A Windshear Warning and Low Fuel Near Disaster

On 3 October 2025, a Malta-registered Boeing 737-800, 9H-QBD, operating for Malta Air on a Ryanair-branded service, left Pisa for Glasgow Prestwick with 171 passengers and six crew, on a day the Met Office had already named Storm Amy. The captain, only 33 years old but with over 6,000 hours total flying time, almost all in type, and the first officer had loaded extra fuel because the forecast called for strong southwest winds, rain, and a realistic chance of holding or diverting. They treated Manchester, not the computer-selected Glasgow, as their working alternate. At Prestwick, they were vectored to Runway 20 with a surface wind of 29 knots gusting 45. Autopilot and autothrottle came off around 1,200 feet. Passing 300 feet, the airspeed began to decay; at 264 feet, a windshear warning fired. The captain flew the escape maneuver, converted it to a missed approach, and entered the hold near Turnberry. They waited to see whether another aircraft of the same type would get in. It did, reporting in spite of a 25-knot airspeed loss at 500 feet. They tried Prestwick again. Around 50 feet, the windshear warning sounded a second time. Company policy forbids a third approach after two windshear escapes, so they asked for Manchester. The request was refused, without an explanation that reached the cockpit. They then proceeded to Edinburgh. On the approach to Runway 24, at 745 feet and a little under six miles from the field, the warning fired a third time. The crew declared "mayday mayday mayday fuel," turned south, and landed at Manchester with about 800 pounds remaining against a 2,400-pound final reserve. From the first missed approach to touchdown, the extra flying lasted an hour and forty-five minutes. No one was injured, and the airplane was undamaged. That is the event, confirmed in the UK Air Accidents Investigation Branch's report on the occurrence (Air Accidents Investigation Branch [AAIB], 2026). The rest is why it unfolded that way.

It's worth pausing on the windshear escape maneuver itself before getting into the decision chain, because pilots outside the industry — and more than a few lawyers reading an incident report for the first time — tend to assume it's a routine event that shows up in the sim and, occasionally, on the line. It isn't. Every pilot trains the escape maneuver in the simulator, typically every training cycle, and every pilot hopes never to fly it for real. I flew jet airliners around the world for 40 years and executed exactly one windshear escape maneuver in all that time. This crew flew three in under two hours. That alone tells you how unusual this evening was — and it's also why the maneuver itself was not the problem here. The captain executed all three correctly, and the equipment worked exactly as designed: the enhanced ground-proximity system manufacturer later confirmed all three warnings were genuine (AAIB, 2026). What turned three well-flown escape maneuvers into a fuel emergency happened almost entirely on the ground and in the decision loop around them, not in the flying.

A standard three-degree glideslope puts an aircraft roughly 300 feet above the ground for every mile from the threshold. That rule of thumb is useful here because it shows how late two of the three warnings arrived. The first came about eight-tenths of a mile from Prestwick at 215 feet, almost exactly where a three-degree path would put the airplane. The second came at 50 feet. The third, at Edinburgh, came at 745 feet when the aircraft was still nearly six miles out — well below a three-degree profile if that mileage was measured to the runway. In other words, the crew was already in the regime where windshear is most dangerous and go-around options shrink by the second.

Escape maneuvers are not ordinary missed approaches. Thrust goes to the stops, configuration stays put, and on this airplane fuel flow reached about 7,940 pounds per hour per engine — roughly 265 pounds a minute for both engines combined. Three of those events, plus a hold and a second diversion, ate through the extra fuel the captain had bought in Italy. Total fuel at Pisa had been about 25,350 pounds, including discretionary fuel worth a little more than an hour in the cruise. The extra was real, and still not enough once the flight path became a sequence of maximum-power climbs.

The operational trap was written into two rules that were each defensible on their own terms. The operator allows a second weather-related approach only if the captain has reason to believe it will succeed, and it prohibits a third approach after two windshear escapes. After the second miss at Prestwick, Manchester was the logical next choice. Manchester's own diversion policy said aircraft would be accepted if there was parking space, that emergencies would always be accepted, and that the handling agent does not decide. Airfield Control nonetheless told air traffic control the handling agent could not take the flight. The AAIB could not establish exactly how that decision was made, but found the refusal appeared to result from a communication breakdown inside Manchester Airport rather than an actual capacity limit — the airport had already accepted nineteen other diversions that evening (AAIB, 2026). Prestwick passed the refusal to the crew without the underlying explanation. Still above the fuel state they believed justified a mayday, they took Edinburgh, whose ATIS showed strong wind but no current windshear. Glasgow was reporting shear. They left the hold without briefing a missed-approach plan for Edinburgh and without telling the company. When the third warning came, remaining at Edinburgh looked, to them, like a choice between flying through a windshear caution to protect reserve fuel or going around into a third approach the manual did not allow. Diverting to Manchester, even knowing they would land below reserve, felt cleaner. The flight-management computer predicted 661 pounds on arrival. Their own estimate had been 1,100 to 1,540 pounds. All of those numbers sat under the 2,400-pound reserve.

The human-factors grain of the flight is plan continuation bias under rising workload. Manchester had been the crew's preferred alternate before they started engines. Once the first request was refused, that airport became unfinished business. They discussed whether the Prestwick warnings might have been spurious, even after a same-type crew reported a large airspeed loss. They discussed declaring minimum fuel or a mayday to force Manchester's hand, and decided the printed criteria were not yet met. En route to Edinburgh, they grew more worried about fuel and another warning, yet they did not give ATC a fuel state or build a branch plan. After the third escape, they were, in their own words, convinced Edinburgh would do it again. When the left-tank low-fuel amber came on — the alert triggers at 1,000 pounds — they noted it, skipped the checklist because the reason seemed obvious, balanced a 220-pound split, and kept going. Newcastle, closer and reporting 22 knots from 290 degrees at the relevant time, never entered the conversation. A high workload, mixed English and Italian on the recorder, and a preferred field already in both pilots' heads will do that. Experienced crews do not become careless so much as they narrow. The narrowing here was toward the airport they had wanted since Pisa.

The legal and regulatory layer here is about who owns the decision when several rule-sets occupy the same cockpit at once. Approved fuel planning exists to protect a landing with at least final reserve after unforeseen events. Three windshear escapes and an unplanned second diversion are unforeseen events by any reasonable definition. The operations manual tells the captain that when reserve can no longer be protected, a fuel emergency should be declared and any landing option explored, including departure from ordinary procedures. The same manual forbids the third windshear approach to the same runway. Those two instructions can be reconciled if an unassured landing at the nearer field is treated as the emergency trigger, rather than waiting for a number on the fuel page to cross a threshold. The AAIB noted that the crew's own stated fear of Edinburgh could have supported an earlier mayday call (AAIB, 2026). The crew read the flowchart more narrowly and waited for the computer's 661-pound prediction instead. That timing matters, because a captain's emergency declaration is not a courtesy extended to air traffic control — it is the switch that converts a commercial diversion request into a safety priority with a different, faster set of obligations attached to it. Manchester's post-incident reminder that handling agents do not decide diversions is, read plainly, an admission that written policy and what actually happened that evening were not the same thing. Naming an alternate on a flight plan does not bind that airport to anything; the crew never formally updated the filed alternate from Glasgow to Manchester, because doing so would have meant a phone call and a new plan, made under pressure, mid-event. The fuel was on board. The paperwork was not. In a regulatory or civil setting, those procedural gaps matter less than the central question: whether landing below reserve at a stable, available runway — rather than pressing a third approach into known windshear — was a reasonable use of the discretion the rules themselves create for a captain in extremis.

The airplane taxied in at Manchester with about 530 pounds indicated before shutdown. The AAIB's serious-incident classification is right. The practical residue of this event is not simply "carry more fuel" — this crew already had. It is: treat a refused diversion as a decision that has to be unpacked immediately, not accepted at face value; treat "landing not assured" as the trigger already printed on the fuel flowchart, rather than a number to be confirmed by the FMC first; and treat a favorite alternate as a working hypothesis, not a destination that survives contact with reality. Windshear does not care about arithmetic or the rules in a manual. The rest of the chain — the one that actually failed here — was still people, checklists, and a phone call that should have happened between an airfield desk and a handling agent, and didn’t. I’ve been “fuel concerned” a few times in my career, and it’s about as uncomfortable as I’ve ever been in an airplane. Add weather and three windshear escape maneuvers to that, and all I can say is, I hope they took some time off on the beach with cold beer. They surely needed it!

References

Air Accidents Investigation Branch. (2026, September 24). AAIB investigation to Boeing 737-800, 9H-QBD [Summary bulletin]. GOV.UK. https://www.gov.uk/aaib-reports/aaib-investigation-to-boeing-737-800-9h-qbd

Air Accidents Investigation Branch. (2026). Boeing 737-800, 9H-QBD (AAIB Bulletin, Ref. AAIB-31283) [PDF]. GOV.UK. https://assets.publishing.service.gov.uk/media/6ab3a323734e2435f202eeee/Boeing_737-800_9H-QBD_10-26.pdf