SR SSD 2002-18
6/2002

Technical Attachment

"VFR Flight Not Recommended"
A Study of Weather-Related Fatal Aviation Accidents

Douglas C. Pearson
NOAA/National Weather Service
Atlanta Center Weather Service Unit

Abstract

This study is a statistical summary of all fatal aircraft accidents that occurred in the United States (including Alaska and Hawaii) and coastal waters which involved general aviation and small commuter aviation aircraft from 1995-2000. These data summarize fatal accidents in which the National Transportation Safety Board (NTSB) cited weather to be a cause or contributing factor. From 1995 to 2000, 4018 people died in aircraft accidents; of which weather related accidents accounted for 1,380 deaths. Essentially, the NTSB cited weather as a factor in 3 of every 10 fatal aircraft accidents during this period.

The National Weather Service Vision states a need to strive to eliminate weather-related fatalities." The NWS theme for 2002 is "Working Together to Save Lives." The yearly average for weather-related fatalities in general aviation (230) is comparable to the combined fatalities due to lightning, tornadoes and floods (213). Results of this study will provide a statistical data base that justifies an expanded outreach program. This program will be targeted to the general aviation community

Introduction

The Federal Aviation Administration Flight Service Manual requires flight service stations (FSS) to use National Weather Service data and products when providing pilots with a flight weather briefing. As part of this briefing, the FSS attendant makes a recommendation on the appropriateness of a Visual Flight Rules (VFR) flight. If the weather is observed or forecast to be Marginal VFR (MVFR) or Instrument Flight Rules (IFR), and VFR flight is doubtful, the attendant will advise the pilot that "VFR Flight Not Recommended (VNR)." MVFR or IFR conditions were a factor in nearly 70 percent of the weather-related fatal accidents. In many of these cases, the pilot either chose to ignore the information provided at the weather briefing or inadvertently flew into adverse IFR weather.

This is not just a recent problem. In their report, Special Study of Fatal, Weather-Involved, General Aviation Accidents, the NTSB cited 2,026 fatal weather-involved accidents that killed 4,714 from 1964 to 1972. These weather-involved accidents represented 36.6 percent of the total fatal accidents. Similarly, the Aircraft Owners and Pilots Association (AOPA) Air Safety Foundation completed an extensive study of general aviation accidents for the period 1982-1993. AOPA results showed a decline in the percent of weather-involved fatal accidents, from a high of 43 percent in 1982 down to a low of 24 percent in 1991. Despite the general downward trend, however, weather-involved accidents averaged 34 percent. Both of these studies also concluded that low ceilings, fog and attempted VFR flight into Instrument Meteorological Conditions (IMC) were the most frequently cited cause or factor in weather-involved fatal general aviation accidents.

The following quote from the August 28, 1974, NTSB study is just as applicable today as it was 28 years ago:

"These accidents occurred with disturbing regularity despite improvements in aircraft, instrumentation, training, training facilities, the air traffic control system, weather facilities, weather services, and navigational aids."

Weather-involved fatal accidents have consistently represented 30-36 percent of the total fatal accidents and, although there has been a slight downward trend over the past 28 years, there remains a need to dramatically reduce these numbers.

Methodology/Data Analysis

The objective of this study is to quantify the significance of adverse weather on fatal accidents involving small aircraft that fall within the category of general and commuter aviation. The magnitude of this sector of aviation is extensive, with over 192,000 registered general aviation aircraft. This class of aircraft is most vulnerable to hazardous weather or weather conditions that exceed the aircraft and/or pilot capabilities.

From 1995-2000 there are 2,605 NTSB records documenting fatal aircraft accidents. From these records, only those accidents that occurred in the United States (including Alaska and Hawaii) and its coastal waters were counted. Thus, 293 accidents were eliminated because the NTSB listed them as occurring outside the United States. Additionally, nine major air carrier accidents were eliminated as was one duplicate record. The remaining 2,302 fatal accident reports and NTSB conclusions were thoroughly reviewed using the NTSB and the FAA National Aviation Safety Data Analysis Center(NASDAC) Web sites. This period of record was used in order to provide the most current data as well as allow for sufficient sample size. Furthermore, the FAANASDAC recommends using the most recent five years for safety analysis and monitoring because the aviation industry is so dynamic and impacted by technology.

The review process initially involved accessing the NTSB accident data base and doing a sort by year for all fatal accidents. The accident summary, the NTSB determined cause and full narrative were reviewed and the data classified. If weather was cited as a cause or factor then the following information was recorded and tabulated on a spreadsheet:

- date
- number of fatalities
- total pilot time
- weather briefing source
- location (city/state)
- cause of accident
- weather phenomena as cause/factor
- phase of flight

After reviewing each record from 1995-2000, weather was cited as a cause or factor in 697 fatal accidents. Note that in some cases weather was not listed as a cause or factor, but was adverse at the time of the accident. The latter cases were not included in the totals cited in this study, but were kept on a separate data sheet. Also, environmental conditions that favored carburetor icing and high density altitude were not included in this study, but again were included on the separate data sheet. Finally, there were 36 accident reports which had not been finalized, most of these from 2000. Each of these preliminary reports was evaluated and only the obvious weather-related accidents were included in this study. The final total reflects a conservative summary of weather related fatal accidents using the following ten adverse weather phenomena:

            - Low ceiling                          - thunderstorms
            - fog                                        - icing
            - rain                                       - updrafts/downdrafts
            -snow                                     - tailwind/crosswind
            - turbulence                            - other 

Each of these accidents was also cross-checked using the FAA NASDAC database to verify information and in some cases supplement the data collection.

Data were analyzed and summarized by month and year, and were averaged over the period of record (1995-2000). Each fatal accident report was evaluated to determine the single most likely associated weather event. Other weather phenomena existing at the time were labeled as two, three and so on. Also, low ceilings and fog were considered as related weather events and they were combined into one category. Similarly, all wind-related (turbulence, updrafts/downdrafts, and tailwind/crosswind) fatal accidents were combined. These categorizations were done to simplify the results and enable the recommendations to be targeted toward a few key areas.

Results

The results of this study focus on quantifying the significance of adverse weather in fatal aircraft accidents, highlighting the long term nature of this problem, and compiling data so targeted recommendations can be identified. A quick look at the statistics for the last six years, as presented in Fig. 1, show the numbers to be fairly consistent. There were typically 350-400 fatal accidents per year during this period, with over 600 fatalities.

ssd02-18fig1.jpg

Fig. 1

Weather was a cause or contributing factor in approximately 100 of these accidents and accounted for over 200 fatalities each year. These data provide convincing evidence that adverse weather still plays a significant role in the day-to-day operations and decision making for pilots.

Figure 2 is a different representation of the values presented in Fig. 1 and shows that targeting adverse weather can provide a focus for addressing 30 to 35 percent of the accidents and fatalities. These results are consistent with the 1974 NTSB Study of Fatal Weather-Involved General Aviation Accidents, and the exhaustive AOPA analysis completed in 1996. The concern here is that in spite of a reduction in the total number of fatalities over the past 28 years, the percentage of these fatalities that is directly related to weather has remained relatively stable. The focus can be further refined quickly by reviewing the specific causes and factors involved with these fatalities.

ssd02-18fig2.gif

Fig. 2

In my review of 697 weather-related fatal accidents there were four causal links that the NTSB repeatedly addressed in their reports. One causal link area reviewed was the quality of the NWS forecast support. This includes all surface observations across the country, the aviation forecast products and terminal forecasts produced at each of the 120 NWS Weather Forecast Offices, and the NCEP Aviation Weather Center (AWC) advisory bulletins which include convective SIGMETs and AIRMETs for IFR/mountain obscuration and turbulence/icing. Additionally, Center Weather Advisories and Meteorological Impact Statements are issued as needed by each of the 21 NWS Center Weather Service Units collocated at FAA Air Route Traffic Control Centers. In these accidents, the NTSB cited NWS related weather support to be a contributing factor in only two of the accidents. This is an indication that the National Weather Service is providing superb support!

A second causal link addressed by the NTSB was the support provided to the pilot by the servicing Flight Service Station. There are over 3,000 FSS personnel providing weather briefing support. It is the responsibility of FSS personnel to provide the pilot with the latest available flight weather data prior to the pilot making his or her decision to fly. Of the 697 weather- related fatal accidents, the NTSB cited this type of support as a contributing factor in only five cases! Thus, it appears as though this segment of the weather support loop is also doing its job.

Another causal link involved the extensive Air Traffic Control (ATC) system and the 17,000 plus personnel directing aircraft. The controllers provide vital assistance to the pilot once he/she is airborne. When a pilot inadvertently flies into adverse weather, the controllers can be the life-saving link in getting the aircraft down safely. In the accidents reviewed, ATC support was faulted as a cause or contributing factor nine times. Taken together then, these three causal links accounted for a very small fraction of weather-related fatal aircraft accidents.

The final causal link was the pilot! The decision-making process and errors made by the pilot dominate the NTSB findings. The most common pilot error was continued flight into IMC, often resulting in loss of control due to spatial disorientation. The NWS is doing its job well, the FSS personnel are providing excellent briefings, and the ATC personnel are very professional in their handling of aircraft. It is impossible to discern the exact number of lives thereby saved. Collectively, these agencies are saving lives every day when a pilot makes a "smart decision" based on available weather data or properly uses these services once in flight. But for all the technological advances, the advancements in forecasting, improved aircraft, rapid communication of data, and better automation in the ATC system, safe flying still boils down to the decision-making process and skill on the part of the pilot.

Figure 3 provides a summary of the weather conditions most likely to cause these fatal accidents and can help the pilot to focus his/her decision making process. On the one hand, pilots most certainly have a greater understanding of the intensity and inherent danger of a thunderstorm. They respect what ice can do to an airplane's aerodynamics. It's the relatively tame and more frequently occurring clouds and fog, however, that are contributing to 63 percent of the fatal accidents. When this factor is combined with rain and snow, another player in the restriction of visibility, IMC problems account for nearly 70 percent of all weather-related fatal accidents. Simply stated, these pilots died because they could not see where they were going!

ssd02-18fig3.gif

Fig. 3

The 2000 AOPA Nall Report stated the same as a perplexing question when addressing VFR flight into IMC:

"What is it about the fact that they can no longer see the ground that pilots don't understand? Because so many of these accidents were fatal, there are few surviving pilots to answer the question."

The FAA Aviation News published a similar article entitled "SIGMETs, AIRMETs, Thunderstorms, and the Force." Though another weather-related accident is highlighted, the "Force" addresses the mysterious decision-making process that every pilot is faced with:

"Unfortunately there is an insidious force that works to flaw our rational decision making. I'll use the technical terminology of "it worked last time" to describe the force. Whether it is inadequate preflight planning, pushing a fuel supply, or taking a "look see" at forecast bad weather, the force often starts out weak and allows bad decisions to pass. But with each exposure the force strengthens and further clouds good decision making. Eventually the force demands a high fee and catches the unwary off guard."

This bad decision-making process often plays a role when dealing with the second most significant category of weather that impacts fatal accidents - the wind. Small aircraft have lower tolerances for maneuverability and handling in strong winds. Tailwinds, crosswinds, gusty winds, updrafts/downdrafts around high terrain, and turbulence contribute to 18 percent of the fatal weather-related accidents.

Figure 4 provides a comparison of fatalities associated with the general public's vulnerability to severe weather and aviation weather-related deaths. These data show the key role that the NWS plays in these two essential areas of public support. By measuring the number of fatalities, it is easy to see where weather impacts the general public. Figure 4 shows that on average the combined number of fatalities associated with tornadoes, lightning and floods are very close to the total number of fatalities in weather-related aviation accidents - 213 and 230 per year, respectively.

ssd02-18fig4.gif

Fig. 4

Analyzing the data in these two groups is not a direct comparison. Causes related to these fatalities and decision-making processes involved are quite different. Sometimes severe weather strikes and claims the lives of helpless victims, such as when a violent thunderstorm or tornado passes through a mobile home park. On the other hand, most of aviation fatalities are preventable because a conscious decision is made to take off. (Perhaps the aviation fatalities can be more directly compared with flash flood deaths, in the sense that too often, at least, individuals put themselves at risk by driving into flooded low water crossings, despite the hazard which should be obvious.) Regardless, both areas of public support benefit from the products and service provided by the NWS.

While there are no obvious trends in weather-related aviation fatalities, there has been a downward trend in severe weather fatalities over the past two years. Identifying any single cause and effect of a trend is difficult, but technology, improved equipment, better forecasts and an aggressive public awareness campaign certainly have contributed to reducing the fatalities associated with severe weather.

The four figures presented here summarize data from over 2000 fatal general aviation and small commuter airplane accidents. These figures should serve to increase the aviation community's awareness toward IMC weather as the single most significant factor in fatal accidents. The data also provide compelling evidence that shows the equal role the NWS plays in two very important areas of public support - the general public and the aviation sector.

Summary and Recommendations

The objective of this study is to show the significance of adverse weather on fatal accidents involving small aircraft. The intent was to review a large amount of data and summarize the results so efforts could be focused in a few critical areas. As indicated here and in previous studies, adverse weather is a primary factor in fatal aircraft accidents. Repeatedly, low ceilings, fog or other visibility restrictions were the leading cause or a contributing factor in these fatal accidents. By all indications, the primary agencies (FAA/ATC, FAA/FSS and NWS) responsible for supporting the pilot in dealing with weather and aviation are doing their job and doing it well. Pilots continue to err in their decision making process. Pilots continue to take off and fly into IMC conditions even though they have access to the latest weather information and have received a complete pilot weather briefing.

Recommendations to remedy this problem are cited in numerous publications, agency goals, and articles. The FAA safety program, Safer Skies - A Focused Agenda, targets weather as a safety issue for general aviation and routinely carries weather safety articles in their publications.

In a multi-agency publication, National Aviation Weather Initiatives, weather is cited as a factor in 23 percent of all aviation accidents and annually costs the country an estimated $3 billion for accident damage, related injuries and delays. This thorough assessment identified 86 initiatives to improve aviation weather safety and services. The AOPA has an aggressive education and awareness program including seminars, recurring publications and articles. The 1974 NTSB study listed 10 recommendations, and seven of those dealt with pilot training or familiarization of aviation meteorology. Their report concluded stating:

"an emphasis on weather awareness is required at all levels of pilot education..."

Education and training concerning inadvertently entering IMC is a common theme in each of these studies, a recurring problem cited in NTSB accident reports, and the primary conclusion of this study. A special emphasis is needed for pilots with little or no IFR experience. For example, seven people died in Alaska in June, 1999, due to a pilot's VFR flight into adverse weather, followed by spatial disorientation and failure to maintain aircraft control. The NTSB investigator asked the chief pilot of the company if he conducted any training for emergency use of basic flight instruments. He replied that he never did and emphasized that the company's policy was to "go down and slow down but never go into instrument conditions." When asked what he would do if he found himself in an IMC situation, the chief pilot indicated he was uncertain because he never intended to be in that situation.

With limited resources, it would be logical for all agencies concerned to target the number one causal factor and use the most cost-effective approach to reduce this factor. The results of this study indicate the focus must be on the pilot's decision-making process when assessing whether to fly when IMC conditions exist or are forecast. This study provides the NWS with ample justification for and the opportunity to partner with the FAA, AOPA and other aviation related organizations to expand an "outreach" program for aviation weather safety. This effort should be comparable to current NWS commitment to public education and the increased awareness on the part of the general public concerning severe weather.

Toward that end, it is suggested that a comprehensive NWS Aviation Weather Safety outreach program must include:

a. Using results of this study to create an urgency within the aviation community. NWS personnel at WFOs and CWSUs should make aviation safety presentations to local flying clubs or flight training schools. These presentations should include weather impact information similar to the results in this study, as well as resources available to the pilot. For example, an outstanding NWS Web site that some pilots are not familiar with is operated by the AWC and is available at http://adds.aviationweather.noaa.gov.

b. Expanding the NWS aviation weather safety program and publishing several pamphlets related to adverse weather and flying safety, using statistics similar to those presented in this study. These pamphlets would provide a forum to address flight safety in IMC conditions. They could be distributed during public presentations, open houses and speaking engagements to ensure widest possible dissemination, and they could also be made available via NWS Web sites.

c. Including the statistics presented in this article on all NWS Web sites. This section could be collocated with the aviation and aviation safety segment of every WFO's homepage.

d. Implement of an Aviation Weather Safety Week, similar to the Severe Weather Awareness Week used to inform and educate the public.

e. Making use of the local media, including newspapers, radio and television, to disseminate aviation weather safety information.

f. Consideration, through partnership, of a special segment on The Weather Channel, Public Television (NOVA), and/or the Discovery Channel that highlights the role the NWS plays in supporting aviation and the impact weather has on flight operations for small aircraft.

g. Continued partnership with FAA, AOPA and other organizations to publish articles and track weather-related accident trends.

The aggressive aviation weather safety education campaign that currently is in place at the FAA and AOPA would be greatly enhanced by the addition of the NWS. A multi-agency assault could be the difference in finally making a dent in this long-term aviation problem. In rough numbers, approximately 100 fatal crashes occur every year that are weather-related. About 70 of these 100 crashes are linked to IMC as a causal or contributing factor. An education campaign targeting this primary problem could save up to 100 lives per year or more. Perhaps a more reasonable three-year goal would be a reduction in fatal crashes to 50 per year with 100 lives lost. Those numbers would be one-half of what the pilots are currently experiencing today. A personal goal shared by all, however, is that this information reaches some pilot out there, and helps that pilot decide to fly smart, avoid IMC and heed the advice from the FSS personnel when they say "VFR Flight Not Recommended."

Acknowledgments

I would like to extend a great deal of appreciation to Warren W. Rodie, Meteorologist-in-Charge of the Atlanta CWSU and Lans Rothfusz, Meteorologist-in-Charge of WFO Peachtree City, Georgia, for their support and encouragement of this effort and their insightful technical review of the results. I would also like to acknowledge my Dad's wisdom and flying skills as a WWII fighter pilot and United States Air Force retiree. Whenever I queried him on flying he always said "I must have been a pretty smart pilot - I'm still alive." When facing the threat of adverse weather, a lot can be said about "flying smart," and his smart flying certainly made this article possible.

References

Aircraft Owners and Pilots Association Air Safety Foundation, 1996: Safety Review - General Aviation Weather Accidents - An Analysis & Preventive Strategies. Bruce Landsberg

Aircraft Owners and Pilots Association Air Safety Foundation, 2001: 2000 Nall Report, General Aviation Accident Trends and Factors for 1999. Bruce Landsberg

Federal Aviation Administration Aviation News, April 2002: SIGMETs, AIRMETs Thunderstorms, and the Force, by Jim McElvain. Vol. 41, No. 3, pp 20-21.

National Aviation Weather Initiatives, Feb 1999, National Aviation Weather Program Council, Office of the Federal Coordinator for Meteorology. Rpt No. FCM-P34-199.

National Transportation Safety Board, August, 1974: Special Study of Fatal, Weather-Involved, General Aviation Accidents. Rpt No. NTSB-AAS-74-2, 22 pp.

National Transportation Safety Board, August, 1981: Special Investigation Report - Flight Service Station Weather Briefing Inadequacies. Rpt No. NTSB-SIR-81-3, 14 pp.

US Department of Transportation, Federal Aviation Administration: Administrator's Fact Book, May 1999. APF-100, 48 pp.

US Department of Transportation, Federal Aviation Administration, Feb 2002: Flight Services, 7110. 10 pp.

Web Sites used for database research:

National Transportation Safety Board, Aviation Accident Databases and Synopses:

http://www.ntsb.gov/ntsb/query.asp

The Federal Aviation Administration, National Aviation Safety Data Analysis Center (NASDAC), Aviation Safety Data, Databases, Data Query Tool: https://www.nasdac.faa.gov

The Federal Aviation Administration, FAA News, April 1998:

http://www.faa.gov/apa/safer_skies/Prsrls.htm

National Weather Service, US Natural Hazard Statistics, 61-Year List of Severe Weather Fatalities: http://www.nws.noaa.gov/om/hazstats.shtml

Note: Weather related or weather-involved fatal accidents refer to accidents in which the NTSB determined weather to be a cause or contributing factor in the accident.