Goa’s runway near miss deserves a fuller answer

Capt. Amit Singh FRAeS, Founder, Safety Matters Foundation

On the morning of 3 April 2026, an Air India Express Boeing 737 with 171 people on board taxied onto the runway at Goa’s Dabolim airport while an Indian Navy fighter was taking off towards it. The fighter lifted off about a kilometre short of the airliner. No one was hurt. The final report of the Aircraft Accident Investigation Bureau (AAIB) concludes that the cockpit crew misread an instruction and entered the runway without clearance.

That conclusion is correct as far as it goes. The crew did enter without a clearance. But a safety investigation exists to answer a harder question: why did a qualified crew, on a clear morning, end up on an active runway with a fighter rolling towards them? On that question the report is incomplete, and the gaps matter well beyond Goa.

Dabolim is a naval air station, INS Hansa, with a civil enclave. The Navy controls the runway; the Airports Authority of India (AAI) controls civil aircraft on the ground. That morning the two systems did not mesh. A ground controller who had just taken over the watch cleared the aircraft to push back for the wrong runway. The handover had been verbal, recorded in the logbook with a single dot. When the error was caught, the aircraft was rerouted via a short taxiway normally used by arriving traffic and told to “expect backtrack” on the runway. The crew heard that as permission.

The recordings show a chain of missed chances. The crew’s readback of the taxi instruction dropped “hold at holding point” and used the wrong flight number; neither was corrected. When told to change to the Navy frequency, the crew replied “entering runway”. That call went unchallenged for the fifteen seconds before the aircraft crossed the holding line. The aircraft was handed to the Navy at the edge of the runway, not at the earlier transfer point the two agencies’ own agreement specifies. Stop bars, the red lights set into a taxiway to block runway entry, are installed at that very junction. They are switched on only in low visibility.

None of this excuses the crew. All of it should have been analysed. Instead, the report’s probable cause rests on the crew alone, and one of its two causal statements cites an instruction to change frequency “at the holding point” that does not appear in the radio recording. The transcript printed in the report also swaps the speakers on the two most critical transmissions.

Here the international rulebook is instructive. In 2007 ICAO, the UN’s aviation agency, published its Manual on the Prevention of Runway Incursions (Doc 9870). It gives investigators a method for grading the severity of an incursion, from A, where a collision is narrowly avoided, to E, and a structured form for identifying causal factors across pilots, controllers, aerodromes and communications. Between 2023 and 2024, an industry effort coordinated by EUROCONTROL and the Flight Safety Foundation within ICAO’s runway safety framework added the Global Action Plan for the Prevention of Runway Incursions, with 127 recommendations. It identifies breakdowns in coordination between controllers and others as a recurring theme, credits stop bars with strengthening pilots’ awareness of runway status, and urges systems that alert controllers to a developing conflict.

The Dabolim report cites neither document. It assigns no severity category. Its recommendations are counselling for the crew, publication of the stop bars in the aeronautical information publication, and a circular on radio phraseology. It proposes no joint runway safety team for an airport where two agencies divide responsibility, nothing on keeping installed stop bars lit, and nothing on the absence of surface radar at the civil ground control.

The deepest problem is one the report itself names: evidence was lost. Video, camera footage and hotline recordings were erased while occurrence reports moved through naval channels, and AAI’s own report, filed three days later, did not mention the fighter. The fighter’s flight data were seen by investigators at the base but not shared with them. The figures on which the report’s account of the near miss depends therefore cannot be verified.

AAIB has asked twice before for a formal framework governing evidence and access between civil investigators and the defence services: after the Gwalior accident in 2021 and the Port Blair airprox in 2024. It asks again now. Dabolim is not unusual. Civil passengers fly in and out of many defence-controlled airfields in India, Pune, Srinagar and Leh among them. Without that framework, the next investigation at a joint-use airfield may also lose its key evidence.

Safety Matters Foundation has written to AAIB setting out these points. We have asked it to correct the transcript, assess the controller and procedural barriers its own findings record, classify the occurrence using the ICAO method, and frame its recommendations against Doc 9870 and the global action plan. We will publish the full critique on this site.

The aim is not to shift blame from two pilots to a controller or to the Navy. ICAO’s rules are explicit that investigations are not about blame. The point is that a runway incursion involving a fighter near take-off speed is a system event, and a report that explains it as a cockpit lapse leaves the system unchanged. The pilots will be counselled. The airfield will look much the same.

Capt. Amit Singh FRAeS is a Boeing 777 and Airbus A320 captain with more than 35 years in civil aviation, and founder of Safety Matters Foundation, an independent aviation safety NGO.