Fatal Crash Karachi May 22nd 2020. PIA flight 8303
A Pakistan International Airlines Airbus A320 carrying 99 people crashed into a residential area of Karachi, Pakistan after reporting loss of thrust on both engines. 97 of those on board died.
It will be some time before details from the flight data recorder and cockpit voice recorder provide a full picture but some parts of the event can be deduced from information in the public domain.
The aircraft made an approach, touched down with the wheels apparently retracted and damaged the engine nacelles on the runway before climbing away again; then the crew – possibly the captain – reported loss of both engines and it crashed into houses, killing all but two of those on board. So far it appears that nobody on the ground died.
From the information available it seems that the aircraft began the approach too high and too fast. Information gathered from radar monitoring systems indicate that at 2000 feet the aircraft speed was almost 240 knots – some 100 knots (114 mph) greater than the final approach and landing speed might have been
The indications are that the undercarriage was not lowered in the first instance, although that cannot be confirmed yet. It would explain why the aircraft did not lose speed however, and on the radio transmissions, a cockpit alarm can be heard that could be the warning that the wheels are not down; it could also be a flap overspeed warning if the crew had extended landing flap without reducing speed.
It seems likely that the damage to the engine pods as they were scraped along the runway, could have in turn damaged fuel supply or electrical components, which resulted in double engine failure.
This outline of a, “Hot & High,” approach as it is often known, seems to fit with subsequent events but it does not explain why a professional crew would carry out such an approach as it would be outside normal procedures by a very long way.
It is standard practice to have theoretical, “Gates,” that must be met during descent and approach and these would conventionally be part of the standard procedures, reinforced in refresher training and form part of an approach briefing between crew member prior to descent.
These gates would culminate in a final one, which would be a requirement to be on the correct approach path (often called the Glideslope) with correct configuration – landing gear down and locked, flaps set for landing – and a stable speed, by 5 miles from touchdown. If that is not achieved at the final 5 mile “gate, “then a go-around must be initiated. At the very least, at 500 feet above the runway all must be stable and a safe landing assured.
Other, “gates,” during decent and approach would also apply but they vary depending on aircraft type and conditions but the principle is the same; if the gate requirements cannot be met then the approach must be abandoned.
This leads us to ask why this was not done in this case. Whilst we are still only able to conjecture, it would seem from the information available that this approach was flawed from the very outset. Some newspapers suggest that the air traffic controller advised the crew to this effect; this should become clear at some point.
Either the pilots were deliberately attempting a high-risk descent profile or there was something seriously amiss. It seems unlikely that the first explanation is the answer and so what other explanation could there be?
One that has been suggested is that lack of flying recently due to the virus lockdown left them vulnerable through lack of practice but that still seems unlikely. If, then, they were simply unaware of the gravity of what they were doing how could that be? It has been suggested that hypoglycaemia as a result o fasting (during Ramadan) may be an explanation but we have no way of knowing if that was the case or even if they had been fasting.
Incapacity for some other reason is a possibility, although the voice on the radio sounded normal and coherent and so it is a puzzle. No doubt a post mortem and toxicology report would answer that.
Some news organisations in the area have expressed concerns over the composition of the investigating body that is being appointed to look into the event and it is not always possible to get full and open reports in circumstances where external influences have an input; we must hope for a swift and competent investigation to reveal more.
One area I have not mentioned is that of Crew Resource Management (CRM) and that cannot be ruled out as a factor here. In a well balanced flight deck both pilots will have equal input into the way the aircraft is operated. By this I mean that whilst the captain carries ultimate responsibility it is no longer considered acceptable to have a dominant person who cannot be questioned or challenged as the commander. In most airlines CRM training to ensure good standards is mandatory.
If either pilot takes decisions that the other is unhappy with, they must speak up and the other should take their point into consideration. In extreme cases a first officer (co-pilot) should take control if they feel the situation warrants it. Was this a factor in this case? At this stage we don’t know but it is a point I will be looking out for in the future, as it seems very strange that neither pilot questioned the way that events were unfolding.
Since posting the above there is now a preliminary report by the Pakistan investigators. Significant items new revealed are:
Latest News, June 2020:
Pakistan International Airlines Airbus A320 – Flight 8303 – Interim report from investigators.
This crash is too recent for a full report but what is known at the time of writing is that the aircraft was en route to Karachi from Lahore and was late to descend for the approach. It was high and fast on the approach and had difficulty in reducing height and speed finally crossing the runway threshold at high speed and with the undercarriage still retracted.
The aircraft touched down a long way into the runway on both engine nacelles, at which point the crew initiated a go around and stated an intention to fly a circuit and land on the second attempt. Probably due to damage during runway contact, both engines failed and the aircraft crashed into a residential area killing all but 2 people on board and miraculously, none on the ground.
It appears that no faults existed with the aircraft prior to the approach and so the questions remain why an experienced crew would fly an approach so far outside correct parameters? Where the pilots unaware that the approach profile was dangerously outside proscribed limits? Or were they used to pushing the limits as this was their home base in good weather and thought they could get away with it?
Either explanation is very worrying indeed. Sadly, we cannot ask them as they died in the crash but the cockpit voice and data recorders provide a little more. There is a preliminary report issued by the Pakistan investigators and some of the points are listed here, verbatim.
- The crew did not follow standard callouts and CRM aspects during most parts of the flight.
- The aircraft ended up higher that the required descent profile.
- Overspeed and EGPWS warnings were triggered.
- At 500 feet the FDR indicates: Landing gear retracted. Slat/flap configuration 3, airspeed 220 Knots indicated airspeed, descent rate 2000 feet per minute.
Here I have cherry picked some salient points as the report is long and technical, containing details that are not directly related to the crew behaviour or the main causes. The Aviation Minister has indicated that the conversations on the flight deck related to coronavirus and that the pilots “were not focused.”
It is a shocking read but I will look mainly at the critical points above. Failure to follow SOPs and callouts is a common thread in crashes, though not all. The aircraft was shockingly high and fast and the configuration that it was in (Undercarriage and speed brakes retracted) made it impossible to regain the correct profile in the distance left to run.
At 500 feet everything should have been completed with the aircraft configured for a safe landing and at the correct speed. The decent rate of 2000 feet per minute is truly shocking. That alone would have been a very hard landing, needing an inspection of the aircraft had they touched down at that descent rate even with gear down. 500 feet per minute would be more normal for final approach.
There will be a full report within 12 months it seems but to summarise. It was a nonstandard operation of the aircraft in which problems multiplied and were not resolved. I cannot tell if they simply didn’t realise how bad it was – although it is reasonable to assume that they did not intend to land without any wheels – or if they just thought they could get away with it. Sadly, all but 2 of the 99 people on board died as a result.
In June 2020 Gulf News reported that around 40% of Pakistani pilots have fake licences. In some cases, others have taken exams for them and GN list other irregularities. A senior pilot in a major airline based in the Middle East tells me that this corruption is not limited to Pakistan. I cannot know if this applies to the pilots of the Karachi crash and of course, his comments are anecdotal and I have no evidence about the other countries that he refers to. The GN article is below.
The European Aviation Safety Agency has now banned PIA from EU airspace for 6 months; the UK authorities have done likewise. The 6 month ban is a preliminary one to allow PIA to address the issues raused by EASA, which have been an ongoing issue for some time. It is possible that the ban could become permanent if EASA is not satisfied. There are 262 Pakstani pilots with licences that are under investigation due to the posibility of “irregularities.”