turn to an easterly heading. Witnesses described the airplane’s climb rate and speed as slow, and they observed the airplane enter a roll and descent that was consistent with a stall. Density altitude at the airport was 6,670 feet. The airplane’s gross weight was calculated to be 84 pounds over the maximum limit at the time of the impact.

The Safety Board determined the probable cause of the accident to be the pilot-in command’s improper decision to take off into deteriorating weather conditions (including turbulence, gusty winds, and an advancing thunderstorm and associated precipitation ) when the airplane was overweight and when the density altitude was higher than he was accustomed to, resulting in a stall caused by failure to maintain airspeed. Contributing to the pilot in command’s decision to take off was a desire to adhere to an overly ambitious itinerary, in part, because of media commitments.

The safety issues identified in this investigation included fatigue, the effects of media attention and itinerary pressure, and aeronautical decision making. As a result of this accident, the Board made a safety recommendation to both the Aircraft Owners and Pilots Association and the National Association of Flight Instructors to :

"Disseminate information about the circumstances of this accident and continue to emphasize to your members the importance of aeronautical decision making."

The Board also recommended that the FAA:

"Incorporate the lessons learned from this accident into educational materials on aeronautical decision making."

Of course, when we refer to aeronautical decision making, we are not talking about a pilot’s skill in controlling an aircraft, rather we are simply talking about good judgement versus bad judgement. But, as the following fatal-accident investigations will show, there continues to be too much poor judgement prevailing over good judgement in the world of general aviation.

These investigations are five other "case studies" from 1996. Let me mention that I did not select these accidents because they all fit a pattern - rather, I had asked our staff to randomly select a few general aviation accidents that might be of interest to a large group. Once selected - it was apparent that there was a common safety issue in all of them.

Example No. 1

Middleburg, Florida
December 22,1996

(View of main wreckage)

The noninstrument-rated 48 yr.-old private pilot of a single-engine airplane departed in dark night conditions on a local VFR flight at about 0530 EST. Instrument meteorological conditions prevailed due to fog. The reported weather was: lowest ceiling- 300 FT. overcast; visibility - 1/2 mile; with fog, blowing spray, and precipitation. The flight did not return and Air Force satellites began receiving an emergency locator transmitter signal from the crash site sometime after 0600. The wreckage was located 1 mile southwest of the departure airport. The aircraft had crashed in a wooded area. During examination of the crash site, all components of the aircraft were located on or around the main wreckage and there was no evidence of preimpact failure or malfunction of the aircraft structure, flight controls, or engine. The engine tachometer showed the aircraft had flown about 38 minutes since departure. No record of the pilot having received a weather briefing was located after the accident.

The Safety Board determined that the probable cause of this accident was :VFR flight by the noninstrument-rated pilot into instrument meteorological conditions (IMC), which resulted in spatial disorientation and an uncontrolled descent into a wooded area. Factors relating to the accident were: failure of the pilot to obtain a preflight weather briefing, darkness, the adverse weather condition (fog and/or low ceiling), and the pilot's lack of instrument experience.

Example No. 2

Hailey, Idaho
June 21, 1996

(View of surrounding terrain. Arrow points to wreckage and indicates direction of wreckage distribution.)

The 43 yr.-old pilot of a Cessna 150B had performed a preflight inspection of the single-engine airplane and had it fully serviced with fuel before taking off to Manila, UT. Postaccident calculations indicated that the airplane was about 50 pounds over its maximum gross weight limit, and its center-of-gravity was behind the aft limit. The airport elevation was 5,315 feet; density altitude at the airport was about 5,840 feet. Immediately after departure in VFR conditions, the pilot flew the airplane toward rising terrain, despite the control tower's advisement to initiate a left turn over lower, level terrain. The controller asked the pilot for his position, and the pilot replied, "I'm heading up the valley . . . east of the airport . . . I'm just about to come out of the valley at the east end." No distress call was received from the airplane. Subsequently, the airplane crashed at an elevation of about 6,700 feet, below a saddle in mountainous terrain. There was evidence that the airplane had stalled. The pilot and his passenger were killed.

The Safety Board determined that the probable cause of this accident was : the pilot's improper planning/decision, and his failure to maintain adequate airspeed during a climb over rising terrain, which resulted in a stall and collision with terrain. Factors relating to the accident were: failure of the pilot to ensure the airplane was within its weight and balance limitations, high density altitude, and the rising/mountainous terrain.

Example No. 3

Perry, Missouri
April 12, 1996

(Impact Point in Water)

The 23 yr.-old pilot and two of his passengers were killed when their single-engine airplane, a Cessna 177B, collided with electrical power lines that were about 700 feet downstream from a dam and about 125 feet above water. The surviving passenger stated that he and the pilot had consumed some beer at his home. After consuming the beer, the passenger said they went to a local restaurant and lounge remaining there until about 0130. The passenger said he and the pilot left the restaurant with two friends that they had met at the restaurant. One of the friends bought a 12 pack of beer en route to the airport. The pilot and three passengers flew for about an hour before descending near a dam. The passenger said the pilot told him he wanted to get a closer look at the dam. Also, the passenger said that before the airplane crashed, he saw the dam above the airplane's front window, then he heard the airplane hit something and saw sparks. Toxicology tests at the FAA CAMI laboratory showed the pilot had ethanol levels of 130 mg/dl in blood, 135 mg/dl in vitreous fluid, and 179 mg/dl in urine.

(mention that an NTSB safety study of alcohol-related general aviation accidents that occurred during 1983-88 indicated that the mean BAC of the pilots was about 150 mg/dl)

The Safety Board determined that the probable cause of this accident was: the pilot's impairment of judgment and performance due to alcohol, and his failure to maintain altitude/clearance from the power transmission wires. Darkness, restricting the pilot's ability to see-and-avoid the transmission line, was a related factor.

Barkhamsted wreckage.

Example No. 4

Barkhamsted, Connecticut
April 6, 1996

(Right-side view of wreckage)

The pilot/owner of a Cessna 172 had been invited to fly to an airport in Hudson, NY and have dinner with two other pilots who were taking their planes. He was hesitant to fly because of a nonfunctioning artificial horizon and a landing light problem. Witnesses stated that the pilot was encouraged by a Certified Flight Instructor (CFI)-rated passenger who wanted to go along but did not have room in the other airplanes.

A witness stated that the owner of the plane did not think it would be wise to take his airplane because of the equipment problems as well as the forecasted weather. The witness recalled that the flight instructor stated that they would have no problem landing without landing lights and if they experienced bad weather they could fly the valleys at 1500 feet. The witness stated that the owner again indicated his concern about making the trip because of his aircraft’s problems. As the witness was departing the airport that afternoon, he stated that he saw the owner again and asked if he was going on the trip. The owner stated that he was. The witness stated that he then asked if the owner had been talked into it by the flight instructor. The witness stated that the owner’s response was in the affirmative.

The flight to the airport in daylight went without incident. The other pilots stated that nothing was mentioned at dinner about problems with the airplane. The three pilots coordinated their night departure in order that the fastest of the three airplanes took off first, followed in succession by the two slower airplanes. The pilot/owner of the Cessna 172, with the CFI-rated passenger seated in front and with his wife in a rear seat, took off as the third airplane. There was approximately 5 to 10 miles between airplanes. The three pilots were talking amongst each other and the lead pilot announced that he had encountered instrument weather conditions and was going to fly an instrument approach. The lead pilot told the other pilots to deviate their routes to the south in order to remain VFR. The second pilot deviated to the south and radioed to the third airplane. The CFI in the third airplane answered and stated that they would also deviate to the south, but that they had lost their artificial horizon and were having problems. The second pilot recommended that they switch to approach control in order to get assistance. The CFI's voice was recorded on the approach controller's frequency stating that he was having problems and had lost his artificial horizon. Within minutes of initial contact, the airplane disappeared off the radar scope. The airplane wreckage was located the next morning.

The Safety Board determined that the probable cause of this accident was : the CFI pilot/passenger's poor judgment which led to spatial disorientation and subsequent loss of control of the airplane. Factors involved in this accident were the night marginal visual flight conditions and continuing flight with a known nonfunctioning artificial horizon.

Reform Accident.Example No. 5

Reform, Alabama
December 5, 1996

(view of main wreckage)

The flight had been airborne 12 minutes when the airplane, a Piper PA-32, had a total engine failure. The 41 yr.-old pilot, the only occupant, was given vectors to an airport and was advised that the airport did not have lighting available. About 1 mile from the airport the pilot could not find the airport in the dark and said, "I'm going for the dark spot...." Radio and radar contact was lost, and there was not further communications with the pilot. A witness had observed the airplane at a low altitude west of the airport, and said: "...it was like it was gliding... no sound...and was descending lower and lower...I didn't see anything, I didn't hear the sound of an airplane. The engine or anything...." The engine, the fuel injection system, the fuel found in the airplane, were all examined, and no discrepancies were found. Both magnetos were tested on a test stand, and the tests revealed that both the coils from each magneto were found cracked and leaking. The aircraft's logbooks showed that the last entry and the last annual were performed on 8/1/88, over 8 years before the fatal flight. Since 1988, the airplane had flown a total of 30 hours. No ferry permit was issued for this flight, and no maintenance was performed before the flight.

The Safety Board determined that the probable cause of this accident was : a complete loss of engine power due to a total failure of both magnetos, and the failure of the pilot-in-command to insure that the airplane was airworthy. The dark night and unsuitable terrain were factors.

That gives you a little of the flavor of what we mean at the Safety Board when we talk about problems in aeronautical decision making.

The picture of pilot decision making in these accidents indicates two things:

One, it’s going to be hard to eliminate this problem, because a lot of it involves the actions of individuals and "human nature."

But two, and much more hopeful, even though these kinds of accidents happen time and time again, year in and year out, when we look back at the accident rates over the years there has been a substantial and significant improvement.

Recent developments in the area of aeronautical decision making have focused on decision making involving real life situations, in which decisions must often be made rapidly in response to changing and ambiguous circumstances. This work has emphasized the importance of experience for rapidly assessing situations and choosing workable alternatives.

The Safety Board is aware of several recent initiatives to upgrade the teaching of decision making to general aviation pilots. For example, the Air Safety Foundation of the Aircraft Owners and Pilots Association (AOPA) has developed a pilot training seminar entitled "Never Again" that is being presented to pilot groups and that focuses on actual weather-related incidents. By using videotape reconstruction and regular audience discussion, the seminar presents decision making issues in a manner that is compelling and closely related to actual pilot experiences. The Safety Board is also aware that the National Association of Flight Instructors (NAFI) is developing a new program in decision making skills aimed at flight instructor recertification training. It will emphasize judgement in concrete situations facing pilots. The Safety Board commends these efforts.

In closing, let me point out that despite the safety improvements we’ve seen and the record low accident rates achieved this past year, I believe that we’re only at the beginning of identifying what kinds of requirements and training are effective in improving the accident rate even further. I believe that your part of the insurance industry will be involved, to the extent that you can help us, the FAA, the aviation industry, and the property/casualty insurers to identify what generates risk in general aviation, and what controls it.

Thank you for your kind attention.


NTSB Home | Contact Us | Search | About the NTSB | Policies and Notices | Related Sites