When ‘I can’t’ is a positive

3985
image: © Marxvile | Wikimedia

Adrian Park on the lessons from the 2013 crash of Agusta 109E G-CRST.

ATC to Agusta 109E G-CRST near Vauxhall Bridge ‘Rocket 2, approved via the London Eye, not above altitude 1,500 feet, VFR … if you can’.

It is 16 January 2013, and air traffic control (ATC) have just radioed to a helicopter flying near London’s central business district. The weather is marginal and the corporate helicopter, an AgustaWestland (AW)109E, is en route to pick up a billionaire client. Protruding ominously into the London fog and cloud are numerous high-rise buildings. Nearly all available landing sites are ‘socked in’.

The pilot replies to ATC:
‘Yeah, we can …’

For this pilot running late—and running the gauntlet between low-cloud and high-rise London concrete—the investigation would show a far better answer would have been ‘no, we cannot’. We cannot maintain visual flight rules (VFR); we cannot maintain sufficient awareness of the city’s skyscrapers; and, with ever-deteriorating visibility, we cannot maintain the necessary margins to avoid an imminent collision. So why the ‘we can’ response? There are many, and complex, answers to this question ranging from cognitive biases to external pressures and cultural conditioning. While the research into such things is often very valuable, perhaps a more direct question is: ‘at the decision-making phase, how many accidents have occurred after I can compared with I cannot?’ Intuitively, I think we all know the answer.

‘It’s still fine here …’. It was never read by the pilot of the AW109, which in a descending right turn, was converging with the undetected jib of a large crane atop a 700 ft high skyscraper.

For that AW109 over London, the accident report would reveal a series of text messages and phone calls between the pilot and various parties on the ground. These texts from the ground included the pilot’s operations staff, a pilot colleague and the client himself. All expressed various degrees of doubt about the ‘can do’ in the pilots ‘I can …’ The client himself, awaiting the 109, and noticing how poor the weather was at the pickup location (i.e. thick fog), called the pilot and suggested delaying the takeoff. The pilot responded that he had already started the engines. A few minutes later the pilot texted the client with ‘I’m coming anyway, will land in a field if I have to’. At this, the client called the pilot again suggesting he remain on the ground, but the 109 lifted and the pilot was soon negotiating weather that London City Airport reported as being 700m visibility in freezing fog with a broken cloud base of 100ft.

‘It’s still fine here …’. It was never read by the pilot of the AW109, which in a descending right turn, was converging with the undetected jib of a large crane atop a 700 ft high skyscraper.

In the next thirty minutes there would be a further 10 text messages sent and received by the aircraft’s mobile phone as the pilot sought information regarding possible diversion points. Weather at the Redhill Aerodrome home base remained fine, but the pilot did not return, evidently wanting to continue to ‘give it a go’. But the fact was the weather en route was far worse then he had anticipated. Radar telemetry of the aircraft showed various and significant altitude and speed changes as the pilot evidently attempted to remain visual around areas of fog and then below ever-lowering cloud. His last text to the client was ‘no holes I’m afraid, heading back to Redhill, at least we tried, chat in 10’. He would never make good on that texted promise.

The final text message from operations to the AW109 pilot was ironically poignant. The text from operator to pilot simply read: ‘It’s still fine here …’. It was never read by the pilot of the AW109, which in a descending right turn, was converging with the undetected jib of a large crane atop a 700 ft high skyscraper. A few seconds later, the main rotor system collided with the jib, and the rotor blades separated catastrophically from the main rotor head. With blade sections dropping down towards the street, the de-bladed helicopter fell ballistically down the side of the building, striking the pavement below and bursting into flames. The pilot was killed on impact, as was a pedestrian engulfed in the explosion. The main rotor head, gearbox and a large section of rotor blade plummeted into the loading bay of a nearby market, striking a delivery van. Most helicopter crashes are spectacularly tragic, but this was even more so because of its central London location.

Again, I’ve often wondered how many accidents occur after the words ‘I can’. Some call it ‘can-do-itis’, others ‘push-on-itis’ and others, normally at a human factors course, call it ‘plan continuation bias’ or ‘confirmation bias’. Whatever you call it, it is easy to point the self-righteous ‘you-should-not-have’ finger at an accident like this, but the difference between the AW109 pilot and myself (or anyone else I’d wager) is not necessarily in having superior (or otherwise) experience. Nor is it necessarily in having superior (or otherwise) judgement and decision-making. The pilot was fifty-years-old with a total of nearly ten thousand rotary wing hours (and over a thousand on type). I doubt very much he would have logged nearly ten thousand hours carrying billionaires around if he had a record of consistently bad judgement. No, the difference between the accident pilot and any other pilot after a ‘can-do’ decision like this is more likely to be one of consequence. The troubling truth is when one holds up the aviation conscience mirror, most of us know deep down it might have been us in this accident (and we should up such a mirror because I don’t know of any helicopter pilot with more than a thousand hours who hasn’t come home at least once with a ‘I think I pushed it’ story). I certainly have a few of these stories. And not just as a junior military pilot early in my career but as an experienced pilot who has spent much of my time facilitating CRM courses and telling others to beware the ‘can-do’ mentality (fair warning: raw moment coming up).

I remember some time ago in a galaxy far, far away (actually it wasn’t that far, but I’ll modify some of the details to protect the guilty—i.e. me) we were providing an aeromedical (AME) rotary wing aircraft for a military exercise. For the ten days we were online the weather insisted on making a nuisance of itself, with almost perpetual low cloud and visibility reducing misty rain. The military like to be in control, and when they are not, they still like to appear as though they are by asking over and over how long it will be before they are in control again (I was in the military for nearly 15 years so I can say such things). So it was I was asked (or nagged) by various commanders on an almost hourly basis ‘can you fly yet?’ This was a new, military twist on the game ‘are we there yet?’ In many ways this was fair enough. I was the pilot in command of the only AME aircraft they had and there were exercise-breaking implications if it couldn’t fly. No fly—due to weather or anything else—meant no live fire (because of their risk management plan). Anyway, after several days of ‘can you fly yet?’, I finally relented and proclaimed that although the forecasts were marginal, we could now fly.

Within an hour we were tasked on a relatively low acuity job to pick up a military patient from a field hospital and transfer him to a nearby civilian hospital. The weather forecasts indicated, much like the accident described above, rubbish weather where we intended to go but where we were right now–no worries. In fact over our location we had blue sky. I explained to the military we would need to pick up the patient relatively quickly and deliver him as quickly as possible because we had only a small weather window before last light.

‘No … we can’t’, I replied.

‘Well, can’t you just get airborne on NVGs and give it a go?’

Of course the transfer was anything but quick, and there were lengthy delays before we were finally on our way. Very soon after take-off the visibility had deteriorated to a point I had not seen in 15 years of flying in the area. The fact of the matter was I had underestimated how bad the en-route weather really was. Favourite landmarks were obscured by misty rain and we forced ever slower and lower. At several junctures we began the set-up for an inclement weather landing in a field, but then it would clear just enough to continue. At another point we began a climb intending to transfer to IFR and finishing the flight via in instrument approach—which would have its own inherent risks—but again it cleared just enough … and we continued because … ‘we can …’.

In the end, we finished the flight at very low altitudes poking along a river that led to the aerodrome and arriving literally on last light. On shutdown, and with the sound of the engines barely fading away, the phone rang again asking if we could get airborne and return to the exercise area immediately. As if for dramatic effect, and drowning out the phone call, the first of several RPT jets reaching the minimas powered up and entered the missed approach unable to get visual. The cloud by now was very close to the ground and the visibility only getting worse.

‘No … we can’t’, I replied.

‘Well, can’t you just get airborne on NVGs and give it a go?’

In what was now darkness, and with another roar of jet engines as yet another airline went around, I held the phone up to the noise for maximum effect.

‘Hear that, that’s another jet going around … we won’t be going anywhere tonight.’

The phone call finished and to say they were very unhappy would be an understatement. It didn’t matter. My relief at being on terra firma significantly outweighed any pressure I felt to continue—my ‘we can’t’ and ‘we won’t’ were now greatly empowered. This is how it often goes with what we could call our ‘acceptability threshold’. It sometimes takes some anxious moments to recalibrate the threshold to a more conservative increment.

Flights like this also remind me it is prudent, and admittedly counter-intuitive, to deliberately look for reasons not to go. As mentioned before there are many psychological reasons why we pilots are so good at finding ‘can-do’ reasons and considering the flow of decisional gravity it is almost always towards ‘I can’—particularly in helicopter AME work—working ‘against gravity’ by voicing reasons not to go is a good thing. Verbalising ‘no-go’ reasons is more likely to reveal deeper truths about actual task complexity. This, in turn, hopefully means a more reasonable assessment of the margins between the demands of the task and the capacity of the crew. This is especially done well when it is done by a crew empowered in assertiveness and who are very much aware of their own limitations as well as those of their teammates. I’ve seen this at work when my medical team says things like ‘this is not a high-priority patient, we don’t need to push it …’ (An articulate no-go reason.) In my ‘can-do’ confessional above, I was given a second chance to say ‘I cannot’ and nobody had to convince me to say it loud and proud the second time around. The words practically leapt out of my mouth. But the fact of the matter was I should have said it the first time. Just like Rocket 2 over Vauxhall, I had every reasonable aviation reason to say ‘no’ and yet I still accepted the job. In the end I had to write a ‘please explain’ regarding the unexpected stopover, but the important lesson I learnt (again) was that no matter what the external pressure is, I can always say I can’t. This is the real lesson from Rocket 2—that saying I can may mean you and your crew never say anything again.

Suggested reading

Aircraft Accident Report 3/2014, Air Accidents Investigation Branch, Report on Agusta A109E, G-CRST, Near Vauxhall Bridge, Central London on 16 January 2013

Leave a Reply