BOEING 777-222ER

Full Narrative

HISTORY OF FLIGHTOn February 10, 2024, about 15:43 EST, UAL1890, a Boeing 777-222ER, N788UA, encountered moderate to severe turbulence while descending to FL190 inbound to EWR. Of the 280 passengers and crewmembers on board, three flight attendants sustained serious injuries. The flight was conducted under the provisions of Title 14 Code of Federal Regulations Part 121 as a scheduled domestic passenger flight from LAX to EWR.

Both flight crewmembers reported being well-rested before the flight and stated that they conducted a thorough preflight briefing, which included a review of forecasted turbulence near the EWR terminal area. The first officer (FO) used the WSI and SkyPath electronic flight bag (EFB) applications to assess turbulence reports, noting light to moderate activity near the usual WILLIAMSPORT3 arrival. Based on this, the crew anticipated and accepted a northerly routing via the FLOSI4 arrival for turbulence avoidance. 

According to Automatic Dependent Surveillance–Broadcast (ADS-B) data, flight UAL1890 departed LAX at approximately 0831 Pacific Standard Time (PST). The FO was the pilot flying (PF) and the captain was the pilot monitoring (PM). The flight crew stated that the departure from LAX was uneventful, and the airplane leveled off at a cruise altitude of FL350. 

About 1334 EST, the flight crew sent a message via Aircraft Crew Alerting and Reporting System (ACARS) to their dispatcher, stating that they were having a “smooth ride past hour or so, how’s descent into EWR.” About one minute later, the dispatcher replied, “so far only complaint is a moderate by a regional jet.” The flight subsequently climbed to its cruise altitude of FL370. There was no additional communication between the flight and the dispatcher until after the encounter with turbulence. The seatbelt sign was cycled ON and OFF in accordance with standard turbulence mitigation procedures, including passenger address (PA) announcements. The WSI and Skypath applications were continuously monitored, and no significant turbulence was indicated along the route or during the approach phase. 
 
The FO stated that he was responsible for briefing the approach/arrival into EWR and in preparation for the brief, checked for potential turbulence using the WSI and SkyPath application to see if there were any reports of turbulence. Neither application revealed any significant indications of turbulence along their route. After the brief, the captain turned the onboard weather radar ON for the eventual descent into EWR. 

As the airplane began its descent into EWR, ATC issued step-down clearances. About 1437, the flight began a descent to FL350 where it remained for about 20 minutes. The flight then turned towards an easterly direction and began a descent to FL330. About 1510, the flight began to descend to FL310.
 
At 1532:30, the crew of ACA548 (an uninvolved airplane) checked in with the R24A controller and was provided with a pilot weather report (PIREP) indicating moderate to heavy precipitation near the Hancock VOR (HNK), with light to moderate chop, in the descent from FL220 to FL190.

At 1535:10, the flight crew of UAL1890 checked in with the R24A controller and indicated that they were level at FL270. The R24A controller did not respond. In a postaccident interview, the R24A controller stated they did not respond at the time because they were completing other coordination. According to the flight crew, while passing through FL270, the captain announced via the PA that the flight attendants prepare the cabin for landing, followed by turning the seatbelt sign ON. 

At 1536:00, the R24A controller instructed the flight to descend to FL240. 

At 1538:14, the R24A controller conducted a position relief briefing with the R24C controller.  

At 1540:38, the R24C controller instructed the crew of ACA548 to contact the Stewart sector (R05) controller. The crew of ACA548 acknowledged the frequency change and advised the R24C controller they had encountered moderate turbulence. The controller did not acknowledge or disseminate this PIREP to other aircraft, including UAL1890, as required by FAA procedures for Pilot Weather Report (PIREPs) solicitation and dissemination. In a postaccident interview, the controller stated they did not hear the PIREP.

At 1540:52, the R24C controller instructed UAL1890 to cross five miles past the HNK VOR at FL190. The crew acknowledged the instructions. The FO stated that he observed an overcast layer of clouds ahead and below that the flight would penetrate during descent. The flight crew stated there was no indication of potential turbulence as nothing showed on the airplane's weather radar or the SkyPath app. Additionally, there were no reports of turbulence provided by ATC or dispatch directly to the flight crew.  

At 1541, a turbulence pilot report (PIREP) was sent to the flight via ACARS stating “YOUR FLIGHT IS IN CLOSE PROXIMITY TO A MOD TURB PIREP EVENT REPORTED BY A B789 AT 2022 (1522 local) Z FL220 LOCATION HNK RMK TB MOD CHOP 220-190 YOU CLOSEST POINT TO EVENT AT 2041Z”.  It could not be conclusively determined if the crew saw or read the message.

 
At 1542:18, the crew of UAL1890 asked the R24C controller to confirm they were instructed to cross five miles past the HNK VOR and the controller responded in the affirmative.
  
At 1543:06, the R24C controller instructed the crew of UAL1890 to contact the R05 controller. The crew acknowledged the frequency change and advised the R24C controller they had encountered “severely moderate turbulence.” The controller did not respond. 
 
At 1544:07, the crew of UAL1890 checked in with the R05 controller and reported descending to FL190. The R05 controller acknowledged and advised the crew of precipitation and light to moderate turbulence along their route of flight. The crew responded and stated that they had encountered moderate turbulence when they were at FL210. 

The flight crew reported that while descending through about FL210 in IMC, the airplane encountered moderate turbulence lasting a few seconds that caused unsecured items on the flightdeck to be thrown about. A visual break in the clouds was observed, and the crew initiated a right turn using heading select mode to exit the turbulent layer. 

Following the event, the captain immediately called the cabin crew and was informed of multiple injuries with one flight attendant sustaining a head laceration. The flight crew declared a medical emergency and requested paramedics meet the airplane at the gate. 

The remainder of the approach, landing, and taxi were uneventful. Upon arrival at the gate, passengers were instructed to remain seated to allow emergency medical services (EMS) personnel to board. Port Authority Police and EMS responded and attended to the injured flight attendants. Three aft-galley flight attendants (FA 2, FA3, FA 9) were transported to the hospital by ambulance. The captain and FO assisted in managing the cabin and supporting the crew and passengers during deplaning. Additional crewmembers later sought medical evaluation for head, neck, back, and extremity pain. Post-flight, two FAs were diagnosed with fracture injuries and a third was diagnosed with a sub-cranial bleed. AIRCRAFT INFORMATIONThe airplane was a Boeing 777-222ER, manufacturing Serial Number 26942. The airplane was manufactured in 1997, delivered to United Airlines in July of 1997 and it held a transport category airworthiness certificate. The airplane was configured with 4 flight crew seats, 14 flight attendant seats, and 276 passenger seats.


Figure 4: Accident Airplane, N788UA (Courtesy of planespotters.net)

At the time of the event, the airplane had two deferrals from the minimum equipment list (MEL). Those deferrals were for the broadband antenna and the autoland system. METEOROLOGICAL INFORMATIONThe National Weather Service (NWS) Surface Analysis Chart, centered over the northeastern United States for 1600 EST, depicted the accident site was located east of a frontal boundary.
 
The Constant Pressure Charts for 1900 EST indicated a mid-level trough over the Great Lakes region. These charts are updated every 6 hours and the 1900 EST time is the closest to the accident time. Troughs and fronts can act as lifting mechanisms to help produce clouds and precipitation when sufficient moisture is present. The 700-hPa constant pressure chart showed westerly wind around 40 knots, becoming a southwest 50 knot wind by 500-hPa, and about 30,000 ft identified the jet stream maximum, which stretched from Michigan eastward into New York, with winds of 150 knots and was located in the vicinity of the accident site.
 
The accident occurred in an area where a chance of general thunderstorms had been predicted.
 
Observations from the BGM Automated Surface Observing System (ASOS), around the accident time, identified marginal visual flight rules (MVFR) to visual flight rules (VFR) conditions around the accident time with IFR conditions due to visibility restrictions when the thunderstorm activity went across BGM between 1353 and 1453 EST. During the period from 1423 through 1453, BGM reported lightning activity associated with the thunderstorms. In addition, winds gusted as high as 31 knots.
 
An HRRR model sounding was created for the approximate accident site coordinates for 1600 EST. The sounding indicated a conditionally unstable atmosphere from the surface through 4,000 ft, transitioning to a stable atmosphere above 4,000 ft. Convective Available Potential Energy (CAPE) was 16 Joules/kilogram (J/kg), with the maximum vertical velocity (MVV) for this atmosphere calculated as 6 meters/second (about 1,181 ft per minute). Rawinsonde Observation (RAOB) indicated cloud cover in two layers between 3,000 ft and 18,000 ft with another cloud layer near 25,000 ft. RAOB also indicated moderate rime and mixed icing between 6,500 ft and 15,000 ft.

Based on the 1600 EST HRRR sounding, wind speed increased to 40 knots by 10,000 ft and to 150 knots by 30,000 ft. Westerly winds increased from 45 knots near 17,000 ft to 75 knots by 19,000 ft, and to 105 knots by 21,000 ft. RAOB data indicated moderate or greater CAT between 18,000 ft and 21,000 ft. At the accident aircraft’s altitude around the time of the event (approximately21,000 ft) the temperature was -27.2°C, the dew point temperature was -47.2°C, and the wind was from 270° at 110 knots.
 
The GOES-16 imagery indicated there was cloud cover above the accident site, with cloud motion moving from west to east. The visible imagery showed a transverse banding within the cloud cover including above the accident site. Figure 1 shows the GOES-16 visible image at 1546 EST at 3x magnification with the accident site highlighted with a red square. The visible imagery indicated a transverse banding in the cloud cover including above the accident site. Figure 2 shows the GOES-16 infrared image at 1546 EST at 8x magnification with the accident site highlighted with a red square.

Cloud cover was indicated above the accident site with lower brightness temperatures (green colors; higher cloud tops) located above and to the northeast through south of the accident site. The brightness temperature of 251 Kelvin22 would have been near 17,000 ft based on the vertical temperature profile provided by the 1600 EST HRRR sounding. A lower brightness temperature of 245 Kelvin was noted in the higher altitude portions of the transverse banding near the accident site and would have corresponded to 23,000 ft based on the vertical temperature profile from the 1600 EST HRRR sounding. It should be noted these figures have not been corrected for any parallax error.

A review of the GOES-16 satellite imagery with Geostationary Lightning Mapper indicated lightning activity in the vicinity of the band of convective clouds moving across New York during the period.

The BGM Weather Surveillance Radar-1988 Doppler (WSR-88D) base reflectivity images for the 6.37° elevation scan initiated at 1540:30 and 1546:38 EST depicted portions of reflectivity values between -10 and 0 dBZ over the accident site location between 19,000 and 22,000 ft. See figures 5 and 6 The reflectivity targets were moving from west to east at the time of the accident.
 
There were no lightning flashes reported within 25 miles of the accident site within 10 minutes prior to and following the accident time.



Figure 5. BGM WSR-88D reflectivity for the 6.37° elevation scan initiated at 1540:30 EST with the accident site marked with the red circle and the location of the accident aircraft in the black circle. The pink dots are the airplane flight track points.


Figure 6. BGM WSR-88D reflectivity for the 6.37° elevation scan initiated at 1546:38 EST with the accident site marked with the red circle and the location of the accident aircraft in the black circle. The pink dots are the airplane flight track points.

A System Service Review (SSR) team conducted a review of the ATC audio between 1454 and 1623 EST for the ATC sector involved in the accident. During that period, 46 aircraft traversed the sector, 27 aircraft required a depicted weather call, and 16 aircraft were issued one correctly. Thirty six of the 46 aircraft required a PIREP to be issued and 26 aircraft received a PIREP as required. Seventeen PIREPs were received at the ATC sector during the timeframe and 8 of those were input and distributed correctly.
 
There was a convective significant meteorological information (SIGMET) valid for the accident site at the accident time. SIGMET 29E issued at 1455 EST forecast an area of embedded thunderstorms with tops to FL 280 with the SIGMET box area moving from 250° at 40 knots, Figure 7 below is a graphical representation of SIGMET 29E from the Aviation Weather Center’s (AWC) observation page. The approximate location of the accident has been outlined with a black circle.


Figure 7. Aviation Weather Center observation page valid at 1545 EST with fronts, PIREPs, and the area of Convective SIGMET 29E highlighted. The location of the accident site is marked by the black circle.

The NWS AWC had text Airmen’s Meteorological Information (AIRMET) advisory Tango and Zulu valid for the accident site at the accident time for above 15,000 ft. The text AIRMET Tango was issued at 1200 EST and forecast moderate turbulence between FL240 and FL420 until 1600 EST. The text AIRMET Zulu was issued at 1024 EST and forecast moderate icing between the freezing level and 17,000 ft until 1600 EST.
 
The Graphical Turbulence Guidance Nowcast (GTGN) data for 1545 EST along with National Center for Atmospheric Research (NCAR) in situ eddy dissipation rate (EDR) algorithm data from the accident aircraft was provided and its data indicated a peak EDR value of 0.431 was noted on the accident aircraft at 1543 EST corresponding to moderate to severe turbulence for a “heavy” class of aircraft.
 
The sector 24 en route automation modernization (ERAM) displayed the accident aircraft in an area of moderate to heavy precipitation between 1543 and 1546 EST. AIRPORT INFORMATIONThe airplane was a Boeing 777-222ER, manufacturing Serial Number 26942. The airplane was manufactured in 1997, delivered to United Airlines in July of 1997 and it held a transport category airworthiness certificate. The airplane was configured with 4 flight crew seats, 14 flight attendant seats, and 276 passenger seats.


Figure 4: Accident Airplane, N788UA (Courtesy of planespotters.net)

At the time of the event, the airplane had two deferrals from the minimum equipment list (MEL). Those deferrals were for the broadband antenna and the autoland system. ADDITIONAL INFORMATIONDispatch:
During an interview with the flight’s dispatchers, one had planned the flight earlier in the day; however, when her shift ended, she passed it to another dispatcher to monitor until it landed at EWR. The dispatcher that planned the flight recalled that on the day of the event they were responsible for releasing about 20 flights during their shift and that they were “busy fine-tuning routes.” She further provided that while a flight was airborne, they can monitor the progress in Fusion, which is a software program developed by The Weather Channel for global flight tracking, weather content, etc. She utilized various functions of the Fusion system including the FPG32 product, which provided turbulence forecast. Additionally, she stated that “pretty much from the Midwest to the East Coast the RPM33 data was showing turbulence.”

Prior to the end of her shift, she had PIREPs being reported as moderate turbulence on descent along the east coast. She had alerted the flights she was monitoring about the turbulence prior to the end of her shift; however, the event flight “was not even halfway across the U.S.” when her shift ended and she briefed and transferred the monitoring of the event flight to another dispatcher.

Following the outgoing dispatcher’s briefing, the incoming dispatcher reviewed the flights he was responsible for, which included the event flight. He reviewed the turbulence on the west coast of the US and also looked at the flights that were headed into EWR. He described the turbulence reports going into EWR as “some light to occasional moderate,” which he characterized as a “typical day.” After the event flight crew sent him an ACARS message about the turbulence and the flight attendant injuries, he began coordinating with emergency services to meet the flight upon arrival at EWR. INJURIES TO PERSONSAll nine cabin crewmembers (identified as FA 1 through FA 9) submitted written statements describing the turbulence encounter during initial descent into EWR. The crew reported that the fasten-seatbelt sign was illuminated and that they were conducting final cabin compliance checks and securing galley equipment at the time of the event. FA 1 reported that the flight deck had earlier advised that turbulence conditions had subsided and normal procedures could resume. 

Several flight attendants located in the forward and mid-cabin galleys (including FA 1, FA 4, FA 5, FA 6, and FA 7) reported feeling light or no turbulence immediately before the airplane experienced a sudden drop. Multiple crewmembers described becoming lifted from their positions, striking the cabin ceiling, and subsequently impacting the floor or galley structures. Crewmembers in the aft galley (primarily FA 2, FA 3, and FA 9) reported the most severe effects, with several sustaining head, back, and leg injuries, and at least one (FA 9) reporting a brief loss of consciousness. 
 
One aft-galley flight attendant (FA 3) reported being thrown upward and subsequently sustaining a shattered femur and spinal compression fractures. Another aft-galley crewmember (FA 9) reported facial trauma and was later diagnosed with intracranial bleeding. Several crewmembers in forward and mid-cabin locations (including FA 1, FA 4, FA 5, FA 6, and FA 7) also reported ceiling contact, head impact, and back and lower-extremity pain. 

Multiple crewmembers reported observing passengers being lifted from their seats, including a passenger in seat 8G reported independently by FA 1, FA 4, and FA 6 as having struck the ceiling. Another passenger reported being inside the lavatory at the time of the event (reported by FA 5 and FA 6) and sustaining neck and head injuries. At least one lap infant reportedly struck the ceiling (reported by FA 8). 

Several flight attendants (including FA 1, FA 2, FA 3, FA 5, FA 6, FA 7, and FA 8) reported crawling or otherwise moving along the cabin floor to reach jumpseats due to continued turbulence and physical injury. Following the turbulence encounter, the flight crew initiated a cabin crew check-in, and crewmembers reported multiple serious injuries to the flight deck. Despite the injuries, aft-cabin crewmembers reported attempting to cover assigned doors for landing.  

No flight attendant reported receiving prior notification from the flight deck to be seated immediately before the turbulence encounter. Several crewmembers (including FA 1, FA 2, FA 3, FA 6, FA 7, and FA 8) reported that the encounter was sudden, severe, and without preceding light or moderate turbulence. COMMUNICATIONSZBW ARTCC was a level 12 enroute ATC facility located in Nashua, New Hampshire, and was part of the FAA’s eastern service area. The radar facility was responsible for approximately 165,000 square miles of airspace that bordered six other ARTCCs including New York (ZNY), Cleveland (ZOB), Washington (ZDC), Montreal (ZUL), Moncton (ZQM) and Toronto (ZYZ). ZBW ARTCC was open 24 hours per day, 7-days per week, year-round. 
The recording timeline was constructed from certified audio recordings and ADS-B data were provided by the FAA. Times are in EST and altitudes below 18,000 feet are in feet above mean sea level (msl), and altitudes of 18,000 feet or greater are in FLs. 

Turbulence Related ACARS Messages:
Almost 2 hours prior to the turbulence encounter, the flight crew sent the following ACARS message “SMOOTH RIDE PAST HOUR OR SO HOWS DESCENT INTO EWR.” About a minute later the dispatcher responded “SO FAR ONLY COMPLAINT IS A MOD BY AN RJ”

A turbulence pilot report (PIREP) was sent to the flight via ACARS at 2041 UTC (1541 local) and reported “YOUR FLIGHT IS IN CLOSE PROXIMITY TO A MOD TURB PIREP EVENT REPORTED BY A B789 AT 2022 Z FL220 LOCATION HNK RMK TB MOD CHOP 220-190 YOU CLOSEST POINT TO EVENT AT 2041Z”. Additionally, at 1547, another ACARS message was sent to the flight crew which stated “YOUR FLIGHT ENTERS ATURBULENCE SIGMET AREAAT 2058Z

FL170LOCATION N4117.4 W07426.4ID TURB 30974ISSUED 10 FEB 1653ZVALID 10 FEB 1653ZTO 10 FEB 2100ZSIGMET AREA IS 70ESEAIR,15N SAX, 20ESEOWENZ,30SSE GVE,70ESE AIRMOVEMENT ENE AT 15KTSTYPE CATSEVERITY MODALTITUDE FL120 - FL180OUTLOOK NCDISCUSSION TURBULENCEREPORTED IN MID LEVELS OVER NEUS.“

At 1553 the accident pilots sent the following ACARS to United Airlines dispatch “PLZ HAVE MEDICAL MEET US HIT HDAVY MOD TURB ON DESCENT THRU FL210 FA BLEEDING ANOTHER MAY BE CONCUSSED.”

About two minutes after the pilots sent the ACARS message, United Airlines responded with “TWO FAS INJURED FROM TURB WILL HAVE EMS ON ARRIVAL.”

At 1611, ACARS indicated that the aircraft brake was set and two minutes later a passenger door was open.

At 1614, United Airlines dispatch sent the following message to the flightcrew “WHEN ABLE PLEASE CALL DISPATCH.” FLIGHT RECORDERSThe NTSB Vehicle Recorder Division received the QAR data from the N788UA airplane in an electronic file format.
The QAR recording contained approximately 4.79 hours of data. Timing of the recording was referenced using subframe reference numbers (SRN), with each SRN corresponding to one elapsed second. The event flight was the only flight contained within the recorded data file. 
The data indicated that between times 1542:35 – 1547:52 EST, the vertical acceleration fluctuated rapidly between -0.55 g and 1.82 g. Vertical acceleration reached a maximum of 1.82 g at 1542:52 and a minimum of -0.55 g at 1542:53 EST. The autopilot and autothrottle did not disengage. At 1547:53, the vertical acceleration began to stabilize around 1.00 g. The remainder of the flight was uneventful. Figures 3 shows a time history plot presenting acceleration and attitude control parameters. SURVIVAL ASPECTSAll injured parties were either not seated or did not have their seatbelt buckled at the time of the turbulence. Injuries were sustained from impacting the ceiling or overhead panel and falling back to the floor or seat during the turbulence event.