Subsonic Aircraft Safety Icing Study

NASA/TM—2008–215107

Subsonic Aircraft Safety Icing Study

Sharon Monica Jones Langley Research Center, Hampton, Virginia Mary S. Reveley Glenn Research Center, Cleveland, Ohio Joni K. Evans ATK, Space Division, Hampton, Virginia Francesca A. Barrientos RIACS, Mountain View, California Click here: Press F1 key (Windows) or Help key (Mac) for help

January 2008

This page is required and contains approved text that cannot be changed.

NASA STI Program ... in Profile Since its founding, NASA has been dedicated CONFERENCE PUBLICATION. Collected to the advancement of aeronautics and space papers from scientific and technical science. The NASA scientific and technical conferences, symposia, seminars, or other information (STI) program plays a key part in meetings sponsored or co-sponsored helping NASA maintain this important role. by NASA.

The NASA STI program operates under the SPECIAL PUBLICATION. Scientific, technical, auspices of the Agency Chief Information Officer. or historical information from NASA It collects, organizes, provides for archiving, and programs, projects, and missions, often disseminates NASA’s STI. The NASA STI concerned with subjects having substantial program provides access to the NASA public interest.

Aeronautics and Space Database and its public interface, the NASA Technical Report Server, TECHNICAL TRANSLATION. English-language thus providing one of the largest collections of translations of foreign scientific and aeronautical and space science STI in the world. technical material pertinent to Results are published in both non-NASA channels NASA’s mission.

and by NASA in the NASA STI Report Series, which includes the following report types: Specialized services also include creating custom thesauri, building customized TECHNICAL PUBLICATION. Reports of databases, and organizing and publishing completed research or a major significant research results.

phase of research that present the results of NASA Programs and include extensive data For more information about the NASA STI or theoretical analysis. Includes compilations program, see the following: of significant scientific and technical data and information deemed to be of continuing Access the NASA STI program home page at reference value. NASA counterpart of peer- http://www.sti.nasa.gov reviewed formal professional papers but has less stringent limitations on manuscript length E-mail your question via the Internet to and extent of graphic presentations. help@sti.nasa.gov TECHNICAL MEMORANDUM. Scientific and Fax your question to the NASA STI Help Desk technical findings that are preliminary or of at (301) 621-0134 specialized interest, e.g., quick release reports, working papers, and bibliographies Phone the NASA STI Help Desk at that contain minimal annotation. Does not (301) 621-0390 contain extensive analysis.

Write to: CONTRACTOR REPORT. Scientific and NASA STI Help Desk technical findings by NASA-sponsored NASA Center for AeroSpace Information contractors and grantees. 7115 Standard Drive Hanover, MD 21076-1320

NASA/TM—2008–215107

Subsonic Aircraft Safety Icing Study

Sharon Monica Jones Langley Research Center, Hampton, Virginia Mary S. Reveley Glenn Research Center, Cleveland, Ohio Joni K. Evans ATK, Space Division, Hampton, Virginia Francesca A. Barrientos RIACS, Mountain View, California National Aeronautics and Space Administration Langley Research Center Hampton, VA 23681

J anuary 2008

Acknowledgments Several persons provided valuable assistance during this study. Tom Ratvasky and Tom Bond provided critical support in locating icing-related studies and reports. Thomas Tibbals, Cathie Zimmerman, Dave Harper , Wayne Black, Andrew Reehorst, and Chuck Ryerson helped us in our last-minute search for military aviation data. Dawn McIntosh, our Aviation Safety Reporting System (ASRS) data mining point of contact, and Jorge Moya from the ASRS group provided assistance with this additional data source. Finally, Sally Viken, our Integrated Resilient Aircraft Control (IRAC) point of contact, supplied the remaining missing information.

Available from: NASA Center for AeroSpace Information 7115 Standard Drive Hanover, MD 21076-1320 (301) 621-0390

TABLE OF CONTENTS

Subsonic Aircraft Safety Icing Study

EXECUTIVE SUMMARY

Of the four types of operations (i.e., Federal Aviation Regulation (FAR) Part 121, Part Scheduled 135, Nonscheduled Part 135 and Part 91) that were examined, Scheduled Part 135 shows the highest percentage of total accidents that involve icing (8.0 percent), the highest percentage of fatal accidents that involve icing (8.2 percent) and the highest percentage of fatalities in icing-related accidents (16.2% percent).

If ramp- and security-related accidents are eliminated from the dataset, 9.5 percent of the total fatal air-carrier (Part 121) accidents are due to icing. This percentage is greater than for general aviation (Part 91) and commuter/air taxi (Part 135), which are 3.7 percent and 8 percent respectively. However, no fatal Part 121 icing-related accidents have occurred since 1994.

Of the six Part 121 icing-related accidents in the dataset, three had turboprop engines, and the others were equipped with turbofan or turbojet engines.

The highest percentage of icing-related reports in the voluntary Aviation Safety Reporting System (ASRS) database occurred in 2005, but the overall number of these events relative to the total is small (less than 1 percent). Of the 663 icing incidents reported, 141 resulted in the crew declaring an emergency.

In Petty and Floyd’s review of airframe icing accidents from 1982-2000, 60 percent of Part 121 accidents were fatal compared with 47 percent of general aviation (Part 91) accidents and 26 percent of Part 135 accidents during the same time period.

Only 0.5 percent of U.S. Army aircraft accidents are icing related, but this number is still of concern due to the Army’s strict regulations that forbid flight into known or severe icing conditions.

A review of several studies by subject-matter experts was summarized into four high-priority icing research areas: (1) improved propulsion system tolerance to weather, (2) development of methods to sense and document actual icing conditions, (3) research to identify realistic ice accumulations to improve aircraft certification and pilot training, and (4) aircraft performance and handling qualities in icing conditions.

Based on the Integrated Resilient Aircraft Control (IRAC) Project goals and objectives, the IRAC project is encouraged to conduct work in all of the high-priority icing research areas that were identified, with the exception of the developing of methods to sense and document actual icing conditions.

Page 6 Subsonic Aircraft Safety Icing Study

1. INTRODUCTION

1.1 Purpose of Study NASA’s Integrated Resilient Aircraft Control (IRAC) Project is one of four projects within the agency’s Aviation Safety Program (AvSafe) in the Aeronautics Research Mission Directorate (ARMD). The IRAC Project, which was redesigned in the first half of 2007, conducts research to advance the state of the art in aircraft control design tools and techniques. The purpose of this research is to ensure flight safety by providing on-board aircraft control resiliency even in the presence of adverse events, including loss of control caused by environmental factors (e.g., wind, icing), and actuator and sensor faults or failures (ref. 1 and 2). The significance of icing in the safety of subsonic transports is inconclusive based on the following statements: “ Note that Icing Challenge is primarily a general aviation problem. Not seen by community as an issue in large transport category aircraft ” (ref. 1).

“ Historical data also indicates icing is primarily considered a general aviation and Part 135 issue ” (ref. 1).

“ General aviation and Part 135 aircraft traditionally fly at lower altitudes and at slower speeds than air transport category aircraft; as a result, they are more likely to encounter icing conditions, including SLD (supercooled liquid droplets) environments ” (ref. 3).

As a result, a “Key Decision Point” (ref. 1) was established for fiscal year 2007 with the following expected outcomes (ref. 1): (1) “Report and document utilizing the most current statistical/prognostic data available from National Transportation Safety Board (NTSB) and Aviation Safety Information Analysis and Sharing (ASIAS), the reports, incidents and accidents associated with icing for subsonic transports.

(2) Report by subject matter experts in icing research on current knowledge of icing effects on control parameters.

(3) Establish future requirements for icing research for subsonic transports including the appropriate alignment (IVHM: identification, IIFD: sensing and IRAC: flight control)” 1.2 Overview of Study Contents The expected outcomes for this study are addressed in sequential order. Outcome 1 is addressed in section 2, which contains statistical analyses of accident and incident data; these analyses have been conducted by NASA researchers for this “Key Decision Point.” Outcome 1 is also addressed in section 3, which contains an examination of icing in other recent statistically based studies. Outcome 2 is the focus of section 4, which is a summary of aviation safety priority lists that have been developed by various subject-matter experts, including the significance of aircraft icing research in these lists. Suggested future requirements for NASA icing research (outcome 3) are presented in section 5. Finally, discussion and the conclusions that have been drawn are provided in section 6.

Page 7 Subsonic Aircraft Safety Icing Study

2. NASA STATISTICAL ANALYSES

As stated previously, the first expected outcome of this study is a report that documents the results of an examination of the most recent statistical/prognostic incident and accident data that is available to determine the significance of icing in subsonic transport accidents and incidents.

This section contains the results of two separate statistical analyses that have been conducted by NASA to address this expected outcome. In the first statistical analysis, publicly available NTSB and Federal Aviation Administration (FAA) accident and incident data were examined. A second statistical analysis was also conducted using the Aviation Safety Reporting System (ASRS) reports as the data source.

2.1 NASA Analysis of NTSB Accident and Incident (A/I) Data A statistical analysis was conducted to look at trends in the number of icing-related events across a 16 year span (1988-2003) within the four areas of flight operation (Part 121, Scheduled Part 135, Nonscheduled Part 135, and Part 91). The word “event” is used here to encompass both accidents and incidents. These types of operations are also known as general aviation (Part 91), commuter/air taxi (Part 135), and air carrier (Part 121). In March 1997, the FAA changed the rules for the classification of major airlines so that air carriers with 10 or more seats, which previously were handled under Part 135 operations, would now be handled under Part 121 operations (ref. 4). The source for accident data is the NTSB Aviation Accident and Incident Data System, while the source for incident data is the FAA’s Accident/Incident Data System. Although both databases contain accident and incident data, the FAA has primary investigative responsibility for incidents, and the NTSB is the authority for accident investigation.

Icing events were selected from the accident data in two ways. First, those accidents that had previously (ref. 5) been assigned the “Ice” category (using the Commercial Aviation Safety Team/International Civil Aviation Organization (CAST/ICAO) Common Taxonomy Team (ref. 6) accident category taxonomy) were separated from the rest of the accidents. These events included both on-ground and in-flight aircraft icing but did not include runway-surface icing.

Some engine icing was included in this group, but only if airframe icing conditions also existed (i.e, rather than just carburetor icing). Also included were events that involved Pitot system icing, windshield icing, or in-flight icing of slush packed in the wheel wells. Second, accidents were selected that included a loss of engine power as a result of carburetor or induction icing, but for which no airframe icing was indicated; this is almost exclusively a general aviation phenomenon.

Selection of icing events from the incident data was accomplished based on the following criteria: (1) if the data record indicated icing as a primary or secondary cause, or (2) if the words “ice,” “icing,” or “freezing rain” were found in the narrative text. All of these records were reviewed; those that involved only ground surface (runway) icing were excluded as were those in which the narrative specified an absence of icing conditions. Events that included a loss of engine power as a result of carburetor icing, induction icing or ice ingestion were not excluded.

A summary of the number of icing events can be found in Table 1. Data for total flight hours per year were obtained from the tables that are published by the NTSB (ref. 7), which are based on data that is supplied by the FAA. Icing contributes to a very small percentage of accidents and incidents, that is, only 3.5 percent of accidents overall and 1.3 percent of incidents. However, fatal icing accidents comprise 4 percent of all fatal accidents, and 4.5 percent of all fatalities occurred in icing-related accidents (aggregate data not shown). Part 135 has the highest Page 8 Subsonic Aircraft Safety Icing Study percentage of icing-related events (as compared with the total number of accidents, both total and fatal) and the largest flight-hour-adjusted rates. By far the largest percentage of icing fatalities (as compared with total fatalities) is within Scheduled Part 135 operations (16.2 percent), and the percentages in other flight categories are remarkably similar to one another (between 4.0 percent and 4.6 percent). Among fatal accidents, the lowest percentage that is attributable to icing is for Part 91 (3.6 percent), which is less than half the rate for Nonscheduled Part 135 operations (7.9 percent).

Table 1. Types of Icing Events by Flight Operation Operation Type of event Part 121 Scheduled Nonscheduled Part 91 Part 135 Part 135 Total flight hours 232,868,640 25,050,928 46,350,000 416,319,000 Total fatal accidents 60 49 278 4599 Fatal accidents with 4 (6.7%) 4 (8.2%) 22 (7.9%) 165 (3.6%) icing conditions Total fatalities 2151 328 664 8724 Fatalities with icing 99 (4.6%) 53 (16.2%) 27 (4.1%) 350 (4.0%) conditions Total accidents 600 213 1070 23349 Accidents with icing 6 (1.0%) 17 (8.0%) 58 (5.4%) 321 (1.4%) conditions Icing accidents per 0.0258 0.679 1.25 0.771 million flight hours Accidents with 3 (0.3%) 467 (2.0%) carburetor icing only Carburetor icing accidents per million 0.065 1.122 flight hours Total incidents 7497 2218 2082 29167 Incidents with icing or carburetor icing 50 (0.7%) 32 (1.4%) 48 (2.3%) 390 (1.3%) conditions Icing or carburetor icing incidents per million 0.215 1.277 1.036 0.937 flight hours Green suggests in his report on icing (ref. 8), that a relationship may exist between icing events and aircraft size. Many of the same aircraft are flown under Part 91 and Part 135, so one might expect the icing rates to be more similar in those groups. Perhaps the pilots who are operating under Part 91 are more able to delay their flights to avoid icing conditions, or perhaps the pilots who are operating under Part 135 overestimate either their own ability or the aircraft’s ability to handle the icing conditions.

Page 9 Subsonic Aircraft Safety Icing Study In prior NASA examinations of aircraft accident rates (ref. 9), ramp-related events were not included because these events were perceived as occupational-health-related events. Security- related events were also excluded because they fall under the domain of the Department of Homeland Security. Therefore, we decided to look within this same data subset at the significance of icing events. Specifically, we eliminated all security-related events (109 total events) and those for which the predominant issue was a ground-handling or ramp error (e.g.

ground-crew errors, pushback/tow issues, inadequate preflight inspection). Ninety-five percent of these ground-handling accidents (274 total events) took place while the aircraft was on the ground or during takeoff; the remainder involved mostly separated doors or engine cowlings.

Part 121 is the only operation that is impacted by this change (see Table 2). The percentage of fatal accidents that are attributable to icing jumps from 6.7 percent to 9.5 percent (now the highest of the four percentages), and the percentage of total fatalities jumps from 4.6 percent to 6.2 percent (now second highest among the four groups).

Table 2. Types of Icing Events by Flight Operation (Without ramp and security events) Operation Type of event Part 121 Schedule Part Nonscheduled Part 91 135 Part 135 Total flight hours 232,868,640 25,050,928 46,350,000 416,319,000 Total fatal accidents 42 48 275 4511 Fatal accidents with 4 (9.5%) 4 (8.3%) 22 (8.0%) 165 (3.7%) icing conditions Total fatalities 1603 327 660 8628 Fatalities with icing 99 (6.2%) 53 (16.2%) 27 (4.1%) 350 (4.1%) conditions Total accidents 534 203 1057 23055 Accidents with icing 6 (1.1%) 17 (8.4%) 58 (5.5%) 321 (1.4%) conditions The icing-related events, shown as a percentage of total fatal accidents, total fatalities, and total accidents in Tables 1 and 2 are displayed in a side-by-side graphical format in the next three figures. For each of the four operations, icing-related accidents and incidents are displayed both with and without ramp- and security-related events. Again, note that the most visually significant difference occurs in the percentage of fatal accidents with icing conditions for Part 121 operations (see Figure 1).

Page 10 Subsonic Aircraft Safety Icing Study Part 91 Part 135 NS Part 135 S Operations Part 121 0 1 2 3 4 5 6 7 8 9 10 Without ramp and security Percent of fatal accidents related to icing With ramp and security Figure 1. Fatal accidents with icing conditions.

Part 91 Part 135 NS Part 135 S Operations Part 121 0 2 4 6 8 10 12 14 16 18 Without ramp and security Percent of total fatalities related to icing With ramp and security Figure 2. Total fatalities with icing conditions.

Part 91 Part 135 NS Part 135 S Operations Part 121 0 1 2 3 4 5 6 7 8 9 Without ramp and security Percent of total accidents related to icing With ramp and security Figure 3. Total accidents with icing conditions.

Page 11 Subsonic Aircraft Safety Icing Study Since 1988, no more than five icing-related events have occurred per year during Part 121 flights, and the overwhelming majority of those were incidents (see Figure 4). In 35 of the 56 events, no damage to the aircraft was noted and no injuries occurred. However, in four of the six accidents, the aircraft was destroyed, and fatal injuries occurred. A total of 99 persons died in icing-related accidents, but no fatal icing-related accidents have occurred since 1994. A downward trend is evident in the flight-hour-adjusted rates for icing events (see Figure 5).

Number of events 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Accidents with carburetor-icing Year of event Accidents without carburetor-icing Incidents Figure 4. Number of Part 121 icing events per year.

0.5 0.4 0.3 0.2 hours 0.1 Events per million flight 0.0 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 yearly rate Year Trend line Figure 5. Rate of Part 121 icing events (per million flight hours).

Among Scheduled Part 135 operations, more than 70 percent (35 of 49) of the icing events that are included in this analysis (see Figure 6) occurred prior to 1995, but just over half of the icing accidents (9 out of 17) have occurred since 1994. The flight-hour-adjusted rates for icing events may be increasing slightly (see Figure 7), or this may be an artifact of the major decrease in flight hours that is noted for Scheduled Part 135 operations since 1997. Of the four types of flight operations that were examined, Scheduled Part 135 operations show the highest Page 12 Subsonic Aircraft Safety Icing Study percentage of total accidents that involve icing (8.0 percent), the highest percentage of fatal accidents that involve icing (8.2 percent), and the highest percentage of fatalities in icing accidents (16.2 percent).

Only one of the four fatal accidents occurred prior to 1990; one-third of the accidents since 1994 (3 of 9) included a fatality. Five of the nine most recent accidents (two of which were fatal) had inadequate ice removal prior to takeoff, which led to an overrun of the runway on an aborted takeoff in one event and a loss of control during the takeoff climb in the remaining four accidents. No injuries occurred in 9 of the 17 accidents, and no damage to the aircraft occurred in 28 of the 32 incidents.

Number of events 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Accidents with carburetor icing Year of Event Accidents with no carburetor icing Incidents Figure 6. Number of Scheduled Part 135 icing events per year.

hours Events per million flight 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Year yearly rate Trend line Figure 7. Rate of Scheduled Part 135 icing events (per million flight hours).

Among Nonscheduled Part 135 flights, slightly more icing-related accidents than incidents occurred in most years (see Figure 8). Of the four flight groups examined, Nonscheduled Part Page 13 Subsonic Aircraft Safety Icing Study 135-NS had the highest percentage of icing-related incidents (2.3 percent) and the second highest percentage of both icing-related accidents (5.7 percent) and icing-related fatal accidents (7.9 percent). Only 27 persons died in Nonscheduled Part 135 icing-related accidents. No injuries occurred in 21 of the 61 accidents, and no damage to the aircraft occurred in 22 of the 48 incidents. A slight downward trend is evident in the flight-hour-adjusted rates for icing-related Nonscheduled Part 135 events (see Figure 9).

Number of events 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Accidents with carburetor icing Year of Event Accidents without carburetor icing Incidents Figure 8. Number of Nonscheduled Part 135 icing events per year.

5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 flight hours 1.0 Events per million 0.5 0.0 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Yearly rate Year Trend line Figure 9. Rate of Nonscheduled Part 135 icing events (per million flight hours).

In the eight years prior to and including 1995, the average number of Part 91 icing-related events was 98. In the next eight years (1996-2003), the average dropped to 49 (see Figure 10).

The largest drop was among incidents (from 36 to 12), and the smallest drop was among the icing accidents that did not involve carburetor icing (from 24 to 16). A significant decrease was noted in the flight-hour-adjusted rates for icing-related events (see Figure 11). Only 3.6 percent of fatal Part 91 accidents were related to icing; this was the lowest percentage among the four Page 14 Subsonic Aircraft Safety Icing Study flight categories that were examined. Although 350 persons died in icing-related accidents, this represents only 4 percent of all Part 91 deaths in this time period (again, the lowest of the four percentages). No injuries occurred in nearly half of the icing-related accidents (390 of 788), and no aircraft damage occurred in 229 of the 390 incidents (59 percent).

Number of events 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Accidents with carburetor icing Year of Event Accidents without carburetor icing Incidents Figure 10. Number of Part 91 icing events per year.

hours Events per million flight 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Yearly rate Year Trend line Figure 11. Rate of Part 91 icing events (per million flight hours).

Each accident report specifies the type of engine that was installed on the aircraft. Because the IRAC project includes engine research, an examination was conducted of all events in the dataset based on engine type. The possible engine types (ref. 10-14) are: • Reciprocating: Also known as a piston engine, the reciprocating engine is a heat engine that uses one or more pistons to convert pressure into a rotating motion.

• Turbojet: Turbojet engines are inefficient and noisy engines that are used by mostly older aircraft and consist of an air inlet, air compressor, combustion chamber, gas Page 15 Subsonic Aircraft Safety Icing Study turbine, and nozzle. These engines are the simplest and oldest kind of general-purpose jet engine.

• Turboprop: Turboprop engines are similar to turbojet engines except that additional stages are included in the turbine to recover more power to turn the propeller. Turboprop engines are generally used on small or slow subsonic aircraft because they are highly efficient at modest flight speeds (i.e., below 450 mph).

• Turbofan: A turbofan engine is similar to a turbojet engine and essentially consists of a ducted fan with a smaller diameter turbojet engine that is mounted behind it to power the fan. Turbofans are used by most large commercial aircraft because of their fuel efficiency and quiet operation.

• Turboshaft: In principle, a turboshaft engine is similar to a turbojet engine, except that it features additional turbine expansion to extract heat energy from the exhaust and convert it to output shaft power. The name turboshaft is most commonly applied to engines that power ships, helicopters, tanks, locomotives, and hovercraft.

For a small number of accidents, the engine type is unknown. Figures 12 through 15 display the percentage of total accidents each year for which the aircraft engine was of a particular type.

Engine type, percent 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Year Reciprocating Turbofan Turboprop Turbojet Unknown Figure 12. Part 121 accidents by engine type.

Part 121 aircraft (see Figure 12) use primarily turbofan engines (62 percent overall), although the number of turboprops engines (19 percent overall) increased in 1997 (when Part 121 was expanded to include flights that previously would have been considered Part 135).

Scheduled Part 135 aircraft (see Figure 13) use either reciprocating engines (52 percent) or turboprop engines (44 percent) almost exclusively.

Nonscheduled Part 135 aircraft (see Figure 14) primarily have used reciprocating engines (79 percent), with turboprop engines (15 percent) becoming more popular in recent years. Part 91 aircraft (see Figure 15) use reciprocating engines almost exclusively (96 percent overall).

Page 16 Subsonic Aircraft Safety Icing Study type, percent Engine 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Year Reciprocating Turboprop Turboshaft Unknown Figure 13. Scheduled Part 135 accidents by engine type.

type, percent Engine 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Year Reciprocating Turbofan Turboprop Turbojet Turboshaft Unknown Figure 14. Nonscheduled Part 135 accidents by engine type.

Engine type, percent 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Year Reciprocating Turbofan Turboprop Turbojet Unknown Figure 15 - Part 91 accidents by engine type.

Page 17 Subsonic Aircraft Safety Icing Study Figures 16 through 19 present the same information as shown in Figures 12-15, for only those accidents that involved fatalities.

For Part 121 operations (see Figure 16), more than half (55 percent) of the fatal accidents occurred in aircraft with turbofan engines, but this is still less than the percentage (62 percent) of aircraft with turbofan engines that were involved in both fatal and nonfatal accidents. The percentage of turboprop engines was also slightly lower (13 percent rather than 19 percent), while relatively more fatal accidents occurred with reciprocating, turbojet, and unknown engine types. In particular, 13 percent of the fatal accidents have unknown engine types. In 2002, no Part 121 accidents included a fatality.

Engine type, percent 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Year Reciprocating Turbofan Turboprop Turbojet Unknown Figure 16. Part 121 fatal accidents by engine type.

For Scheduled Part 135 operations (see Figure 17), the percentages for fatal accidents are similar to those for all accidents with respect to engine type. The fatal accidents occurred predominantly on aircraft with reciprocating or turboprop engines (96 percent), with slightly more fatal accidents on aircraft with turboprop engines (51 percent). No fatal accidents occurred in 1998 and 2002.

Engine type, percent 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Year Reciprocating Turboprop Turboshaft Unknown Figure 17. Scheduled Part 135 fatal accidents by engine type.

Page 18 Subsonic Aircraft Safety Icing Study For both Nonscheduled Part 135 (see Figure 18) and Part 91 operations (see Figure 19), the distribution of engine type in fatal accidents is remarkably similar to that for fatal/non-fatal accidents combined. The percentages for fatal accidents in turboprop aircraft were slightly higher than for the combined group (16.5 percent versus 15 percent for Nonscheduled Part 135; 4 percent versus 2 percent for Part 91 respectively).

Engine type, percent 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Year Reciprocating Turbofan Turboprop Turbojet Unknown Figure 18. Nonscheduled Part 135 fatal accidents by engine type.

Engine type, percent 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Year Reciprocating Turbofan Turboprop Turbojet Unknown Figure 19. Part 91 fatal accidents by engine type.

Recall that when ramp- and security-related accidents were eliminated from the dataset (see Table 2), 9.5 percent of the total fatal air carrier (Part 121) accidents were caused by icing. This percentage is greater than that for general aviation (Part 91) or commuter/air taxi (Part 135), which were 3.7 percent and 8 percent respectively. The aviation community generally attributes the large number of icing accidents in Part 121 operations to aircraft with turboprop engines and because a reduction in the use of turboprop engines has been noted for Part 121 operations during the past 10 years, aviation experts assume that icing is no longer a significant problem for Part 121 operations. This assumption is disputed for two reasons. First, although the majority of Part 121 operators indeed use jet aircraft, turboprop and piston engines still Page 19 Subsonic Aircraft Safety Icing Study comprise a significant portion of the this operating category. For example, in 2004, 59 percent of the passenger aircraft used by regional airlines had jet engines, but turboprop and piston (i.e.

reciprocating) engines accounted for an additional 41 percent of the aircraft in operation (ref.

15). Second, although two of the four fatal Part 121 icing accidents that occurred during the period of study were in turboprops, the other fatal accidents involved aircraft that were equipped with turbofan and turbojet engines (see Table 3).

Table 3. Part 121 Fatal Icing Accidents Event date Location Airline Engine type 3/15/89 West Lafayette, IN Mid Pacific Turboprop 2/17/91 Cleveland, OH Ryan International Turbojet 3/22/92 Flushing, NY US Air Turbofan 10/31/94 Roselawn, IN American Eagle Turboprop Page 20 Subsonic Aircraft Safety Icing Study 2.2 NASA Analysis of ASRS Data Database The data in this section of the report were gathered from the ASRS database. As of the writing of this report, the database archives incident reports that were submitted between January 1988 and February 2007. The total number of reports in the database from this time period is a little less than 137,000.

The ASRS is a voluntary reporting system, and reports can be submitted by anyone who is involved in an incident. The majority of the reports appear to be filed by the flight crew. All of the reports that are accepted into the ASRS database have been reviewed, screened, and coded by an analyst. The proportion of reported incidents that are rejected is unknown.

The ASRS database is accessible online at http://asrs.arc.nasa.gov.

Methodology The data set used in this analysis consists of 663 reports that are highly likely to have icing as a significant factor. To retrieve these reports, we used the ASRS online search tool. The query had two parts: 1) Primary problem: weather 2) Free text search string: "ic% OR rime OR rhime OR de-ice OR deice OR sleet OR frost" The Primary problem field has16 possible category selections which are mutually exclusive. The free text search is performed across three fields: narrative, synopsis, and keywords. The narrative is a part of the raw report, and is retained just as it was submitted by the reporter. The synopsis and keywords are added by an analyst later.

While developing the search string, we checked possible alternate spellings of sleet —“slete,” “sleat,” and “sle%t” - and found no occurrences of these alternate spellings. In his report, Green's (ref. 8) search string included the term glaze . We did not include this term because each report that contained the word glaze in reference to an environmental factor also contained the word ice . (Five such reports met this criterion.) In the single report in which the word glaze appeared without the term ice, it referred to the condition of someone's eyes.

Although most or all of the reports in the data set appear to involve engine or wing icing, the reports have not been reviewed exhaustively. From our experience with the database, we know that weather-related ice problems also can refer to ice on the runway as well as the aircraft.

Icing-Related Incidents Versus All Weather-Related Incidents We compared the number of icing-related incidents with both the total number of incidents and the total number of weather-related incidents that were reported over the last 10 years (see Table 4). We use the term weather incidents to mean the incidents that are retrieved by searching for items that have weather as the primary cause.

In all years, the number of icing-related incidents is very small relative to the total number of incidents. However, icing is a significant fraction of all weather-related incidents. Although the last three years show fewer weather incidents, note that fewer incidents overall were recorded for the same time period.

Page 21 Subsonic Aircraft Safety Icing Study Table 4. Icing Versus Weather Incidents in ASRS Number of Percentage of Percentage of Number of icing- incidents Total incidents weather- related related to related to related Number of Year incidents icing incidents weather incidents 1997 52 0.65 8018 16 332 1998 34 0.40 8428 12 276 1999 29 0.29 9866 12 245 2000 38 0.48 7998 16 232 2001 21 0.24 8612 11 197 2002 26 0.36 7198 13 201 2003 32 0.39 8143 12 266 2004 27 0.44 6200 17 161 2005 25 0.71 3524 23 109 2006 11 0.21 5196 9 188 Incidents by FAR Operation Incident reporters are responsible for documenting the FAR under which they were operating at the time of the incident. The applicable regulations for this field in the ASRS database are Parts 91, 119, 121, 125, 129, and 135; an other choice is provided for all others. In the icing data set, none of the aircraft were reported as operating under FAR Part 125.

Table 5 shows the number of icing-related incidents for selected FAR operations. The total number of incidents and the total number of weather-related incidents are included in this table for comparison purposes.

Table 5. Subset of ASRS Weather-Related Incidents by Select FAR Operation Part Part Part Part Number of weather- Number of total Year 91 119 121 135 related incidents incidents 1997 15 0 29 7 332 8018 1998 15 0 13 4 276 8428 1999 15 1 9 2 245 9866 2000 12 0 24 2 232 7998 2001 10 0 6 2 197 8612 2002 8 0 15 3 201 7198 2003 20 0 12 0 266 8143 2004 14 0 12 1 161 6200 2005 13 0 8 3 109 3524 2006 7 0 3 1 188 5196 Incidents by Flight Phase For the Flight phase field, incident reporters can select multiple flight phases. A visual scan of the reports shows that nearly all of the reports specify multiple flight phases. This field allows primary categories and subcategories. Counts for the primary categories are given in Table 6.

Page 22 Subsonic Aircraft Safety Icing Study Table 6. ASRS Icing-Related Incidents by Flight Phase Flight Phase Number of Incidents Climb-out 98 Cruise 325 Descent 193 Ground 106 Landing 133 Incidents by Resolutory Actions Resolutory actions are the actions taken by the flight crew or controller or other circumstances that occurred to resolve an anomaly during the flight. Of the 663 icing-related incidents, 141 resulted in the crew declaring an emergency, and 16 resulted in an emergency landing. In this analysis, we provide counts of the categories for those actions that included crew action but not for other types of actions (Table 7). These actions are not mutually exclusive; incident reporters could select more than one entry for this field.

Table 7. ASRS Icing-Related Incidents with Resolutory Actions Resolutory Action Number of Icing-Related Incidents Declared emergency 141 Landed in emergency 16 Landed as precaution 7 Rejected takeoff 9 Executed go around 2 Incidents That Involved Carburetor Icing Carburetor icing incidents were found by searching for the string “carb” within the narrative section of the report (Table 8). Of the 663 icing reports, this search returned 9 incidents that mentioned carburetor icing.

Table 8. ASRS Incidents With Carburetor Icing Year Count 1990 1 1994 1 1996 1 1998 2 2000 1 2001 1 2003 2 Incidents by Anomaly Classification Incident reporters can select multiple entries for the A nomaly field. Of the 663 icing-related incidents, 172 reports contained A ircraft equipment problem as one of the anomalies. This anomaly can be further subcategorized into “critical” or “less severe”; 121 incidents were classified as critical, and 51 incidents were classified as less severe.

Page 23 Subsonic Aircraft Safety Icing Study Although the anomalies cab be subcategorized, only the primary anomalies are listed shown Table 9.

Table 9. ASRS Icing-Related Incidents by Anomaly Field Number of icing-related Anomaly incidents Equipment problem 172 Airspace violation 5 Altitude deviation 165 Cabin event 8 Conflict (airborne or ground) 28 Excursions 156 Ground encounters 12 Incursion 7 In-flight encounter 532 Nonadherence 267 Other 415 Incidents by Engine Type Table 10 shows the 663 incidents broken down by engine type.

Table 10: ASRS Icing-Related Incidents by Engine Type Engine Number of icing-related Type incidents reciprocating 223 turbojet 208 turboprop 87 unknown 46 Page 24 Subsonic Aircraft Safety Icing Study

3. REVIEW OF PRIOR AVIATION STATISTICAL STUDIES

A number of previous studies have been conducted that examine historical aviation data to determine the impact of icing on accidents and incidents. One such study was funded by NASA (ref. 8). Two others were led by other agencies, such as the FAA; however NASA employees were members of the analysis teams. The remaining study was conducted by the NTSB without any known NASA interaction. Although many of the studies used the same databases for their analyses, considerable differences exist in their results due to the variations in the respective study assumptions and constraints.

3.1 Green Study of Inflight Icing At the 2007 American Institute of Aeronautics and Astronautics conference in Reno, NV, Steven Green presented the results of his study of inflight icing accidents and incidents in the United States (ref. 8). Three databases were used for the study: (1) the NTSB Accident Database and Synopses, (2) the FAA Accident/Incident Data System (AIDS), and (3) the NASA Aviation Safety Reporting System (ASRS). Green searched for the terms “icing,” “freezing,” “rime, “ “glaze,” “sleet” and “frost” in all three databases and narrowed the set of events to those that occurred between January 1, 1978, and December 31, 2002. A variety of parameters were examined, including: weather (i.e., METAR, also known as “aviation routine weather reports”), aircraft characteristics from Type Certificate Data Sheets or the FAA’s Aircraft Registry, level of pilot experience, and operating rules (e.g., FAR Parts 91 and 135).

Some notable observations from this data analysis are: • Freezing rain was involved in 33 percent of the total events that involved precipitation, • The ASRS reports indicated that although pilots attempt to minimize the amount of ice accretion by using rapid descent, their efforts are often hampered by Air Traffic Control requirements.

3.2 Petty/Floyd Review of U.S. Airframe-Icing Accidents A review of U.S. airframe-icing accidents was conducted by the NTSB (ref. 3) and the overall objective was to update a study of aircraft icing hazards that focused on accidents that occurred in the 1970’s and the 1980’s (ref. 18) . Petty and Floyd (ref. 3) used NTSB accident data from 1982 - 2000 and examined three types of aircraft operations: Part 91, Part 121, and Part 135.

They also looked at the month, phase of flight, location (U.S. state) in which the icing accidents occurred, and the highest certification of the pilot in command at the time of the accident. The majority of airframe accidents occurred in Part 91 (80.6 percent), followed by Part 135 (17.6 percent), and Part 121 (1.7 percent). However, during the time period that was studied, 60 percent of Part 121 accidents were fatal compared with 47 percent of Part 91 and 26 percent of Part 135 airframe-icing accidents. Only 10 fatal Part 121 accidents occurred in the period of observation. The number of Part 91 accidents dropped from a high of 49 at the beginning of the review period to 17 accidents in the year 2000, which correlated with the overall decrease in Part 91 accidents.

Petty and Floyd (ref. 3) stated in their report that GA and Part 135 are more likely to encounter icing conditions because they “fly at lower altitudes and slower speeds than air transport aircraft,” but they did not provide data in their report to support this assumption. For this reason Page 25 Subsonic Aircraft Safety Icing Study the engine type was also examined in the IRAC Accident/Incident Study, and the results were presented and discussed in Section 2.

3.3 Army Aircraft Icing Report The Army Safety Center searched its database of 54,081 Army aircraft accidents and incidents that occurred from fiscal year 1985 to 1999 for events that were coded as “aircraft icing” or “significant weather,” such as sleet or freezing rain (ref. 17). A total of 255 events resulted from this query; however, based on the narrative summaries only 172 of these accidents were truly icing related. The remaining non-genuine icing accidents were not removed from the total icing accident data set because icing could not be definitively ruled out as a factor in the accidents.

Of the 255 icing events, 85 involved fixed-wing aircraft and the remaining two-thirds involved helicopters. If we base our assumptions purely on the data, we can conclude that icing is not a high-frequency safety problem for the Army because it accounts for only 0.5 percent of the total accidents during the study period. However, these relatively small numbers are still of concern due to the Army’s strict regulations in regard to aircraft and icing conditions: “… aircraft will not be flown into known or forecast severe icing conditions” (Aviation Flight Regulations, AR 95).

The regulation also states that, if a flight is to be made into known or forecast moderate icing conditions, the aircraft must be equipped with adequate deicing or anti-icing equipment.

To mitigate the risks of icing-related events, better in-flight icing detection and preflight deicing capabilities were recommended in the report.

3.4 NASDAC Review of NTSB Weather-Related Accidents The National Safety Data Analysis Center (NASDAC) is the former name of the current ASIAS (Aviation Safety Information Analysis and Sharing) system. The purpose of ASIAS is to enable the exchange of multiple types of aviation safety data and information. ASIAS staff members examined NTSB aviation accident and incident data from 1994 to 2003 to identify any events in which weather was a causal or contributing factor (ref. 18).The ASIAS staff also looked at the relationship between the types of weather that were involved in the events and other parameters, such as type of operation and phase of flight.

Weather was cited as a contributing or causal factor in 21.3 percent of the 19,562 accidents that occurred in the study period. Out of the 4,159 weather-related accidents, icing accounted for 7.0 percent of the total accidents. The significance of icing differs based on the operation type. For example, only 0.8 percent (i.e. one accident) of the 124 weather-related part 121 accidents involved icing, but icing was a factor in 10 percent and 7.1 percent of the Part 135 and Part 91 accidents, respectively.

3.5 Summary of Statistical Study Parameters One of the reasons that a comparison of the results from various studies can be difficult is the differences in the study constraints and assumptions (see Table 11). Although many of these studies used the same accident and incident databases, often a subset of the database (e.g., Page 26 Subsonic Aircraft Safety Icing Study specific event years, FAR operations) is used for analysis. For example, we examined the differences between the results of the ASIAS weather study (see section 3.4) and those from the NASA analysis of the recent NTSB accident and incident data (see section 2.1). The results of these studies differ for two primary reasons. First, the ASIAS study covers a subset (years 1994 – 2003) of the total accidents that were examined in the NASA study (years 1988 – 2003).

In the NASA study data, six Part 121 icing-related accidents were identified (Two with loss of control in flight, two with loss of control during takeoff, one with loss of control during approach/landing and one with extensive engine damage). However, even if the data set is reduced to the same period as that used in the ASIAS study, three Part 121 accidents were identified in the NASA study, which still differs from the ASIAS results. The reason for this discrepancy is that the ASIAS analysis relied solely on NTSB accident-classification codes or primary accident cause to identify accident categories. Thus, according to the ASIAS analysis, the accident on October 31, 1994 in Roselawn, IN would be the only Part 121 event that is icing related. However, according to the NASA study, the Part 121 accidents that occurred on March 14, 1997 (ice ingestion in both engines), and March 19, 2001 (failure to maintain airspeed during an encounter with severe icing conditions), also are icing related.

Table 11. Summary of Statistical Study Parameters Total number Author/ study Operation Country of Database(s) Event years of icing title type operations events NTSB Accident, Part 91 645 ( A/I) Green FAA AIDS, United States 1978 – 2002 Part 135 299 (ASRS) NASA ASRS GA Petty/ Part 135 NTSB Accident United States 1982 – 2000 583 Floyd Part 121 Army Data Worldwide (U.S.

Part 91 Army Aircraft A/I 1985 – 1999 255 Army) Center Part 91 Part 135S NSTB Accident United States 1994 – 2003 ASIAS Part 135NS Part 121 Part 91 NASA Part 135S NTSB Accident, 872 (NTSB) Analysis of United States 1988 – 2003 Part 135NS FAA AIDS 520 (AIDS) NTSB Data Part 121 NASA Part 91 Part 121 NASA ASRS United States 1988 – 2007 663 Analysis of Part 135 ASRS Data Page 27 Subsonic Aircraft Safety Icing Study

4. REVIEW OF AVIATION SAFETY PRIORITY LISTS

In recent years, subject-matter experts have conducted a number of studies that included recommendations for improving the safety of the air transportation system. This section contains a review of these safety priority lists with special emphasis on the significance of icing in improving aviation safety.

4.1 Decadal Study of Civil Aeronautics The Aeronautics Space and Engineering Board within the National Research Council used a modified quality function deployment (QFD) process to identify and rank research and technology challenges. This study (ref. 19) which was sponsored by NASA, began with a September 2005 kick-off meeting in Herndon, VA. A steering committee, along with several supporting panels, conducted a series of additional meetings, which culminated in a June 2006 report in which a total of 51 high-priority challenges were identified. Two of these high-priority challenges, A6 (Aerodynamics robust to atmospheric disturbances and adverse weather conditions, including icing) and D6 (Improved onboard weather systems and tools), are related to icing.

Challenge A6 is based on the need for additional research to develop techniques to monitor and mitigate adverse weather conditions, such as icing, wind sheer, and free- stream turbulence.

The expected results of this research include the development of three-dimensional icing- prediction tools and systems with better measurement and mitigation of upstream environmental conditions. In particular, better icing-prediction methods would help reduce the high cost of aircraft and helicopter icing certification. The recommendation was made that NASA conduct research in this area because the agency’s icing wind tunnels and research aircraft would complement the infrastructure that is found in industry and academia. This challenge was also rated as being highly relevant to NASA’s Mission.

Improved on-board weather systems and tools (challenge D6) are necessary to cost effectively integrate real-time weather information into four dimensional integrated control of flight. The goal is for aircraft to share on-board sensor data about wind, icing conditions, lightning, and turbulence via data links with ATC and other aircraft. This information would be used in the air traffic management (ATM) system to reduce delays, especially in high-traffic areas, and to prevent weather-related accidents. The rationale for NASA conducting research in this area was because the agency has “an outstanding research facility for icing tests and evaluation and the infrastructure to develop and test weather-related tools.” The report also stated that this research is central to NASA’s Safety and Capacity missions.

A third research and technology challenge that is related to icing, B13 (Improved propulsion system tolerance to weather, inlet distortion, wake ingestion, bird strike, and foreign-object damage) was also identified. One of the proposed outcomes of this challenge is the development of physics-based models that can more accurately predict the effects of adverse weather (e.g., rain, hail, ice) on fans, compressors, and combustor stability. Although this challenge received high scores as a national priority, it did not make the overall high-priority list because NASA has few facilities, such as icing tunnels, that are relevant to this challenge.

Page 28 Subsonic Aircraft Safety Icing Study 4.2 Review of NASA’s Aerospace Technology Enterprise The National Research Council was asked by NASA and the Office of Management and Budget (OMB) to review NASA’s former Aerospace Technology Enterprise (currently known as the Aerospace Research Mission Directorate.) The review began in early 2003 and, at the time, the Aerospace Technology Enterprise comprised three programs: Vehicle Systems, Airspace Systems, and Aviation Safety. At the time of the review, the Aviation Safety Program contained an aircraft icing subproject, which the committee deemed “the best technical work in the Aviation Safety Program” and “a unique national asset that is vital to the air transportation community, both civil and military.” In the final report, which was published in 2004, the committee recommended that NASA continue the research to understand and mitigate aircraft icing. Some specific future research areas that were identified in the report include additional investigations to determine the effects of icing on aircraft aerodynamics, the development of methods to sense and document actual icing conditions, and smart icing systems. The report also advocated increased research in the areas of anti-icing fluids and the assessment of holdover times.

4.3 CAST/JIMDAT Safety Implementation Plans The Commercial Aviation Safety Team (CAST) was formally established on June 23, 1998. Its mission is to provide government and industry leadership to develop and focus implementation of an integrated, data-driven strategy to improve commercial aviation safety. The members of CAST are a variety of stakeholders, including representatives from government, industry, pilot groups, air traffic controllers, and others. The Joint Implementation Data Analysis Team (JIMDAT) was formed by CAST to develop methods for prioritizing safety-enhancement implementation plans. JIMDAT accomplishes this mandate by: • Estimating the potential risk decrease.

• Tracking those projects that are selected for implementation.

• Evaluating project outcomes to determine whether the estimated risk was lessened.

• Assessing any improvements that are identified outside of CAST.

• Identifying other areas of interest beyond the six prioritized FAA Safer Skies Agenda accident categories.

In its original list of recommendations to CAST, JIMDAT proposed a Safety Enhancement (SE) to release the risk of loss of control caused by icing: SE 39: Criteria for Flight in Icing Conditions for New Airplane Designs To reduce fatal accidents due to loss of control, recommend and support the development of amended icing certification criteria, for new airplane designs not equipped with evaporative (i.e. hot wing) systems, that include performance and handling qualities requirements for the following: - Residual ice - Intercycle ice - Delayed anti-icing/deicing system activation - Deicing/anti-icing system malfunction JIMDAT considers SE 39 to be an ongoing enhancement that has not been completed. Over the years, JIMDAT has made modifications to its original list of safety enhancements and has Page 29 Subsonic Aircraft Safety Icing Study identified additional SE’s that are either appropriate for long-term research or address the “remaining risk” in the air transportation system. One of the research SE’s is related to icing: SE 119: Loss of Control – Icing (research) For aircraft that incorporate non-evaporative ice protection systems, develop systems that sense the presence of ice accretion on the aircraft, automatically activate and manage the ice protection systems, and provide the pilot with feedback including the effect on measured aircraft performance, stability, and control.

- Provides annunciation that alerts the crew to respond appropriately to the icing hazard.

- Ground- and aircraft-based means of detection of meteorological icing conditions.

- Define the effects of all ice accretions, with particular emphasis on the roll effect due to ice-contaminated wings.

- Understand the effects of super-cooled large droplets (SLD).

The three icing-related “remaining risk” SE’s that JIMDAT has recommended for approval by the CAST are as follows: SE 133: Turboprop Aircraft Ice Detection Systems For all turboprop aircraft with non-evaporative ice protection systems and non- powered flight controls used in Part 121 operations, manufacturers should: (1) For new type designs, adapt and implement systems that automatically detect ice, measure the rate of ice accretion, and provide annunciation to the flight crew and (2) For current turboprop production aircraft and existing type design, conduct a study to determine the feasibility of installing systems that automatically detect ice, measure the rate of ice accretion, and provide annunciation to the flight crew SE 134R1: Aircraft Design – Avionics Avionic equipment manufacturers and aircraft manufacturers should develop and install smart pitch guidance systems on new type designs to prevent over rotation in conjunction with a low energy state or aerodynamic degradation due to the presence of ice on critical flight surfaces. The smart pitch avionics system should provide to the flight crew appropriate flight guidance information, the current aircraft energy state and performance margins with respect to stall speed and maximum angle of attack for all icing conditions for which an aircraft has been certified.

SE 136: Training – Engine Surge Recovery To prevent fatal accidents resulting from an engine surge caused by ice ingestions, airlines should provide adequate training for flight crews to ensure appropriate responses to this event. This training should include engine-out identification, engine surge recovery procedures, and associated aircraft recovery in all the varying combinations. Key to this training is that it be accomplished prior to the pilot being assigned to the line or introduced to new equipment.

Page 30 Subsonic Aircraft Safety Icing Study To understand the relative significance of these icing enhancements, Figure 20 contains the potential impact of all of the recommended enhancements on reducing risk. Note that only 6 out of the 19 total enhancements have higher risk-reduction scores than the 3 icing enhancements.

Four of the higher scoring enhancements are related to cargo and the other two are related to maintenance/systems.

% Reduction of Fatality Risk That Could Be Mitigated by Individual RR SEs (1987-2000 Part 121 Hull Loss and Fatal Accidents) 2.50 2.00 1.50 1.00 0.50 of Part 121 Fatality Risk Eliminated % 0.00 163) See (SE162) (SE130) Advanced ( 101) Arc (SE-175) Regulation and avoid 121 TCAS Supression (SE159) (SE127) Fire (SE164) Part Maintenance (SE policies and Processing Procedures – SE (SE129) Legal Safety Culture (SE165) TCAS Fault Breakers (SE131) Cargo Design Inflight (SE133) Aircraft (SE-169) Policy Design – OEM (SE-172) Policy (SE121) Cargo & Airspace design (SE-170) Aircraft and Procedures Loading Trn and (SE125) Hazmat (SE134) Avionics (SE136) Training CARGO Midair Icing Maintenance/Systems Figure 20. JIMDAT “remaining risk” safety enhancements.

( Source: Rob Noges, JIDMAT member) Two additional icing-related enhancements have been examined by JIMDAT but have not been submitted for CAST approval. One of the factors for delaying their recommendation is the expected cost of implementing these SE’s as compared with their impact on risk reduction. The impact of these additional enhancements on the risk to the air transportation system has not been calculated. The two additional enhancements are: SE 135: Ground Operations To prevent fatal accidents resulting from ice-contaminated wings, control surfaces, and other critical surfaces, aircraft operators, airport authorities, regulators, and air traffic control should develop and implement ground operations policies and procedures to reduce the likelihood of ice contamination on flight critical surfaces.

SE137: Weather Information To prevent fatal accidents and incidents resulting from inadequate aircraft deicing and anti-icing and runway and taxiway snow removal operations due to insufficient information on surface weather conditions, weather information providers should provide airport-zone-specific advanced- detection and forecast Page 31 Subsonic Aircraft Safety Icing Study products (e.g., Weather Support to Deicing Decision Making (WSDDM)) to operators, flight crews, and airport authorities for ground operations.

Aircraft Icing Research Alliance (AIRA) The Aircraft Icing Research Alliance is an organization that was formed in 2002 to increase coordination of aircraft icing research. Members of the AIRA include representatives from NASA, the FAA, the National Oceanic and Atmospheric Administration (NOAA), Transport Canada (TC), Environment Canada (EC), the National Research Council of Canada (NRC) , National Defense of Canada, and Defense Science and Technology Laboratory (DSTL) of the United Kingdom. Industry, academia, and other nations also participate in the AIRA either by assisting with setting research priorities or via joint research as part of an AIRA subcommittee.

The AIRA created a list of 87 icing-related research focus areas (RFA’s), which are divided into six themes: (1) in-flight icing, (2) aircraft icing while on the ground, (3) runway winter contamination, (4) operations, human factors, and training, (5) safety and economic analysis, and (6) emerging technologies. The AIRA identified 17 of these as priority RFA’s, and results are shown in Table 12. Note that three of the themes do not contain priority RFA’s and the in- flight icing theme is further divided into seven sub-themes. (ref. 20) Page 32 Subsonic Aircraft Safety Icing Study Table 12. AIRA Priority Research Focus Areas (RFA’s) Aircraft icing while on the Operations, human factors In-flight icing ground and training Aerodynamic performance ƒ Fluid integrity detection ƒ Flight crew training module and handling qualities system for operations in icing ƒ Aerodynamic performance ƒ Performance effects of conditions and handling qualities in icing limited contamination Atmospheric characterization ƒ High ice-water content Weather and forecasting ƒ Nowcasting development and validation ƒ Dissemination of weather information ƒ Forecasting development and validation ƒ Remote sensing systems for icing detection Facilities, simulators and instrumentation systems ƒ Comparison of tools, techniques and facilities and the establishment of standards Fundamental ice physics ƒ Ice formation mechanics ƒ Scaling Ice shedding ƒ Ice shedding mechanisms Propulsion and Power Plant Icing ƒ High ice-water ingestion Page 33 Subsonic Aircraft Safety Icing Study 4.5 NTSB Most-Wanted List The NTSB publishes an annual list (ref. 21) of the most-wanted transportation safety improvements for various modes of transportation (e.g., highway). In 1996, a recommendation was added to the aviation list to “reduce dangers to aircraft flying in icing conditions.” The Board’s specific recommendation to the FAA about this issue is as follows: Use current research on freezing rain and large water droplets to revise the way aircraft are designed and approved for flight in icing conditions.

Conduct additional research with NASA to identify realistic ice accumulations and incorporate new information into aircraft certification and pilot training requirements.

NTSB icing expert Dan Bower stated in 2005 (ref. 22) “...the industry may not yet fully appreciate the danger of even a small amount of ice on upper wing surfaces…Research results have shown that fine particles of frost or ice, the size of a grain of table salt and distributed as sparsely as one per square centimeter over an airplanes wing’s upper surface can destroy enough lift to prevent that airplane from taking off.” The FAA responded to this recommendation by making revisions to and developing new rules to address icing, which resulted in a total of five rule-making activities. Beyond these activities, the FAA contends that additional actions are not necessary in regard to flight in icing conditions.

The NTSB recommends additional actions and although the number of icing-related accidents and serious incidents is small, considers icing to be a significant aviation danger. In June 2007, the NSTB asked the FAA to complete the following actions: (1) complete efforts to revise icing certification criteria, testing requirements, and restrictions on operations in icing conditions and (2) evaluate all aircraft that are certified for flight in icing conditions using the new criteria and standards. (ref. 23) 4.6 Summary of Icing Research Priorities Based on the prioritized lists that were discussed earlier and on additional information in regard to icing-related atmospheric problems (ref. 24-26), the following are the highest priorities for aviation icing-related research: • Improved propulsion system tolerance to weather − Develop more accurate physics-based models that can predict the effects of adverse weather on fans, compressors, and combustor stability.

− Conduct research on high ice-water ingestion.

− Conduct research on high-altitude ice-particle ingestion.

• Development of methods to sense and document actual icing conditions − Develop systems with better measurement and mitigation of upstream environmental conditions.

− Develop improved on-board weather systems and tools (including incorporation of on- board sensor data about wind, icing conditions, etc.)

− Develop and validate better methods for nowcasting.

Page 34 Subsonic Aircraft Safety Icing Study − Develop and validate better methods for forecasting.

− Develop dissemination of weather information.

− Develop remote sensing systems for icing detection.

• Research to identify realistic ice accumulations to improve aircraft certification and pilot training − Conduct research on freezing rain and large water droplets to revise the manner in which aircraft are designed and approved for flight in icing conditions.

− Develop flight crew training module for operations in icing conditions.

− Conduct research on ice formation mechanics.

− Measure performance effects of limited contamination.

− Assess holdover times.

• Aircraft performance and handling qualities in icing − Research the effects of icing on aircraft aerodynamics.

− Study ice shedding mechanisms.

− Research anti-icing fluids.

− Define the effects of ice accretions, especially the effect on roll characteristics.

− Develop smart pitch guidance systems.

− Develop smart icing systems.

Page 35 Subsonic Aircraft Safety Icing Study

5. FUTURE ICING RESEARCH REQUIREMENTS

The expected outcome of this analysis is to provide input for future icing research requirements, including the appropriate alignment within the AvSafe Program. Specifically, these requirements should address whether future icing research should be conducted through the IRAC project or through another AvSafe project, such as the Integrated Vehicle Health Management (IVHM) Project or the Integrated Intelligent Flight Deck (IIFD) Project. The rationale is that although most of the icing research priorities are in the area of flight control, others are related to sensing and identification, which are interests for the IIFD and IVHM projects, respectively. This section summarizes the published goals and objectives for the IRAC, IVHM, and IIFD projects, examines their relevance to the icing research priorities that have been presented in the previous section, and presents suggestions for future NASA icing research.

5.1 IRAC Goals and Objectives The objectives of the NASA Aviation Safety Program (AvSafe) (ref. 2) are: • “Conduct long-term, cutting-edge research that will produce tools, methods, and technologies to improve the intrinsic safety attributes of current and future aircraft.” • “Overcome safety technology barriers that would otherwise constrain full realization of the Next Generation Air Transportation System.” According to the IRAC Technical Plan (ref. 1), the IRAC project will conduct research to advance the state of aircraft flight control to provide onboard control resilience for ensuring safe flight in the presence of adverse conditions. The goal of the IRAC project is to arrive at a set of validated multidisciplinary integrated aircraft control design tools and techniques for enabling safe flight in the presence of adverse conditions (e.g. faults, damage, upsets). The objective is to advance the state of the art of adaptive controls as a design option to provide enhanced stability and maneuverability margins for safe landing. Adverse events include loss of control that is caused by environmental factors and actuator and sensor faults or failures and will expand to include more complicated damage conditions.

The IRAC project goals and objectives (refs. 1 and 27) are graphically depicted using an objectives tree format (ref. 28) in Figures 21 and 22.

Page 36 Subsonic Aircraft Safety Icing Study To improve aircraft safety for current and future aircraft (AvSafe Program Goal #1) To ensure safe flight of current and next generation subsonic transports in the presence of adverse conditions (faults, damage, and/or upsets) To provide on-board To provide enhanced control resilience stability and maneuverability margins for safe IRAC landing To advance the project state of aircraft goals and flight control objectives To advance the To advance the state of the art of state of the art of adaptive controls adaptive controls as a design as a design option option To develop a set of validated multidisciplinary integrated aircraft control design tools and techniques Key Will help to achieve Figure 21. NASA IRAC objectives tree.

Page 37 Subsonic Aircraft Safety Icing Study To develop a set of validated multidisciplinary integrated aircraft control design tools and techniques To develop To conduct To develop To investigate how To validate advanced fundamental integrated planning, guidance, technologies methods for research on diagnostic and and control concepts that are modeling and engine prognostic can be used to developed by control of control and aeroservoelastic increase safety of IRAC for aircraft engine methods flight under recovery from performance emergency loss-of-control situations that are flight conditions caused by adverse and damage conditions scenarios Validation Integrated Integrated Airframe Intelligent and dynamics propulsion and flight verification and flight control structural planning methods control and dynamics and and dynamics guidance testbeds Figure 22. NASA IRAC subproject objectives tree.

5.2 IVHM Goals and Objectives The goals of the IVHM project are to: (1) reduce system and component failures as causal and contributing factors in aircraft accidents and incidents and (2) provide continuous on-board situational awareness of the state of vehicle health for use by the flight crew, ground crew, and maintenance depot. The purpose of this research is to develop technologies that will enable the early detection of system/component degradation and damage and thereby provide the ability to prevent or gracefully recover from in-flight failures. The objectives of the IVHM project are as follows (ref. 29): • Develop tools and techniques to: − Determine the state of subsystems such that the state of the entire vehicle can be determined for accurate prognosis.

− Diagnose coupled degradation/malfunction/failure/hazard conditions and predict their effects on vehicle safety.

− Mitigate damage/degradation/failures inflight.

• Develop a public database and testing capabilities for IVHM technologies.

5.3 IIFD Goals and Objectives The IIFD project within the AvSafe Program conducts research that is shared with government and industry to create future flight decks that are safer and more capable. One of the advanced features that is envisioned for these future flight decks is the integration of external hazard Page 38 Subsonic Aircraft Safety Icing Study sensors with decision-aiding functions. The objectives tree for the IIFD project is shown in Figure 23 (ref. 30).

To develop flight deck technologies that mitigate operator, automation and environment induced hazards for future operational concepts To develop To develop To develop fail-safe To enable robust detection of crew/vehicle monitoring methods for changing external hazards with interface technologies the operator/ sufficient time-to-alarm for technologies to enable automation roles in safe maneuvering to avoid that reduce the detection of the presence of the hazards risk of pilot unsafe detected disability error behaviors states To develop a comprehensive surveillance system design Figure 23. NASA IIFD project objectives tree.

5.4 Mapping of Goals to Icing Research Priorities High-priority icing research issues were summarized in section 4. Based on the current stated goals and objectives of the AvSafe IRAC, IIFD, and IVHM projects, Table 13 provides the suggested mapping of these goals to the icing research issues. Note that certain research issues do not map to an AvSafe project; this is a result of the strict alignment to the current project objectives and should not be interpreted as a lack of technical expertise within NASA in these icing research areas.

Page 39 Subsonic Aircraft Safety Icing Study Table 13. NASA AvSafe Project Goals Mapped to Icing Research Priorities Aviation safety program Icing Research Priorities IRAC IVHM IIFD Flight On-board Remote control sensing sensing Improved propulsion system tolerance to weather Development of more accurate physics based models that can predict the effects of adverse weather on fans, compressors X and combustor stability High ice water ingestion X High altitude ice particle ingestion X Development of methods to sense and document actual icing conditions Systems with better measurement and mitigation of upstream environmental conditions Improved onboard weather systems and tools (including incorporation of onboard sensor data about wind, icing X X conditions, etc.)

Nowcasting development and validation Forecasting development and validation Dissemination of weather information X Remote sensing systems for icing detection X Research to identify realistic ice accumulations to improve aircraft certification and pilot training Research on freezing rain (ZR) and large water droplets to revise the way aircraft are designed and approved for flight in X icing conditions Flight crew training module for operations in icing conditions Ice formation mechanics X Performance effects of limited contamination X Assessment of holdover times Aircraft performance and handling qualities in icing Research on the effects of icing on aircraft aerodynamics X Ice shedding mechanisms Anti-icing fluids Define the effects of ice accretions, especially the effect on roll X Smart pitch guidance systems X Smart icing systems X X X Page 40 Subsonic Aircraft Safety Icing Study

6. DISCUSSION AND CONCLUSIONS

Although FAR Part 135 operations have the highest percentage of icing-related events and no fatal Part 121 accidents have occurred since 1994, icing is still a safety concern across all categories of subsonic aircraft. As previously stated, when ramp- and security-related accidents are eliminated from the NTSB data set, 9.5 percent of the total fatal Part 121 accidents were attributable to icing, which is more than for Part 91 (3.7 percent), Part 135 Scheduled (8.3 percent) and Part 135 Nonscheduled (8.0 percent). Three of the six icing accidents in the NASA NTSB data study (see section 2.1) involved turboprop engines, while the remaining three accidents involved turbofan or turbojet engines. Aircraft with turboprop and reciprocating engines still comprise a significant portion of Part 121 operations. In 2004, 41 percent of the aircraft used by regional airlines had turboprop or reciprocating engines. Of the 663 icing incidents reported in the voluntary ASRS database for which an engine type was provided, 223 had reciprocating engines, but 208 had turbojet engines, and 87 had turboprop engines.

Further, 141 of the 663 total icing-related incidents resulted in the crew declaring an emergency.

In the Petty and Floyd review (ref.3) of airframe-icing accidents from 1982 to 2000, 60 percent of accidents for Part 121 were fatal compared with 47 percent for GA and 26 percent for Part 135 during the same time period. According to the Green Study (ref. 8), ASRS reports indicate that pilots try to minimize the amount of ice accretion by using rapid descent but their efforts are often hampered by ATC requirements. Even in military operations, icing still appears to be a problem. Only 0.5 percent of U.S. Army aircraft accidents are icing related, but this number is still of concern due to the Army’s strict regulations that forbid flight into known or severe icing conditions.

A review of studies by subject-matter experts from the NRC, the CAST/JIMDAT, the AIRA, and the NTSB was summarized into four high-priority icing research areas: (1) improved propulsion system tolerance to weather, (2) development of methods to sense and document actual icing conditions, (3) research to identify realistic ice accumulations to improve aircraft certification and pilot training, and (4) aircraft performance and handling qualities in icing.

Based on the current stated goals and objectives of the AvSafe IRAC, IIFD, and IVHM projects, the suggested mappings of these goals to the icing research issues were listed in Table 13. As shown in Table 13, the IRAC project is encouraged to conduct work in all of the identified areas except for the “development of methods to sense and document actual icing conditions.” Page 41 Subsonic Aircraft Safety Icing Study

REFERENCES

1. Totah, Joseph; Krishmakumar, Kalmanje; and Viken, Sally: Integrated Resilient Aircraft Control Project Implementation Plan , National Aeronautics and Space Administration, April 13, 2007. http://www.aeronautics.nasa.gov/programs_avsafe.htm Accessed June 7, 2007.

th 2. Schlickenmaier, Herb: “Aviation Safety Program”, presented at the 45 AIAA Aerospace Sciences Meeting, Reno, Nevada, January 8-11, 2007.

3. Petty, Kevin R.; and Floyd, Carol D. J.: “A Statistical Review of Aviation Airframe Icing Accidents in the U.S.,” Proceedings of the 11th Conference on Aviation, Range, and Aerospace Meteorology , Amer. Meteor. Society, 2004.

4. National Transportation Safety Board: Annual Review of Aircraft Accident Data, U.S. Air Carrier Operations Calendar Year 1997 , NTSB9ARC-02901, PB2002-106890, Notation 7425, January 2002.

5. Evans, Joni K.: “An Application of CICCT Accident Categories to Aviation Accidents in 1988- 2004,” NASA CR-2007-214888, July 2007.

6. Commercial Aviation Safety Team/International Civil Aviation Organization (CAST/ICAO) Common Taxonomy Team: Aviation Occurrence Categories: Definitions and Usage Notes , version 4.1.3, February 2006. http://www.intlaviationstandards.org/ Accessed March 7, 2006.

7. National Transportation Safety Board: Aviation Accident Statistics , March 13, 2007.

http://www.ntsb.gov/aviation/Stats.htm Accessed on May 14, 2007.

8. Green, Steven D.: A Study of U.S. Inflight Icing Accidents and Incidents 1978 to 2002, AIAA- 2006-82, January 9-12, 2006 .

9. Jones, Sharon Monica; and Reveley, Mary S.: “An Overview of the NASA Aviation Safety Program Assessment Process,” Proceedings of the AIAA 3rd Annual Aviation Technology, Integration and Operations (ATIO) Forum , Denver, CO, November 2003.

10. Benson, Tom, editor: Beginner’s Guide to Propulsion , last updated July 20, 2007.

http://www.grc.nasa.gov/WWW/K-12/airplane/bgp.html Accessed August 22, 2007.

11. Berton, Jeff, curator: Types of Gas Turbine Engines , last updated on August 13, 2007, http://asao.grc.nasa.gov/reengine/types.html Accessed August 22, 2007.

12. Hill, Philip G.; and Peterson, Carl R.: Mechanics & Thermodynamics of Propulsion , Addison- Wesley, Reading, Massachusetts, 1965.

13. Mattingly, Jack D.: Elements of Gas Turbine Propulsion , McGraw-Hill, New York, New York, 1996.

14. Springer, Edwin H.: Constructing a Turbocharger Turbojet Engine , Turbojet Technologies, 2001.

Page 42 Subsonic Aircraft Safety Icing Study 15. Velocity Group: Summary of Passenger Aircraft in Regional Airline Use–2004 , Regional Airline Association.

http://www.raa.org/client_files/Carriers_services/Summary_Passenger_Aircraft.pdf Accessed September 17, 2007.

16. Cole, Jeffrey A.; and Sand, Wayne R.: Statistical Study of Aircraft Icing Accidents, AIAA-91- 558, January 1991.

17. Peck, Lindamae; Ryerson, Charles C.; and Martel, C. James: Army Aircraft Icing , U.S. Army Corps of Engineers Technical Report, ERDC/CRREL TR-02-13, September 2002.

18. Federal Aviation Administration: NASDAC Review of NTSB Weather Related Accidents , modified April 19, 2007.

http://www.asias.faa.gov/aviation_studies/weather_study/studyindex.html Accessed August 21, 2007.

19. National Research Council (U.S.): Decadal Survey of Civil Aeronautics: Foundation for the Future , The National Academies Press, Washington, DC, 2006.

20. Aircraft Icing Research Alliance Web site, http://icingalliance.org Accessed July 19, 2007.

21. National Transportation Safety Board: NTSB Most Wanted – Aviation , February 2007.

http:///www.ntsb.gov/Recs/mostwanted/air_ice.htm Accessed August 16, 2007.

22. “List of NTSB ‘Most Wanted’ Safety Improvements Languish,” Air Safety Week , November 21, 2005.

23. Rosenker, Mark V.: Testimony of Mark V. Rosenker, Chairman, National Transportation Safety Board before the Committee on Transportation and Infrastructure Subcommittee on Aviation, U.S. House of Representatives , June 6, 2007.

24. Liu, Guosheng; and Curry, Judith A.: “Remote Sensing of Ice Water Characteristics in Tropical Clouds Using Aircraft Microwave Measurements,” Journal of Applied Meteorology , Vol.

37, April 1998.

25. Mason, Jeanne G.; Strapp, J. Walter; and Chow, Philip: The Ice Particle Threat to Engines in Flight, AIAA-2006-206, January 2006.

26. Veillette, Patrick R.: “High Altitude Engine Flame-Outs,” Business and Commercial Aviation, July 2007.

27. Totah, Joseph; Krishmakumar, Kalmanje; and Viken, Sally: “Aviation Safety Program, Integrated Resilient Aircraft Control Project: Project Overview and Status,” presented at NASA Glenn Research Center, Cleveland, Ohio, June 15, 2007.

28. Gibson, John E.; Scherer, William T.; and Gibson, William F.: How to Do Systems Analysis , John Wiley & Sons, Inc., Hoboken, New Jersey, 2007.

29. Srivastava, Ashok; and Grady, Joe: Aviation Safety Program, Integrated Vehicle Health Management: Technical Plan Summary, National Aeronautics and Space Administration, July 3, Page 43 Subsonic Aircraft Safety Icing Study 2007. http://www.aeronautics.nasa.gov/nra_pdf/ivhm_tech_plan_c1.pdf Accessed August 31, 2007.

30. Young, Steven D.; and Quon, Leighton: Aviation Safety Program, Integrated Intelligent Flight Deck: Technical Plan Summary , National Aeronautics and Space Administration, 2007.

Page 44 Form Approved REPORT DOCUMENTATION PAGE OMB No. 0704-0188 The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number.

PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS.

1. REPORT DATE (DD-MM-YYYY) 2. REPORT TYPE 3. DATES COVERED (From - To) 01 - 2008 01- Technical Memorandum 4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER Jones, Sharon Monica; Reveley, Mary S.; Evans, Joni K.; and Barrientos, Francesca A.

5b. GRANT NUMBER 5c. PROGRAM ELEMENT NUMBER 6. AUTHOR(S) 5d. PROJECT NUMBER Subsonic Aircraft Safety Icing Study 5e. TASK NUMBER 5f. WORK UNIT NUMBER 457280.02.07.07.02 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION REPORT NUMBER NASA Langley Research Center Hampton, VA 23681-2199 L-19435 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR'S ACRONYM(S) National Aeronautics and Space Administration NASA Washington, DC 20546-0001 11. SPONSOR/MONITOR'S REPORT NUMBER(S) NASA/TM-2008-215107 12. DISTRIBUTION/AVAILABILITY STATEMENT Unclassified - Unlimited Subject Category 03 Availability: NASA CASI (301) 621-0390 13. SUPPLEMENTARY NOTES An electronic version can be found at http://ntrs.nasa.gov 14. ABSTRACT NASA’s Integrated Resilient Aircraft Control (IRAC) Project is one of four projects within the agency’s Aviation Safety Program (AvSafe) in the Aeronautics Research Mission Directorate (ARMD). The IRAC Project, which was redesigned in the first half of 2007, conducts research to advance the state of the art in aircraft control design tools and techniques. A “Key Decision Point” was established for fiscal year 2007 with the following expected outcomes: document the most currently available statistical/prognostic data associated with icing for subsonic transport, summarize reports by subject matter experts in icing research on current knowledge of icing effects on control parameters and establish future requirements for icing research for subsonic transports including the appropriate alignment. This study contains: (1) statistical analyses of accident and incident data conducted by NASA researchers for this “Key Decision Point”, (2) an examination of icing in other recent statistically based studies, (3) a summary of aviation safety priority lists that have been developed by various subject-matter experts, including the significance of aircraft icing research in these lists and (4) suggested future requirements for NASA icing research. The review of several studies by subject-matter experts was summarized into four high-priority icing research areas. Based on the Integrated Resilient Aircraft Control (IRAC) Project goals and objectives, the IRAC project was encouraged to conduct work in all of the high-priority icing research areas that were identified, with the exception of the developing of methods to sense and document actual icing conditions.

15. SUBJECT TERMS Accidents (aircraft); Air safety; Aircraft accidents; Aircraft safety; Icing (aircraft) 19a. NAME OF RESPONSIBLE PERSON 18. NUMBER 17. LIMITATION OF 16. SECURITY CLASSIFICATION OF: OF ABSTRACT STI Help Desk (email: help@sti.nasa.gov) a. REPORT c. THIS PAGE b. ABSTRACT PAGES 19b. TELEPHONE NUMBER (Include area code) 45 (301) 621-0390 U U U UU Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std. Z39.18