2. Analysis

Table of Contents

2.1 General

Table of Contents

The flight and cabin crewmembers were properly certificated and qualified under federal regulations. No evidence indicated any preexisting medical or physical condition that might have adversely affected the flight crew’s performance during the accident flight.

The accident airplane was equipped, dispatched, and maintained in accordance with federal regulations.

The LGA departure controller chose to display only correlated targets on his radar display. If he had chosen to display both the correlated and uncorrelated targets, he would not have been able to effectively control traffic because a large amount of extraneous information would have been shown on the display. Additionally, he would not have been able to determine whether the additional targets were birds, boats, precipitation, or any other item the radar detected. Therefore, the NTSB concludes that the LGA departure controller’s decision to display only correlated primary radar targets on his radar display was appropriate.

Examinations of the recovered components revealed no evidence of any preexisting engine, system, or structural failures. The airplane met the structural ditching certification regulations in effect at the time of its certification, and the engine met the bird-ingestion certification regulations in effect at the time of its certification, as well as an anticipated additional regulation that it was not required to meet at that time.

The investigation determined that the airplane’s descent rate at the time it impacted the water was 12.5 fps, more than three times the descent rate of 3.5 fps assumed during ditching certification, resulting in external pressures on the aft fuselage, primarily from FR55 to FR70, which significantly exceeded the values established to demonstrate compliance with the certification criteria. These external pressures were sufficient to cause the large-scale collapse and failure of the aft fuselage frames, cargo floor, and passenger floor struts and to initiate cracking of the lower fuselage skin, allowing water to enter the airplane. Further, the water ingress and continued forward motion of the airplane through the water resulted in postimpact pressures and suction forces that caused additional damage, including the failure of the lower fuselage skin panel and aft pressure bulkhead. Therefore, the NTSB concludes that the airframe damage was caused by the high-energy impact at the aft fuselage and the ensuing forward motion of the airplane through the water.

The term, “ditching,” is not defined in federal regulations. The NTSB addressed this issue previously in its Safety Study 85/02, “Air Carrier Overwater Emergency Equipment and Procedures,” which stated the following:

[ditching] usually means a planned water event in which the flight crew, with the aircraft under control, knowingly attempts to land in water. In contrast to an inadvertent water impact, in which there is no time for passenger or crew preparation, ditching allows some time for donning life preservers, etc.

The NTSB considers this accident to be a ditching because the pilots clearly intended to ditch on the Hudson River. The accident event falls between a planned and unplanned event in that, although the pilots did not have time to complete each step of the applicable checklist, they did have sufficient time to consult the QRH, begin checklist execution, transmit radio calls, determine a landing strategy, configure the airplane for the ditching, and alert the flight attendants and passengers to “brace for impact.”

Although the airplane impacted the water at a descent rate that exceeded the Airbus ditching parameter of 3.5 fps, postaccident ditching simulation results indicated that, during an actual ditching without engine power, the average pilot will not likely ditch the airplane within all of the Airbus ditching parameters because it is exceptionally difficult for pilots to meet such precise criteria with no power. Further, the water swell tests conducted on Mercure airplanes indicated that, even with engine power, water swells and/or high winds also make it difficult for pilots to safely ditch an airplane, and these factors were not taken into account during certification. (See section 2.6 for a more detailed discussion of this issue.)

Although both engines experienced an almost total loss of thrust after the bird encounter, the flight crew was able to ditch the airplane on the Hudson River, resulting in very few serious injuries and no fatalities. Further, all of the airplane occupants evacuated the airplane and were subsequently rescued. Consequently, this accident has been portrayed as a “successful” ditching. However, the investigation revealed that the success of this ditching mostly resulted from a series of fortuitous circumstances, including that the ditching occurred in good visibility conditions on calm water and was executed by a very experienced flight crew; that the airplane was EOW equipped even though it was not required to be so equipped for this particular flight; and that the airplane was ditched near vessels immediately available to rescue the passengers and crewmembers. The investigation revealed several areas where safety improvements are needed.

The analysis discusses the flight crew performance and safety issues related to the following: in-flight engine diagnostics, engine bird-ingestion certification testing, emergency and abnormal checklist design, dual-engine failure and ditching training, training on the effects of flight envelope limitations on airplane response to pilot inputs, validation of operational procedures and requirements for airplane ditching certification; and wildlife hazard mitigation. Also analyzed are survival-related issues, including passenger brace positions; slide/raft stowage; passenger immersion protection; life line usage; life vest stowage, retrieval, and donning; preflight safety briefings, and passenger education.


2.2 Engine Analysis

Table of Contents

2.2.1 General

FDR data indicated that, during ground operation and takeoff, the N1 and N2 speeds of both engines accelerated in unison during the throttle advancement to full takeoff power and that these speeds were similarly matched and stable during takeoff and initial climb until about 1 minute 37 seconds into the flight. Although the right engine had recently experienced an engine compressor stall, US Airways had corrected the problem in accordance with maintenance manual practices, and no FDR evidence indicated that a compressor stall occurred before the bird encounter.

2.2.2 Identification of Ingested Birds

The Smithsonian Institution analyzed the feather and tissue samples from both engines and determined that the left engine contained both male and female Canada geese remains, indicating that the engine ingested at least two geese. (The average weight of a male Canada goose is from 8.4 to 9.2 pounds, and the average weight of a female goose is from 7.3 to 7.8 pounds.) The Smithsonian Institution report stated that only male Canada goose remains were found in the right engine, suggesting that it might have only ingested one bird; however, a comparison of the physical features and quantity of the damage in the two engines, which will be discussed in the following sections, indicated that the right engine ingested at least two Canada geese.132

2.2.3 Engine Damage

The engines were certificated to withstand the ingestion of birds of a specified weight in accordance with the certification standards and still produce sufficient power to sustain flight. (The certification requirements are discussed in section 2.2.5.) However, during this event, each engine ingested at least two Canada geese weighing about 8 pounds each, which significantly exceeded the certification standards, and neither engine was able to produce sufficient power to sustain flight after ingesting these birds. This section will discuss the progression of the damage to the engine parts, starting with the engine spinners and moving to the cores, to explain at which point in the bird-ingestion sequence the damage occurred that prevented the production of sufficient power to sustain flight.

2.2.3.1 Engine Spinner, Fan Blade, and Fan Inlet Case Damage

If a bird enters the engine inlet near the inner radius (near the spinner), a portion of it may be ingested by the engine core because of the radius’ proximity to the core. Both engine spinners on the accident airplane exhibited soft-body impact damage, indicating that both engines ingested a bird very near the inner radius of the engine inlet and that some of that bird mass entered the engine core. Although all of the left and right engine fan blades were present and intact, three of the left engine fan blades and five of the right engine fan blades exhibited damage indicating that both engines ingested a second bird near the fan midspan, but, because it was ingested near the edge of the fan blades, none of that bird mass entered the core.

When a turbofan engine ingests birds and no fan blades are fractured, the damage to the fan blades is generally localized because the bird will affect only those fan blades that actually impact or slice it as it passes through the fan plane. The number of fan blades affected by the impact is determined by the bird size, the relative bird velocity with respect to the airplane, the rotational fan speed, and the bird-impact angle. As the fan blades impact and slice the bird, the impact forces against the fan blades can be high enough to permanently deform and twist them as they bend and vibrate in response to the impact. Although the fan blades of both engines showed evidence of bird ingestion and subsequent mechanical damage, as noted, no significant fan blade damage or fractures were found.

Gouging was found on both engines’ forward acoustical panels in the fan inlet case. Turbofan engine fan blades are designed to accelerate only compressible materials, such as air. When rotating fan blades contact a denser, noncompressible material, such as water, they will “bite” into the water, which will cause the blades to bend forward and cause gouging. Therefore, the fan rotors of both engines were rotating upon water impact.

2.2.3.2 Engine Core Damage

Because the fan spins rapidly, the fan blades protect the engine core by centrifugally slinging foreign objects outward into the bypass duct; therefore, most foreign objects that enter the engine inlet strike the fan blades and exit through the bypass duct, causing only fan blade damage. The spinner shape is also designed to deflect foreign objects outward to the bypass duct. However, only foreign objects of a limited size and consistency can be centrifuged or deflected from the engine’s core.

Disassembly and examination of the engines revealed that two LPC IGVs in each engine had fractured because of the bird ingestion and were subsequently ingested into the engine cores, where they initiated secondary damage to the LPC and HPC. Immediately thereafter, the engine cores were incapable of supplying power to the fans; therefore, the fans could no longer rotate and produce sufficient thrust to sustain flight.

In addition, damage to the left engine HPC VGVs resulted in the blockage of most of the airflow through the compressor. The insufficient airflow into the combustor to cool the engine and through the LPT to drive the fan resulted in the loss of left engine power. Although the airflow was not blocked in the right engine as it was in the left engine, the destruction of all of the HPC VGVs and the fracture of several compressor blades caused the loss of directional control of the airflow into the compressor, causing it to stall continuously, with no recovery possible, and, eventually, to lose power.

In summary, the NTSB concludes that both engines were operating normally until they each ingested at least two large birds (weighing about 8 pounds each), one of which was ingested into each engine core, causing mechanical damage that prevented the engines from being able to provide sufficient thrust to sustain flight.

2.2.4 In-Flight Engine Problem Diagnostics

FDR data indicated that, although the engine power and fuel flow decreased immediately after the bird ingestion, both engines’ LPC spools continued to rotate, and no loss of combustion occurred. According to FDR and CVR data, after the bird ingestion, the first officer followed the Engine Dual Failure checklist and spent about 30 to 40 seconds trying to relight the engines; however, since engine combustion was not lost, these attempts were ineffective in that they would not fix the problem, and the N2 speeds could not increase during the remainder of the flight. The flight crew was unaware that the extent and type of the engine damage precluded any pilot action from returning them to operational status. If the flight crewmembers had known this, they could have proceeded to other critical tasks, such as completing only the Engine Dual Failure checklist items applicable to the situation. (See section 2.3.1 for information about the Engine Dual Failure checklist and the flight crew’s accomplishment of it.)

The NTSB notes that it is unreasonable to expect pilots to properly diagnose complex engine problems and take appropriate corrective actions while they are encountering an emergency condition under critical time constraints. Many modern engines are equipped with engine sensors and full-authority digital engine controls (FADEC) that can be programmed to advise pilots about the status of an engine so that they can respond better to engine failures.

However, currently, no commercially available engines have diagnostic capabilities to identify the type of engine damage (sensors and FADECs can only identify that a problem exists) and recommend mitigating or corrective actions to pilots; yet, work has been performed to develop this technology for both military and civilian applications. For example, in 1998, the Department of the Navy, in conjunction with industry and the FAA, initiated the Survivable Engine Control Algorithm Development project, which was tasked, in part, to develop technology that would inform flight crews about an engine’s condition following foreign-object or bird ingestion that resulted in engine gas path damage. The intent was to use existing engine sensors to define the type of engine damage and then apply appropriate mitigation through changing control schedules within the FADEC. Although a successful demonstration of this technology was conducted on the U.S. Navy’s GE F414 turbofan engine, the project was terminated because of a lack of funding. In 2007, similar work was conducted on the GE T700 turboshaft engine; however, this project was also terminated before it was completed because of funding shortfalls.

Commercial applications for this type of technology were investigated in 2002 by NASA’s Aviation Safety and Security Program, which initiated the CEDAR (Commercial Engine Damage Assessment and Reconfiguration) project using a GE CF6-80C2 engine to develop damage detection algorithms. Again, initial efforts were terminated because of a lack of funding and shifted priorities.

The NTSB concludes that, if the accident engines’ electronic control system had been capable of informing the flight crewmembers about the continuing operational status of the engines, they would have been aware that thrust could not be restored and would not have spent valuable time trying to relight the engines, which were too damaged for any pilot action to make operational. Therefore, the NTSB recommends that the FAA work with the military, manufacturers, and NASA to complete the development of a technology capable of informing pilots about the continuing operational status of an engine. The NTSB further recommends that, once the development of the engine technology has been completed, as asked for in Safety Recommendation A-10-62, the FAA require the implementation of the technology on transport-category airplane engines equipped with FADECs.

2.2.5 CFM 56-5B4/P Bird-Ingestion Certification Tests

Each accident engine ingested one 8-pound bird into its core, preventing the engines from providing sufficient thrust to sustain flight, indicating that an engine of this size cannot withstand the ingestion of such a large bird into the core and continue to operate. Further, informal discussions with industry and the FAA revealed that it would not be practical to build an engine that could withstand ingesting a bird of this size into the core because of performance and weight penalties that such a design would entail. These discussions also revealed that ingesting one 2 1/2-pound bird into the engine core, which is the current engine core ingestion test requirement, is already considered a stringent test of the engine core.

The NTSB concludes that the size and number of the birds ingested by the accident engines well exceeded the current bird-ingestion certification standards.

The accident event highlighted other considerations that could be addressed during the tests related to small, medium, and large flocking birds. These considerations are discussed below.

The test requirements contained in 14 CFR 33.76(c) for the ingestion of small and medium flocking birds require that, for an engine of this size, one 2 1/2-pound bird be volleyed into the core and four 1 1/2-pound birds be volleyed at other locations on the fan disk. Each accident engine ingested one 8-pound Canada goose through to its core, much more than the weight used in the current certification tests; therefore, the accident engines sustained a significantly greater impact force than that for which they were certificated. FDR data indicated that the fan speed of both engines just before the bird ingestion was only about 80 percent, which is consistent for the airplane and atmospheric conditions at that point in the flight and is well below the bird-ingestion test fan-speed requirement of 100 percent.

Current Section 33.76(c) small- and medium-flocking-bird certification tests require that 100-percent fan speed be used; this condition involves the highest kinetic energy of the bird relative to the fan blade, which is likely the most critical condition for damage to the fan blade itself. However, an additional consideration for the severity of a core ingestion event is the volume or bird mass. Therefore, the lowest operational fan speed should be used during the tests related to small, and medium, flocking birds so that a larger portion of the bird mass passes through the fan blades. Additionally, a slower fan speed would cause less centrifuging of the bird mass as it passes through the fan, which would allow a larger portion of the bird mass to pass through to the IGVs and other core components, causing higher impact forces on them. Reducing the fan speed during the certification tests to that expected during takeoff conditions would allow more bird mass to enter the engine core.

The NTSB concludes that the current small and medium flocking bird tests required by 14 CFR 33.76(c) would provide a more stringent test of the turbofan engine core resistance to bird ingestion if the lowest expected fan speed for the minimum climb rate were used instead of 100-percent fan speed because it would allow a larger portion of the bird mass to enter the engine core. Therefore, the NTSB recommends that the FAA modify the 14 CFR 33.76(c) small and medium flocking bird certification test standard to require that the test be conducted using the lowest expected fan speed, instead of 100-percent fan speed, for the minimum climb rate. Further, the NTSB recommends that EASA modify the small and medium flocking bird certification test standard in JAR-E to require that the test be conducted using the lowest expected fan speed, instead of 100-percent fan speed, for the minimum climb rate.

Current Section 33.76(d) large flocking bird certification tests require the ingestion of one large flocking bird. However, during this test, the bird is not directed into the core; therefore, only the fan blades, flammable fluid lines, and support structure are tested. Further, the test is limited to engines with inlet areas greater than 3,875 square inches; smaller transport-category airplane engines, such as the CFM56-5B4/P, with an inlet area of 3,077 square inches, are exempt from this test. The evidence from this accident shows that large flocking birds can be ingested into smaller transport-category airplane engines and pose a threat to the engine core as well as the fan blades; however, the large flocking bird tests are not required as part of the certification process for this size engine.

The NTSB concludes that additional considerations need to be addressed related to the current 14 CFR 33.76(d) large flocking bird certification test standards because they do not require large flocking bird tests on smaller transport-category airplane engines, such as the accident engine, or a test of the engine core; the circumstances of the accident demonstrate that large birds can be ingested into the core of small engines and cause significant damage. The NTSB notes that the FAA engine and propeller directorate, jointly with EASA, initiated a reevaluation of the existing engine bird-ingestion certification regulations by tasking a working group to update the BRDB to include events through the end of 2008. Once the BRDB update is completed, the group is expected to perform a statistical analysis of the raw data and evaluate whether the current regulations still meet FAA and EASA safety objectives and whether additional actions or rule changes are necessary. Therefore, the NTSB recommends that, during the BRDB working group’s reevaluation of the current engine bird-ingestion certification regulations, the FAA specifically reevaluate the 14 CFR 33.76(d) large flocking bird certification test standards to determine whether they should 1) apply to engines with an inlet area of less than 3,875 square inches and 2) include a requirement for engine core ingestion. If the BRDB working group’s reevaluation determines that such requirements are needed, incorporate them into 14 CFR 33.76(d) and require that newly certificated engines be designed and tested to these requirements. Further, the NTSB recommends that, during the BRDB working group’s reevaluation of the current engine bird-ingestion certification regulations, the EASA specifically reevaluate the JAR-E large flocking bird certification test standards to determine whether they should 1) apply to engines with an inlet area of less than 3,875 square inches and 2) include a requirement for engine core ingestion. If the BRDB working group’s reevaluation determines that such requirements are needed, incorporate them into JAR-E and require that newly certificated engines be designed and tested to these requirements.

2.2.6 Bird-Ingestion Protection Devices for Engines

Engine design changes and protective screens have been used or considered in some engine and aircraft designs. For example, certain small turbofan engines, such as the GE CF-34 and some later model Honeywell TFE-731s, incorporate a hidden- or partially hidden-core inlet. The hidden-core inlet design hides the IGVs behind the fan hub rather than placing them directly into the airflow path; thus, all foreign objects pass over the IGVs into the bypass duct and cannot be ingested into the core. However, the hidden-core inlet design results in significant design compromises that increase as the size of the engine increases. The design requires that the engine be longer and heavier because the core inlet duct must be longer to direct the airflow into the core without separation from the duct walls and because the structure, bearings, and shafts must be lengthened. Additionally, the associated engine attachment structure and the airplane structure itself must be strengthened to account for the weight increase, resulting in an increase in fuel consumption. Another compromise the design creates is a nonoptimum relight envelope, which requires that the aircraft be put into a steep dive, an undesirable behavior in a passenger aircraft, to build up sufficient static pressure in the inlet to maintain engine core rotation for a successful emergency relight.

In addition, protective screens are currently used on some modern turbopropeller airplanes and on some turboshaft helicopter engines; however, the type of protective device used on these engines cannot be incorporated into turbofan engines because of the engine construction layout. No manufacturers have developed an inlet screen to protect turbofan engines, such as the accident engine, from bird ingestion. Several technical issues related to performance, weight, and reliability must be considered to determine whether protective screens can be used effectively and safely on turbofan engines, and these issues are summarized as follows:

Impact on engine performance. Screens can block, impede, or distort the airflow just in front of the engine, negatively impacting engine performance and exhaust emissions. Screens can cause erratic engine behavior in crosswind or gusty conditions, increasing the likelihood of a stall.

Impact on in-flight restart envelope. Screens can require a higher airplane restart airspeed to reach the desired engine windmilling rotor speeds, which reduces the restart envelope of the airplane.

Impact of vibration stresses. Screens can disturb the upstream airflow into the engine and induce airflow oscillation, resulting in high airfoil vibrations within the engine and causing premature fatigue and fracture of the fan blades or other airfoils in the engine.

Impact of icing behavior. Screens can accrete ice very easily when they pass through a moist, cool atmosphere. Unless the screens are electrically heated to prevent ice formation, a high risk of screen ice blockage exists. The heat required to deice a screen in extreme icing conditions would require additional generator capacity and large, heavy electrical hardware to deal with the extra power requirements.

Impact of screen and additional structural weight. During informal discussions with engineers from Honeywell and Boeing, it was estimated that the addition of a screen, support structure, electrical harness, and generator would add at least 1,000 pounds per engine installation. Further, the size of the pylon and wing structure would also need to be increased to accommodate the additional weight of the engine, resulting in even more weight being added to the airframe to structurally accommodate an inlet screen.

Screen failure. The reliability of any component can never be 100 percent; therefore, the risk of a screen failure and its subsequent ingestion in the engine inlet must be considered in any design. If a screen were ingested into the engine, it could cause more damage than bird ingestion, leading to a catastrophic engine failure. Damage to the flight control surfaces on the wing or rudder/stabilizer is also a possible hazard.

The NTSB concludes that, although engine design changes and protective screens have been used or considered in some engine and aircraft designs as a means to protect against bird ingestion, neither option has been found to be viable on turbofan engines like the accident engine.


2.3 Flight Crew Performance

Table of Contents

2.3.1 Decision to Use Engine Dual Failure Checklist

At 1527:23, about 12 seconds after the bird strike, the captain took control of the airplane. Five seconds later, the captain called for the QRH Engine Dual Failure checklist, and the first officer complied. Even though the engines did not experience a total loss of thrust, the Engine Dual Failure checklist was the most applicable checklist contained in the US Airways QRH, which was developed in accordance with the Airbus QRH, to address the accident event because it was the only checklist that contained guidance to follow if an engine restart was not possible and if a forced landing or ditching was anticipated (starting from 3,000 feet). However, according to postaccident interviews and CVR data, the flight crew did not complete the Engine Dual Failure checklist, which had 3 parts and was 3 pages long. Although the flight crewmembers were able to complete most of part 1 of the checklist, they were not able to start parts 2 and 3 of the checklist because of the airplane’s low altitude and the limited time available.

The Engine Dual Failure checklist was designed assuming that a dual-engine failure occurred at a high altitude (above 20,000 feet). According to Airbus, the checklist was so designed because most Airbus operations were at high altitude, and, therefore, a dual-engine failure would most likely occur at altitudes above 20,000 feet. Airbus had not considered developing a checklist for use at a low altitude, when limited time is available before ground or water impact. Discussions with A320 operators and a manufacturer also indicated that low-altitude, dual-engine failure checklists are not readily available in the industry.

In 2005, Airbus amended the Engine Dual Failure checklist by including two parallel steps, one for a fuel remaining scenario that included steps to attempt to relight an engine and one for a no fuel remaining scenario that did not include steps to attempt to relight an engine, and by incorporating the ditching procedures, which had previously been located in a separate checklist. Although the amendment allowed pilots to use one checklist, instead of several, for a dual-engine failure, it resulted in a lengthy checklist.

As noted, the Engine Dual Failure checklist did not fully apply to a low-altitude, dual-engine failure and was unduly long for such an event given the limited time available. In fact, the first officer spent about 30 to 40 seconds attempting to relight the engines (as indicated in part 1 of the checklist) because he did not know the extent of the engine damage. Further, the flight crew never reached the ditching portion of the checklist, which most directly applied to the accident situation. A checklist for a dual-engine failure or other abnormal event occurring at a low altitude would increase the chances of a successful ditching and omit many of the steps that took up the flight crew’s limited time.

The NTSB concludes that, although the Engine Dual Failure checklist did not fully apply to the accident event, it was the most applicable checklist contained in the QRH to address the event and that the flight crew’s decision to use this checklist was in accordance with US Airways procedures. The NTSB further concludes that, if a checklist that addressed a dual-engine failure occurring at a low altitude had been available to the flight crewmembers, they would have been more likely to have completed that checklist. This accident demonstrates that abnormal events, including a dual-engine failure, can occur at a low altitude and, therefore, that a checklist is clearly needed to address such situations. Therefore, the NTSB recommends that the FAA require manufacturers of turbine-powered aircraft to develop a checklist and procedure for a dual-engine failure occurring at a low altitude. Further, the NTSB recommends that EASA require manufacturers of turbine-powered aircraft to develop a checklist and procedure for a dual-engine failure occurring at a low altitude. In addition, the NTSB recommends that, once the development of the checklist and procedure for a dual-engine failure occurring at a low altitude has been completed, as asked for in Safety Recommendation A-10-66, require 14 CFR Part 121, Part 135, and Part 91 Subpart K operators of turbine-powered aircraft to implement the checklist and procedure.

Although the flight crew was only able to complete about one-third of the Engine Dual Failure checklist, immediately after the bird strike, the captain did accomplish one critical item that the flight crew did not reach in the checklist: starting the APU. Starting the APU early in the accident sequence proved to be critical because it improved the outcome of the ditching by ensuring that electrical power was available to the airplane. Further, if the captain had not started the APU, the airplane would not have remained in normal law mode. This critical step would not have been completed if the flight crew had simply followed the order of the items in the checklist.

The NTSB concludes that, despite being unable to complete the Engine Dual Failure checklist, the captain started the APU, which improved the outcome of the ditching by ensuring that a primary source of electrical power was available to the airplane and that the airplane remained in normal law and maintained the flight envelope protections, one of which protects against a stall.

2.3.2 Decision to Ditch on the Hudson River

At the time of the bird strike, the airplane was about 4.5 miles north-northwest of the approach end of runway 22 at LGA and about 9.5 miles east-northeast of the approach end of runway 24 at TEB. During postaccident interviews, both pilots indicated that they thought the Hudson River was the best and safest landing option given the airplane’s airspeed, altitude, and position.

About 1 minute after the bird strike, it was evident to the flight crew that landing at an airport may not be an option, and, at 1528:11, the captain reported to ATC that he did not think they would be able to land at LGA and that they might end up in the Hudson. At 1529:25, the captain told ATC that they would also be unable to land at TEB. Three seconds later, he stated to ATC that the airplane was going to be in the Hudson. During postaccident interviews, the captain stated that, “due to the surrounding area,” returning to LGA would have been problematic and that it would not have been a realistic choice. He further stated that, once a turn to LGA was made, “it would have been an irrevocable choice, eliminating all other options,” and that TEB “was too far away.” The NTSB notes that a direct return to LGA would have required crossing Manhattan, a highly populated area, and putting people on the ground at risk.

Simulation flights were run to determine whether the accident flight could have landed successfully at LGA or TEB following the bird strike. The simulations demonstrated that, to accomplish a successful flight to either airport, the airplane would have to have been turned toward the airport immediately after the bird strike. The immediate turn did not reflect or account for real-world considerations, such as the time delay required to recognize the extent of the engine thrust loss and decide on a course of action. The one simulator flight that took into account real-world considerations (a return to LGA runway 13 was attempted after a 35-second delay) was not successful. Therefore, the NTSB concludes that the captain’s decision to ditch on the Hudson River rather than attempting to land at an airport provided the highest probability that the accident would be survivable.

2.3.3 Descent and Ditching Airspeed

As noted, the flight crew was not able to initiate part 2 of the Engine Dual Failure checklist, which contained airspeed guidance for pilots to follow if an engine restart is considered impossible and a ditching is anticipated. The checklist states that, when an engine restart is considered impossible, the optimum airspeed at which to fly is the green dot speed.

Despite not reaching this portion of the Engine Dual Failure checklist, the captain stated during postaccident interviews that he thought that he had obtained green dot speed immediately after the bird strike, maintained that speed until the airplane was configured for landing, and, after deploying the flaps, maintained a speed “safely above VLS,” which is the lowest selectable airspeed providing an appropriate margin to the stall speed. However, FDR data indicated that the airplane was below green dot speed and at VLS or slightly less for most of the descent, and about 15 to 19 knots below VLS during the last 200 feet.

The NTSB concludes that the captain’s difficulty maintaining his intended airspeed during the final approach resulted in high AOAs, which contributed to the difficulties in flaring the airplane, the high descent rate at touchdown, and the fuselage damage. (See additional discussion in section 2.7.1.)

During emergency situations, such as the accident event, pilots experience high levels of stress resulting from high workload, time pressure, and noise. Stress can distract pilots from cockpit duties and result in pilot errors or performance degradation.133 For example, stress can lead to a phenomenon known as “tunnel vision,” or the narrowing of attention in which simple things can be overlooked (for example, airspeed and descent rate) and an individual focuses on a narrow piece of information perceived to be most threatening or salient (for example, surrounding terrain and a suitable landing location).134 During the emergency, the flight crew was faced with a series of GPWS and TCAS aural alerts and many ATC communications, which can also present distractions during an emergency. Further, during postaccident interviews, the captain stated that, during the emergency situation, time was very compressed and that, because he was intensely focused on maintaining a successful flightpath, his attention was narrowed.

To alleviate a pilot being overloaded by aural warnings, Airbus designed alert prioritizations to determine when cues are made available to pilots. However, in this accident, the low-speed warning was inhibited by the GPWS warnings, so that the flight crew was not made aware of the low-speed state. (See section 2.6.1.) Although a visual low-speed indication was available on the airspeed tape, the NTSB acknowledges that the flight crewmembers were overloaded with other visual cues (for example, engine parameters and outside visual references, such as buildings and bridges), which might have affected their ability to continuously monitor the airspeed tape. When the airspeed is high enough, such as the airspeed recommended in the QRH, the AOA never reaches the flight envelope protection activation threshold.

The NTSB concludes the captain’s difficulty maintaining his intended airspeed during the final approach resulted, in part, from high workload, stress, and task saturation.

2.3.4 Decision to Use Flaps 2 for Ditching

The Airbus and US Airways engine dual failure checklists indicated that only blue hydraulic power would be available and, therefore, that only slats would extend when configuring for landing. Although the dual-engine failure certification assessed this worst-case scenario, the possibility of having green and yellow hydraulic systems available was also considered. FDR data indicated that, during the accident event, all three (green, blue, and yellow) hydraulic systems were available and that the flight crew was able to extend flaps and slats. In the accident scenario, the NTSB notes that the selection of flaps 3 would have allowed the airplane to fly at a lower airspeed.

At 1529:45, when the airplane was at an altitude of about 270 feet, the captain instructed the first officer to set the flaps. The first officer then stated that they were at flaps 2 and asked the captain if he “want[ed] more?” The captain replied, “no, let’s stay at 2.” About 1 minute later, the airplane was ditched on the Hudson River.

During postaccident interviews, the captain stated that he used flaps 2 because there were “operational advantages to using flaps 2.” He stated that using flaps 3 would not have lowered the stall speed significantly and would have increased the drag. He stated that he was concerned about having enough energy to successfully flare the airplane and reduce the descent rate sufficiently. He stated that, from his experience, using flaps 2 provides a slightly higher nose attitude and that he felt that, in the accident situation, flaps 2 was the optimum setting.

The NTSB concludes that the captain’s decision to use flaps 2 for the ditching, based on his experience and perception of the situation, was reasonable and consistent with the limited civilian industry and military guidance that was available regarding forced landings of large aircraft without power.

2.3.5 CRM and TEM During the Accident Sequence

Both pilots indicated that CRM was integral to the success of the accident flight. The first officer stated that they each had specific roles, knew what each other was doing, and interacted when necessary. The captain indicated that, because of the time constraints, they could not discuss every part of the decision process; therefore, they had to listen to and observe each other. The captain further stated that they did not have time to consult all of the written guidance or complete the appropriate checklist, so he and the first officer had to work almost intuitively in a very close-knit fashion. For example, the captain stated that when he called for the QRH, about 17 seconds after the bird strike, the first officer already had the checklist out. The captain stated that the US Airways CRM and TEM training, which was integrated into all aspects of US Airways training, including ground school and flight training, gave pilots the skills and tools needed to build a team quickly, open lines of communication, share common goals, and work together.

CVR data indicate that the communication and coordination between the captain and first officer were excellent and professional after the bird strike. Further, the flight crew managed the workload by making only pertinent callouts to ATC and the cabin crew as time permitted. In addition, CVR data showed that each pilot adhered to his role and responsibilities during the accident sequence. The first officer progressed through the checklist while the captain was flying the airplane, communicating with ATC, and determining a suitable landing point. In addition, the captain used the first officer as a resource by requesting his input during the accident sequence.

The NTSB concludes that the professionalism of the flight crewmembers and their excellent CRM during the accident sequence contributed to their ability to maintain control of the airplane, configure it to the extent possible under the circumstances, and fly an approach that increased the survivability of the impact.


2.4 Abnormal and Emergency Events Checklist Design

Table of Contents

NASA researchers have studied the difficulties inherent to designing checklists and procedures for emergency and abnormal situations. A 2005 NASA report noted that, although checklists and procedures cannot be developed for all possible contingencies, checklists should be developed for emergency and abnormal situations “for all phases of flight in which they might be needed.”135