reducing "Crew-caused"
approach and landing
accidents 

Pilot-in-charge Monitored Approach

1990 B707 fuel exhaustion New York USA

Brief account : 

After extensive holding, the B707 ran out of fuel while re-positioning for a second approach following a go-around at very low altitude in poor weather including poor visibility, low cloud and wind-shear.  The Captain was highly experienced, and the Flight Engineer somewhat less so, but the First Officer had only 1600 hours total and 64 hours on type.  

The investigation was hampered by the absence of flight recorder data but position and altitude were obtained from radar traces. The probable cause of the accident was inadequate management of the airplane's fuel load, and the crew's failure to communicate their emergency fuel situation to air traffic control before fuel exhaustion occurred.

Fuel planning had been inadequate from departure but when the flight then encountered  serious delays due to weather and traffic density the reserves became exhausted and the ability to reach an alternate was lost. The First Officer had made numerous remarks to ATC indicating concern about their fuel state but did not formally declare an emergency using standard phraseology, so the situation was not recognised by ATC. Communication both between crew members and from crew to ATC were seriously inadequate. 

Crew-related factors : 

It is likely that the Captain (PF) conducted the initial ILS approach without autopilot or flight director due to defects in this equipment. The flight path was unstable with multiple GPWS alerts and resulted in a go-around from DH altitude, at a time when the fuel state was already very unsatisfactory.

After 9 minutes positioning for a second approach the aircraft lost power on both starboard engines due lack of fuel, hitting the ground 30 seconds later.  

The enquiry considered that poor performance on the ILS was a consequence of the presence of windshear, fatigue resulting from a long flight, possibly flown without the benefit of an operable autopilot, and stress aggravated by concern about the remaining fuel. The Captain's workload had reached very high levels but he was not receiving adequate support from the other crew members.  

It is likely that if the crew had been using a PiCMA procedure: 

1) the Captain's workload would have been lower throughout the initial descent. This would have permitted better strategic planning and consideration of available options given the weather and ATC congestion in the whole area. It would have allowed more opportunity to consider the fuel situatoin in collaboration with the experienced Flight Engineer.

2) the Captain would have been in direct communication with ATC throughout, and better able to ensure that his concerns and decisions were understood.  

Together these 2 factors might well have ensured an earlier approach clearance or diversion and precluded any need for an emergency declaration.  

3) The First Officer would have been able to concentrate on accurately flying the ILS approach with the Captain as a more effective monitor, even given the non-availability of A/P or F/D. The Captain might well have better prepared for the windshear by briefing the inexperienced F/O regarding its possible effects.  As it was, it appeared that the approach became seriously de-stabilised after a (presumed) windshear at about 900ft, with excursions well above and below the glideslope. The first GPWS "Pull up" occurred at about 600ft with a steep descent resulting in the aircraft reaching the 200ft DH 2.25 miles out.

4) With no requirement for the F/O to attempt to obtain visual references, the severe below-glideslope deviation might have been reduced. As it was, the Captain said "lights" just below 400ft but almost immediately lost sight of them as the aircraft reached DH, but was well over 2 miles from the runway. It appears that all 3 crew had been head-up as both the F/O and F/E reported seeing nothing after the go-around was initiated.

In combination, a lower workload combined with a more stable approach path might well have allowed the Captain  to reach a safe decision to land, given that most other aircraft had been able to do so in the same conditions. This would have obviated the need for a go-around with inadequate fuel.  

Type: 
B707-320
Where: 
New York USA
Expected weather: 
Instrument
Pilot in charge: 
Capt
Early transition: 
No
Go-around : 
At or above DH/A
Damage: 
Serious
PicMA potential: 
Major
Year: 
1990
Time: 
Night
Deterioration: 
No
Vert Guidance: 
G/S
Both Head Up: 
Yes
LoC: 
No
Operator: 
Avianca
Fully prepared: 
Yes
Actual Weather: 
Fog
CCAG: 
High
Source material: