CALHOUN, GA, USA
N948VV
Douglas DC-9-32
The properly certificated and qualified flight crew received flight release documents containing weather information from airline dispatch that was incomplete and did not contain pertinent Convective SIGMETs and a tornado watch bulletin. However, the flight crew stated that before departure, they looked at a computerized weather display and observed a line of adverse weather conditions along their intended route of flight. During initial climb after takeoff, the flight crew noticed that they were approaching a line of thunderstorms. The captain initially planned to fly around this line but noticed a 10-mile gap in the line that was depicted on the airplane's on-board weather radar display. The captain decided to fly through the gap, which he could also see when he looked outside of the airplane. The first officer did not challenge his decision. Both flight crewmembers stated that the weather radar was functioning normally. It is unknown exactly what was depicted on the aircraft radar. Radar and weather data indicate that the airplane penetrated an extreme weather echo (VIP level 6) that likely contained a severe thunderstorm, hail, and severe to extreme turbulence. The thunderstorm top was above 45,000 feet. Neither the flight attendants nor the passengers received an adequate or timely briefing to remain seated and to prepare for the possibility of turbulence. The aft flight attendant and a passenger were seriously injured during the turbulence encounter. Hail shattered the front windshield and damaged the pitot system, radome, wings, tail, and engines. The flight crew successfully performed an emergency landing at a nearby airport. The investigation revealed that the captain had been involved in two other air carrier incidents involving adverse weather conditions. It also revealed that the airline lacked adequate training and guidance regarding hazardous weather interpretation and avoidance, as well as adequate procedures to notify flight attendants about potential turbulence. According to FAA guidance, flight crews should avoid intense or extreme level radar echoes by at least 20 miles. This guidance also instructs pilots that such echoes should be separated by at least 40 miles before flight is attempted between the echoes.
HISTORY OF FLIGHT On May 7, 1998, about 1920 eastern daylight time, a Douglas DC-9-32, N948VV, operated by AirTran Airlines, Inc., as flight 426, encountered turbulence and hail near Calhoun, Georgia, while climbing through 20,000 feet above mean sea level (msl). Of the 2 flight crewmembers, 3 flight attendants, and 82 passengers who were on board, 1 flight attendant and 1 passenger received serious injuries during the encounter, and the remaining occupants were uninjured. The airplane was substantially damaged by hail. The flight crew performed a successful emergency landing at the Chattanooga Metropolitan Airport (CHA) near Chattanooga, Tennessee. Flight 426 was being operated as a scheduled domestic passenger flight under the provisions of 14 Code of Federal Regulations (CFR) Part 121 and was destined for Chicago Midway Airport (MDW), Chicago, Illinois, from Atlanta Hartsfield International Airport (ATL), Atlanta, Georgia. On the evening before the accident, the flight crew arrived at Dallas-Fort Worth International Airport (DFW), Texas, after flying the first day's schedule of a two-day trip. The overnight stay at DFW was a scheduled rest period of 14 hours, and both pilots stated that they were well rested the next day. On the following afternoon, the flight crewmembers arrived at the DFW airport about 1400 to begin their preparations for a flight to ATL. Following a routine flight preparation, including a check of the airplane's weather radar, they departed DFW about 1500. The flight crew stated that the flight to ATL was uneventful and that they did not see any adverse weather conditions to the north of ATL. They arrived at ATL about 1740 and were scheduled to be on the ground for one hour before the next flight. At the gate in Atlanta, the flight crew was given departure papers for the flight to MDW. The departure papers contained a flight plan, a preliminary load manifest and a weather package. The captain stated that he looked at a weather display in the AirTran operations area and observed a "line of weather" along the Georgia and Tennessee borders. He stated that he was not sure if the weather depiction was from a satellite or radar, but the line was significant. He stated that the line had red radar displays in north Georgia and was more like a "blob" instead of single cell. The first officer stated that he observed a radar display in the AirTran operations area that depicted a line of weather with red, yellow, and green colored cells along the Georgia and Tennessee borders. The captain stated that after returning to the airplane, he briefed a flight attendant, telling her that he did not notice any hazardous weather around ATL, that MDW weather was improving, and that there was some adverse weather en route to MDW. After a brief ground delay, the flight departed from runway 26L at 1907 and was assigned a 350-degree heading and a clearance to climb to 14,000 feet. The captain stated that during climb, the weather radar was on and selected to the 80-mile range in the "normal" mode with the gain set to "auto." He stated that the weather radar was showing a solid line with dark returns in northwest Georgia. He also visually observed the line by looking out of the cockpit windows. The captain stated that he decided to delay the 10,000-foot departure announcement to the flight attendants because he believed that it would have prompted them to begin meal service and he wanted them to remain in their seats. The pilots stated that they heard the forward flight attendant get up from her seat, so they opened the cockpit door to tell her to sit down. Between 10,000 and 12,000 feet, the flight was cleared to continue the climb to 23,000 feet. At that time, the crew asked for and received a 330-degree heading to remain clear of weather. The pilots stated that they were in visual meteorological conditions at this time and that their radar showed a light area of precipitation west of the line of weather. They stated that their radar also showed an approximately 10-mile gap between two storm cells in the line of weather and that, through this gap, they were able to see that no adverse weather conditions were on the other side. The flight crewmembers reported that as they proceeded through the gap, they observed an egg-sized piece of hail hit the center windshield, causing its outer pane to shatter. At the same time, they began experiencing turbulence that lasted about 10 seconds. They stated that approximately 1 to 2 seconds after the encounter with the first piece of hail, the airplane encountered significant hail, which lasted about 3 to 5 seconds and shattered the outer panes of the captain's and first officer's windshields. Both pilots stated that the hail caused significant damage to the skin of the airplane but that no control problems were noted. They stated the nose radome cover was torn off and that the noise level in the cockpit was high. They also stated that the airspeed indication was zero and the altimeter indications were erratic. Data from the digital flight data recorder (DFDR) indicate that at 1916:05, about 3 minutes before the hail encounter, the airplane's pitch angle decreased from 4 degrees to -4 degrees while the airplane climbed from 15,000 to 16,000 feet. The airplane's airspeed increased from 305 knots to 350 knots, its pressure altitude decreased from 15,600 feet to 14,600 feet, its vertical acceleration increased to 1.5 g, its longitudinal acceleration decreased to 0 g, its roll changed from 0 degrees to -6 degrees, its control column pitch became erratic, and its engine pressure ratios (EPR) decreased from 1.8 to approximately 1.4. About 1 minute later, at 1917:05, the DFDR parameters stabilized. At 1918:30, while climbing through 20,500 feet, many of the DFDR parameters became erratic again. Roll degrees reached an extreme of about -40 degrees, and continued to oscillate between 20 and -10 degrees. Pitch angle increased to about 10 degrees, then decreased to -6 degrees, and remained erratic throughout the incident. Vertical acceleration oscillated between 2.9 g and -.9 g. Longitudinal acceleration fluctuated between 0 g and .15 g. Control column pitch became erratic, though generally increased, to a maximum of approximately 12 degrees. Indicated airspeed became erratic but generally decreased from 350 knots to 280 knots. At 1919:30 the EPRs decreased from approximately 1.9 to 1.5. Pressure altitude increased during the incident from approximately 19,500 feet to approximately 23,000 feet. The flight crewmembers stated that as they emerged from the weather encounter, they told ATC what had happened, requested vectors to fly direct to CHA, declared an emergency, and asked to have emergency equipment standing by when they landed. ATC assigned a vector to runway 20 at CHA and a lower altitude. They requested that the controller provide a ground controlled approach and verbal ground speed readouts every 10 seconds. The first officer performed the normal checklists for landing and tuned in the frequency for the runway 20 instrument landing system. The captain and first officer stated that, when they were about 10 to 12 miles from the airport, they began to see the runway through small, unshattered areas of the windshields. They reported that while on the approach, they compared their altimeter data with the altitude information that was verbally provided by ATC and that the altimeters appeared to be providing correct altitude readouts. The landing was successful. After landing, the fire department performed a fire inspection of the airplane and the airplane was taxied to a gate. The crew requested medical assistance. Upon arrival at the gate, medical personnel came on board to attend to the injured. DAMAGE TO AIRPLANE The airplane's radome had separated and portions of it had been ingested into the right engine. All three outer panes of the cockpit front windshield were shattered. The wing leading edge devices, horizontal stabilizer leading edge, vertical stabilizer leading edge, and both left and right engine inlet cowls were dented and damaged. Both engines sustained foreign object damage. PESONNEL INFORMATION The Captain The captain, age 57, was hired by Valujet Airlines on January 23, 1993, as a DC-9 captain. He remained with the airline after it was acquired by AirTran Airways. Before his employment with Valujet, the captain served 21 years as a pilot for Eastern Air Lines and 6 years as a pilot in the U.S. Air Force. At the time of the accident, the captain held an airline transport pilot (ATP) certificate with type ratings in the DC-9. He accumulated about 7,200 hours in the DC-9 and about 4,800 hours as a DC-9 pilot-in-command. The captain's most recent Federal Aviation Administration (FAA) first-class medical certificate was issued on February 25, 1998, with the limitation that he wear corrective lenses. A review of FAA and National Transportation Safety Board records revealed that on August 15, 1990, the captain failed the oral portion of his DC-9 type rating examination. He was retested and subsequently passed. On September 15, 1990, the captain failed the entire flight test portion of his DC-9 type rating check ride. He was retested and subsequently passed. On January 28, 1994, the captain was involved in an air carrier incident in which he lost control of a DC-9 during a takeoff roll in adverse weather conditions. The takeoff was aborted and the airplane departed the runway onto a grassy area. On July 5, 1994, the captain was involved in an air carrier accident in which he was in command of a DC-9 that encountered moderate convective turbulence during climbout, causing a flight attendant to suffer multiple leg fractures. After the May 7, 1998, accident, the captain was reassigned as a first officer. The First Officer The first officer, age 30, was hired by Valujet Airlines on August 21, 1995, as a DC-9 first officer. He remained with the airline after it was acquired by AirTran Airways. Before his employment with Valujet, the first officer flew multiengine turboprop airplanes for USAir Express (a 14 CFR Part 135 scheduled commuter operator). At the time of the accident, the first officer held an ATP certificate with type ratings in the DC-9 and the British Aerospace BA-3100. He accumulated a total of about 5,500 flight hours, including 1,500 hours in the DC-9 and 2,000 hours as pilot-in-command. The first officer's most recent FAA first-class medical certificate was issued on August 29, 1997, with no limitations or waivers. A review of AirTran records revealed that on April 21, 1998, the first officer had an unsatisfactory oral examination during a requalification on the DC-9. As a result, he received an additional 5 hours of instruction. He was retested and subsequently passed. AIRCRAFT INFORMATION The accident airplane, a Douglas DC-9-32, was equipped with a three-color weather radar. No unresolved maintenance discrepancies with the weather radar or any other pertinent systems were reported. METEOROLOGICAL INFORMATION The Safety Board obtained reflectivity images from the Weather Surveillance Radar-1988 (WSR-88D) system at Peachtree City, Georgia, and overlaid these images with AirTran flight 426's ground track from ATC radar data. The images were recorded at 1919 at three elevation angles: 0.5, 1.5, and 2.4 degrees. The 0.5 degree elevation scan of base reflectivity that was completed at 1919:15 (when the airplane was at 21,300 msl) showed the maximum weather echo intensity observed at 9,140 feet msl along flight 426's ground track as extreme (level 6). The 1.5 and 2.4 degree elevation scans of base reflectivity that were completed about 1919:05 to 1919:15 (when the airplane was at 21,100 to 21,300) showed the maximum weather echo intensity at 17,140 and 24,840 msl along flight 426's ground track as extreme (level 6). Visible and infrared data from the Geostationary Operational Environmental Satellite (GOES) 8 were reviewed on the Safety Board's Man Computer Interactive Data Access System (McIDAS) workstation. (McIDAS is an interactive meteorological analysis and data management computer system administered by personnel at the Space Science and Engineering Center at the University of Wisconsin at Madison.) The visible image for 1915 indicated that flight 426's flight track went through an area of increased cloud brightness. There is also evidence of overshooting cloud tops in this area. GOES 8 infrared images for 1902, 1915, and 1932 showed cold radiative temperatures (high cloud tops) in the area of flight 426's flight track. Review of the GOES 8 infrared image recorded at 1915 indicated that the track of AirTran flight 426 passed within about 3 nautical miles west of an estimated cloud top of 44,000 feet. The closest track points were at times of 1919:05 and 1920:06 at altitudes of 21,100 feet and 22,900 feet respectively. Lightning strike data recorded from 1900 to 1940 indicated that there were 563 cloud to ground lightning strikes within a 20-mile radius of the estimated location of the hail encounter. Preliminary local storm reports from the National Weather Service (NWS) in Peachtree City, Georgia, noted that at 1950, the citizens in Jasper, Georgia, reported hail that measured 1.00 inch in diameter. At 2000, hail measuring 1.75 inches in diameter was reported 2.5 miles south of Jasper. Jasper is located about 22 nautical miles east of flight 426's ground track location time of 1918:05. A report from Ball Ground, Georgia, indicated hail measuring 1.25 inches in diameter at 2002. Ball Ground is located about 21 nautical miles east-southeast of flight 426's ground track location time of 1918:05. Several NWS Convective SIGMETs (Significant Meteorological Information) (attached) were issued before the accident for times and locations consistent with those of the accident flight. These advisories noted severe thunderstorms, tornadoes, hail, and strong wind gusts, and many of them were included in the flight release provided to the flight crew before the accident flight. In addition, the NWS issued Tornado Watch No. 302, which was in effect for the time and location of the hail encounter. The NWS also issued a tornado warning at 1858 that was valid until 1930 for Eastern Gordon County, which is the location of the hail encounter. This warning was issued after the pilots had received their flight release documents and began the flight. Examination of the information contained in the flight release documents for AirTran flight 426 revealed that three pertinent convective SIGMETs and a tornado watch report were not included. According to AirTran representatives, a programming flaw in the airline's DC-9 dispatch computer system caused these pertinent weather reports to be omitted from the flight release. SURVIVAL ASPECTS According to interviews with the flight attendants (attached), 2 flight attendants were seated with their seat belts fastened at the front of the airplane, and the third was seated with her seat belt fastened at the rear of the airplane before the accident. The forward flight attendants stated that they did not hear the flight crew announce the "Departure Check," which would have been their signal to begin the meal service. One of the forward flight attendants stated that she stood because she thought the captain had forgotten to make the announcement and was immediately told by the captain to sit back down. The aft flight attendant stated that when she saw the forward flight attendants get up from their jumpseats, she thought she had missed the Departure Check announcement and stood up to begin the inflight service. One of the forward flight attendants contacted her via interphone to tell her to sit down, which she did. She stated that as she was fastening her lap belt, she was lifted out of her seat by turbulence and her head struck the ceiling, causing a serious injury. A female adult passenger was also seriously injured when she was struck by bags that fell out of an overhead compartment during the turbulence encounter. She was seated on the right side of the airplane in the first row of economy seats. ADDITIONAL INFORMATION FAA Guidance Regarding Thunderstorms The following are summarized
The failure of the flight crew to maintain adequate separation from hazardous meteorological conditions. Factors contributing to the accident were: (1) the failure of AirTran Airlines to provide adequate crewmember training and guidance regarding hazardous weather encounters; (2) the failure of the flight crew to provide an adequate and timely briefing to the flight attendants regarding turbulence; and (3) the presence of hail and turbulence.
Source: NTSB Aviation Accident Database
Aviation Accidents App
In-Depth Access to Aviation Accident Reports
