Boeing 737

Full Narrative

HISTORY OF FLIGHTOn December 6, 2018, about 0903 Pacific standard time, Southwest Airlines (SWA) flight 278, a Boeing 737-7H4, N752SW, overran the end of runway 8 during the landing roll at Bob Hope Airport (BUR), Burbank, California, and came to rest in the engineered materials arresting system (EMAS) at the departure end of the runway. The 2 pilots, 3 flight attendants, and 112 passengers were not injured. The airplane sustained minor damage. The regularly scheduled domestic passenger flight was operating under the provisions of Title 14 Code of Federal Regulations (CFR) Part 121. Instrument flight conditions prevailed at the time of the incident, and an instrument flight rules flight plan was filed.

The flight departed from Metropolitan Oakland International Airport, Oakland, California, about 0810. The captain was the pilot flying, and the first officer was the pilot monitoring.

About 0822, the flight crew requested, via the aircraft communications addressing and reporting system (ACARS), the weather at BUR and received automatic terminal information service (ATIS) information Hotel, which was issued at 0753. The ATIS information indicated that the wind was from 280° at 5 knots, visibility was 1 1/4 miles in heavy rain and mist, and the ceiling was broken clouds at 1,100 ft above ground level (agl).

About 0824, the crew received, via ACARS, a landing data report generated from the company’s performance weight and balance (PWB) system. The report was based on the reported wind from the most recent meteorological aerodrome report (290° at 5 knots) converted to bearing from magnetic north (278° at 5 knots), the airplane’s estimated landing weight (120,800 pounds), the planned flap setting (40°), and the reported runway condition (good). (Note: The wind direction in meteorological aerodrome reports and similar products is relative to true north, and the wind direction provided by air traffic control [ATC] and in ATIS reports is relative to magnetic north. At BUR, magnetic north is 12° east of true north.) The report indicated that maximum autobrakes should be used for landing on runway 8 (which was 5,802 ft in length) and that the stopping margin (that is, the distance between the location where the airplane should come to a stop, plus a 15% safety factor, and the end of the runway) would be 245 ft. The stopping margin calculation assumed, among other things, a touchdown at 1,500 ft from the runway threshold. According to the cockpit voice recorder (CVR), the flight crewmembers were concerned about the stopping distance given the wet runway, the tailwind conditions, and the runway length. Also, the captain and first officer discussed that, because of the runway conditions, the braking would be “pretty abrupt.”

At 0848:14, the CVR recorded a controller from the Los Angeles Air Route Traffic Control Center informing the flight crew about “moderate to heavy to extreme” precipitation between the airplane’s position at the time and BUR. The center controller transferred the flight to the next ATC facility, the Southern California Terminal Radar Approach Control; at 0851:27, the approach controller instructed the flight crew to descend the airplane to 8,000 ft and expect the “ILS [instrument landing system] Zulu” runway 8 approach to BUR. The first officer acknowledged the instructions and advised the controller that they had ATIS information Hotel. Upon reaching MIKEI, an initial approach fix for the approach, the controller cleared the flight crew to conduct the approach. Between 0854:37 and 0854:51, the controller advised the flight crew that, about 10 minutes earlier, the pilot of King Air airplane reported a 15-knot loss of airspeed on final approach to BUR and that a corporate jet had just conducted a goaround at BUR. About 2 minutes later, the controller informed the flight crew that the goaround occurred because of the wind. The first officer acknowledged this information.

According to the company’s PWB system, the reference landing speed was 126 knots (indicated airspeed), and the target speed was 131 knots. Flight data recorder (FDR) data showed that, about 0858, the airplane turned onto its final heading and captured the glideslope and localizer for the ILS approach. About 1 minute later, the flight crew selected a flap setting of 40°, and the airplane’s indicated airspeed was between 133 and 136 knots.

At 0858:54, the approach controller instructed the flight crew to contact the BUR ATC tower, and the first officer acknowledged this instruction and made initial contact with the tower at 0859:17. Between 0859:21 and 0859:36, the tower controller advised the flight crew about “moderate to heavy precipitation” between the flight’s position at the time and the airport, stated that the wind was 260° at 9 knots, and cleared the airplane to land on runway 8. The controller also advised the flight crew that, 10 minutes earlier, the pilot of a Boeing 737 airplane reported braking action as “good.” Less than 1 minute later, the controller further advised that heavy precipitation was directly over the airport. The first officer acknowledged each of the transmissions from the controller.

At 0900:43, the controller stated that the wind was from 270° at 10 knots. Five seconds later, the first officer stated to the captain that 10 knots “is the max,” referring to the 10-knot tailwind limitation for landing in the SWA B737NG Aircraft Operating Manual (AOM); the captain also stated “max” and noted that the autopilot “has got a handle on it” thus far. At 0901:33, the captain asked that the windshield wipers be set to high. Three seconds later, the controller stated that the wind was from 270° at 11 knots. At 0901:52, the first officer stated, “we got eleven knots. You want to call it good?” The captain replied, “yeah.” During postincident interviews, the captain stated that he and the first officer agreed that the tailwind component would be about 9 to 10 knots and thus within the limit, and the first officer described the 11-knot wind as being “right on the edge” of the tailwind limit.

During a postincident interview, the captain reported that, when the airplane was about 400 ft agl, the airplane exited the clouds, and he had a clear view of the runway. The captain also reported that, at the decision altitude of 1,007 ft mean sea level (about 280 ft agl), he disconnected the autothrottles and the autopilot and continued to descend using the precision approach path indicator lights as a descent path reference. (The CVR recorded the captain stating, “we need some kind of lights. I got some lights,” about that time.) The captain further reported that, during the landing flare (when the airplane was about 5 ft agl), the rain intensity “picked up and visibility decreased a little bit.”

The National Transportation Safety Board’s aircraft performance study for this incident found that the airplane crossed the runway threshold at an altitude of 54 ft agl and an indicated airspeed of 137 knots. (The ILS runway 8 approach procedure noted a threshold crossing height of 60 ft when following the 3° glideslope.) The captain reported (during a postincident interview) that he was aware of the 1,000ft markers as the airplane crossed the runway threshold. FDR data showed that the airplane touched down at 0902:35 at an indicated airspeed of 126 knots and a groundspeed of 151 knots.

During postincident interviews, the captain estimated that the airplane touched down between 1,300 and 1,500 ft from the runway threshold based on the timing from about 10 ft agl to touchdown (and not runway markings). The first officer stated that the airplane touched down within the touchdown zone, which he defined as 1,000 to 1,500 ft from the threshold, and was “pretty confident” that the airplane touched down by 1,500 ft. The aircraft performance study found that the airplane touched down 2,504 ft from the runway threshold. A controller in the BUR tower observed the airplane touch down near the taxiway D7 intersection with runway 8, which was about 2,600 ft from the runway threshold. This observation was consistent with the related finding from the aircraft performance study.

According to the FDR, the airplane’s autobrakes, auto speedbrakes, and ground spoilers were activated at touchdown. Both thrust reversers were fully deployed within 3 seconds of touchdown. Brake pressure reached its maximum of 3,000 psi in about 6 seconds. At 0902:45, the CVR recorded the first officer stating, “come on baby,” which was followed by the captain stating “whoa-ah” and then expletives. The CVR recorded sounds of impact at 0902:56.

The captain stated that, after touchdown, he used maximum reverse thrust and that he heard and felt the brakes “chattering.” He verified that the speedbrake handle was extended and started “blending in manual brakes after that” because the airplane was not slowing with reverse thrust and maximum autobrakes. The captain became concerned about the airplane stopping as it passed runway 15/33 (which intersected with runway 8 about 3,765 ft from the runway 8 threshold). The captain noted that, with 1,000 to 500 ft remaining on the runway, the airplane stopped decelerating. The captain thought that he might be able to make a turn onto taxiway D1 (which intersected runway 8 about 5,340 ft from the runway threshold), but, as he tried to turn using left rudder and the tiller, the airplane did not respond. The captain reported that the nosewheel was skidding and that he straightened it so that the airplane would enter the EMAS without the nosewheel turned sideways.

FDR data showed that, during the first 17 seconds after touchdown (between 0902:35 and 0902:52), the airplane reached and maintained a deceleration rate between about 0.3 and 0.4 G, and the airplane’s groundspeed decreased from 151 to 50 knots. During the next 5 seconds, the deceleration rate decreased to between 0.15 and 0.20 G as the airplane crossed over the left edge of the grooved runway onto smooth pavement, and the groundspeed slowed to about 24 knots at 0902:57. The deceleration rate then increased to a maximum of about 0.6 G from the time that the airplane entered the EMAS to the time that the airplane came to a stop. The airplane came to rest about 71 ft into the EMAS, 144 ft past the end of runway 8, and 107 ft to the left of the runway centerline. The crewmembers and passengers evacuated the airplane via airstairs.
PERSONNEL INFORMATIONThe captain reported that he flew into BUR typically every month and estimated that he had landed there between 80 and 100 times. The first officer reported that he had flown into BUR at least 100 times.
METEOROLOGICAL INFORMATIONBUR had an automated surface observing system (ASOS) that was augmented by ATC personnel. The ASOS is the official weather reporting source at the airport. ATIS and ATC reports provide ASOS wind information that has been converted from true north to magnetic north for runway orientation.

The 0853 meteorological aerodrome report (about 10 minutes before the incident) is shown in the data field above. The 5-minute observation at 0905 (about the time of the incident) reported that the wind was from 280º at 13 knots, visibility was 1 1/4 miles in heavy rain and mist, and the hourly precipitation since the 0853 observation was 0.13 inch. The average 2minute wind reported by the ASOS at 0905 was from 279º at 13 knots with the peak gust at 18 knots in heavy rain. The special weather observation that was issued at 0915 (about 12 minutes after the incident) stated that the wind was from 290° at 8 knots, visibility was 1 mile in heavy rain and mist, and the hourly precipitation since the 0853 observation was 0.26 inch.

The captain reported that he had not previously encountered conditions such as those on the day of the incident; his previous flights into BUR occurred in visual flight rules flight and without any significant precipitation, although he had occasionally landed with a tailwind on runway 8. The first officer reported that he did not usually encounter weather at BUR that was similar to that during the incident flight. The first officer also reported that he had previously landed at BUR with “a bit of a tailwind” and with rain. The first officer estimated that he encountered tailwind or rain conditions during 5% to 10% of the landings there.
WRECKAGE AND IMPACT INFORMATIONAccording to SWA, the following components sustained minor damage as a result of this incident: both engines, both engine inboard c-ducts, the air conditioning heat exchangers, the left engine inlet cowl, the left engine inboard fan cowl, three pitot probes, two angle-of-attack probes, and a total air temperature probe.
ADDITIONAL INFORMATIONSWA took several actions as a result of this incident. On May 6, 2019, SWA issued a flight safety alert for BUR regarding the runway 8 touchdown point. The safety alert stated the following:

RWY 08 is one of the shortest runways (5,802’) in our system. For valid PWB stopping margins, ensure that touchdown occurs within the first 1,500’ of the runway. If touchdown occurs beyond the 1,500’ mark, the ability to stop on the remaining runway may be compromised and a go-around is the better option.

The safety alert also stated the following:

The color contrast between the concrete and asphalt (approx. first 500’ of RWY 8) may cause a visual illusion, which could affect your touchdown aim point. In addition to the painted runway markings, another visual reference to ensure a timely touchdown is to use TWY [taxiway] D8/TWY C8, which is approximately 1,400’ past the threshold of RWY 08.

On November 27, 2019, SWA issued a flight operations bulletin that discussed that a go-around would be required if an airplane touched down beyond 1,500 ft with an insufficient PWB system-computed stopping margin. The bulletin defined “touchdown zone” as “the area of the runway 500 ft to 3,000 ft beyond the landing threshold not to exceed the first one-third of the runway”; according to the bulletin, this definition aligned with regulatory and manufacturer guidance.

The bulletin stated that the definition for touchdown zone should not be confused with AOM guidance regarding the planned touchdown point of 1,000 to 1,500 ft from the runway threshold (consistent with the AOM version that was current at the time of the incident). Further, the bulletin stated the following:

A go-around must be executed if the aircraft touches down beyond the most restrictive of the following distances:

First one-third of the available runway length.
First 3,000 ft of the available runway length.
1,500 ft plus the planned PWB System-computed stopping margin.

In addition, on June 17, 2020, SWA reported that it took actions, including the following, to mitigate the current and potential risks identified after the incident:

Incorporated the above-referenced flight operations bulletin into the Flight Operations Manual revision that became effective in March 2020.

Emphasized aeronautical decision-making under rapidly changing conditions in its 2019 distance learning.

Asked company check airman to add runway landing distances and runway remaining distances to line checks.

Included, in the 2020 pilot continuing qualification training, ground school discussions about plan continuation bias and rapidly changing conditions and emphasized, in maneuvers observation briefings, runway markings and PWBcomputed stopping margins.

Incorporated enhanced landing exercises into the 2020 line-oriented evaluation briefing to provide a better understanding of (1) the definition of the touchdown zone, including how it relates to landing or go-around decision-making; (2) the difference between the planned touchdown point and the touchdown zone; (3) the relationship of PWB data to the touchdown point, landing distance required, stopping margin, and distance remaining; and (4) the effect that “less than optimum control” of variables such as airspeed, glidepath, and landing flare can have on landing distances and runway remaining distances.
MEDICAL AND PATHOLOGICAL INFORMATIONAfter the incident, the captain and the first officer were tested for drugs and alcohol in accordance with 14 CFR Part 120, Drug and Alcohol Testing Program. The test results were negative.
TESTS AND RESEARCHThe aircraft performance study for this incident found that the airplane’s indicated airspeed over the runway 8 threshold was 137 knots—the reference landing speed (Vref) of 126 knots (as calculated by the PWB system) plus 11 knots. As stated in the Organizational and Management Information section, the target speed for tailwind landings was Vref plus 5 knots; thus, the airplane’s 137-knot airspeed over the runway threshold was 6 knots above the target airspeed for the incident landing (and the target airspeed calculation provided by the PWB system for the flight).

The airplane touched down about 10 seconds later at an indicated airspeed of 126 knots (Vref). The touchdown point was 2,504 ft past the runway threshold, as determined from FDR latitude, longitude, and weight-on-wheels data. (The result was also consistent with the sound similar to landing gear touchdown that the CVR recorded at 0902:35.) The touchdown point was about 1,000 ft beyond the nominal 1,500-ft touchdown point assumed in the PWB system calculation.

According to the true airspeed and the groundspeed computed from FDR data, a tailwind of 13 to 18 knots—not the 5-knot tailwind assumed in the PWB system calculation—was over the runway during the ground roll. When the airplane crossed the runway 8 threshold, the true airspeed was 140 knots, and the groundspeed was 156 knots, corresponding to a 16-knot tailwind. When the airplane touched down, the true airspeed was 133.5 knots, and the groundspeed was 151 knots, corresponding to a 17.5-knot tailwind. The study noted that the calculated tailwind was consistent with the wind recorded by the ASOS on the field but that the tailwind that ATC reported to the crew during the flight ranged between 5 and 11 knots.

The ground roll distance that the PWB system calculated was 3,145 ft. Runway 8 was 5,802 ft in length, so the touchdown point that would have resulted in the airplane stopping with no runway remaining was 2,657 ft from the threshold. Even though the airplane touched down about 150 ft before that point, the higher airspeed and greater tailwind than those assumed in the PWB system calculations prevented the airplane from stopping within the systemcomputed 3,145-ft ground roll distance.

The study concluded that the higher-than-expected tailwind, the longer-than-normal touchdown point, and the faster-than-nominal approach speed all contributed to the overrun, with the long touchdown point being the most significant contributor. The study found that the airplane could have stopped on the runway with the faster-than-nominal approach speed and the tailwind even if the touchdown point had been up to 312 ft longer than the normal touchdown point of 1,500 ft from the runway threshold. Because the touchdown point was 2,504 ft from the threshold, either the tailwind or the faster approach speed by itself would have prevented the airplane from stopping before the end of the runway.

The study also examined the airplane’s braking performance on the runway. The study found that the right wheel brake pressure was steady near 0 psi until 0902:34 and then increased to about 600 psi at 0902:35, 800 psi at 0902:39, and 3,000 psi (corresponding to full manual braking) at 0902:41, where the brake pressure remained until 0902:58 (1 second before the airplane came to rest). The left wheel brake pressure was steady near 400 psi until 0902:36 and increased steadily, reaching and maintaining 3,000 psi about the same time as the right brake pressure.

In addition, the study found that the FDR “autobrake applied” parameter changed from zero to one at 0902:35.6 and then changed back to zero at 0902:36.6, indicating that the autobrake system engaged after touchdown but then disengaged almost immediately as a result of pilotapplied pressure to the brake pedals. The study concluded that the maximum braking performance available from the runway was achieved despite the 6-second duration between the time of touchdown and the time that the left and right brakes reached their maximum brake pressure.
ORGANIZATIONAL AND MANAGEMENT INFORMATIONPerformance Weight and Balance System

SWA implemented its PWB system in two phases. The first phase, which involved the PWB landing module (for use in calculating landing data), began in late 2016. Full use of the system began in 2017.

According to the AOM, the performance weight and balance system “is a real-time performance and weight and balance data computation system used to produce data required to operate an aircraft within the limitations specified in the Airplane Flight Manual CFRs.” The system integrates five technology platforms, including PWB data centers, which provide realtime calculations that flight crews use to obtain, via ACARS, time-of-arrival landing performance information. The PWB system also produces landing data reports for flight crews to use during a flight to prepare for landing. Such reports provide crews with the information necessary to determine the suitability of the planned landing runway based on various factors, including the speeds for the approach and landing and the stopping margin for the selected runway. A safety factor of 15% is added to the PWB system-calculated distance for the airplane to come to a stop. The stopping margin is the difference between the calculated landing distance (including the 15% safety factor) and the runway length available. Although the PWB system will automatically use the wind information from the current meteorological aerodrome report when flight crews request landing data from the system, crews have the option to manually enter the wind information.

During a postincident interview, the captain stated that he requests an updated data report from the PWB system if the runway condition changes from dry to wet or the intended landing runway changes. The captain also stated that flight crewmembers can request multiple landing data reports from the PWB system but noted that, in the case of a wind change, crewmembers might receive the same information if the weather information that the PWB system uses has not been updated. In addition, the captain stated that he had not previously received a stopping margin from the PWB system that was as low as the one (245 ft) for the incident flight. The first officer stated that BUR “always has a low [PWB system] number for stopping margin,” which he considered to be a concerning but not a significant issue.

In-flight Weather Information

The SWA Flight Operations Manual stated that the pilot monitoring (the first officer for this flight) should obtain ATIS information before reaching the top of descent. The dispatch release for this flight showed that the top of descent would occur about 21 minutes before landing. The FDR showed that the airplane began its descent from cruise altitude about 0839, which was 17 minutes after the flight crew had received ATIS information Hotel and about 23 minutes before landing.

According to SWA, the following ATIS reports were also in effect during the incident flight:

ATIS information India, which became effective at 0841;
ATIS information Juliet, which also became effective at 0841 and contained the same information as the previous ATIS along with a windshear advisory; and
ATIS information Kilo, which became effective at 0853 and included a windshear advisory.

The ATC transcript for this incident showed that, about 0856, another SWA flight checked in with the Southern California Terminal Radar Approach Control while inbound to BUR, and the controller advised that flight crew that ATIS information Juliet had just become current. The incident flight crewmembers were on the same frequency at the time, but the available evidence did not indicate whether they heard this transmission. (The CVR transcript did not reflect the transmission.) The CVR transcript showed that the controller did not notify the incident flight crew when ATIS information Juliet and Kilo had become effective. During postincident interviews, both the incident captain and first officer stated that they were aware of the last wind report from the controller (270° at 11 knots), which they received about 1 minute before touchdown.

Approach Speed Information

According to the AOM, the PWB system defines the target speed as Vref plus one-half the headwind component plus the full gust component and sets a minimum value of Vref + 5 knots. A note in this part of the manual states that, for tailwind landings, the target speed should be Vref  + 5 knots. Another note states, “the purpose of the +5 knots is to ensure that V2 [the takeoff safety speed] is met in case of a go-around. Under adverse conditions, the Captain may elect to fly at VREF.”

Normal Landing Procedures

The AOM instructed the pilot flying to “touch down between 1,000 and 1,500 ft from the landing threshold on centerline” and stated the following:

PWB stopping margin information is based on the assumption that aircraft touchdown occurs not later than 1,500 ft from the usable end of the runway. If the aircraft lands beyond the 1,500-ft mark, the PWB System-computed stopping margin will be invalid, and in some cases, the runway length will be insufficient to stop the aircraft.

If touchdown occurs beyond 1,500 ft, the ability to stop on the remaining runway may be compromised. Also, if the current conditions are significantly different than the anticipated conditions at the time of PWB programming…the PWB System-computed stopping margin may be invalid…[and] higher than planned braking may be needed to account for the reduced or insufficient stopping margin. The situation becomes more critical on shorter runways; therefore, a go-around is the better option. A go-around is possible until the thrust reverser levers are raised.

Use of Autobrakes

The AOM stated that autobrake use was required if the landing data report from the PWB system specified that they be used (which was the case for the incident flight). The AOM also provided the following guidance regarding the transition from autobrakes to manual braking:

The intent of using the autobrake system for landing is to let the system automatically brake the aircraft to an appropriate speed, not to override the system shortly after touchdown. Autobrakes relieve the PF’s [pilot flying] workload by automatically initiating wheel braking at touchdown and maintaining an appropriate deceleration rate throughout the landing roll. This allows the PF to concentrate on speedbrake deployment, reverse thrust application, directional control, and overall stopping performance. Once the landing roll is stabilized, transition to manual brakes by overriding the autobrake system. The speed at which this transition is accomplished will vary with runway and environmental conditions. With adverse conditions, transition at a slower speed. At the appropriate speed, apply and maintain brake pedal pressure to override the autobrake system.

Go-Around/Missed Approach Requirements

According to the AOM, the pilot flying must execute a go-around/missed approach when, among other reasons, “the pilot determines that a landing in the touchdown zone cannot be safely accomplished because…the aircraft touches down beyond 1500 ft. with an insufficient PWB System-computed stopping margin.” The AOM also advised pilots to monitor wind information during the final approach phase and not to continue the approach if known windshear exists.

The SWA Flight Operations Manual instructed pilots to “maintain a stable approach path” and “if stabilized approach criteria are not met, execute or direct a go-around/missed approach.” The manual also stated that “momentary deviations of glidepath, course, airspeed, and sink rate do not require an immediate go-around.” In addition, the manual stated, “anytime the approach or landing appears unsafe, execute or direct a go-around/missed approach.”