Runway safety in the United States usually lives in checklists, tower logs but the recent close call in Boston forced it into public view. Delta Flight 2351 was on final approach to Boston Logan recently on June 20, lined up, descending and configured to land, when the crew saw and heard something they did not like. On a crossing runway, American Airlines Flight 3161 was already accelerating for takeoff. The crossing geometry meant the two aircraft were on paths that would intersect on concrete rather than on radar screens. Controllers reacted, clearances were revised, and the Delta crew did what training turns into reflex, they pushed the thrust levers forward, pitched the nose up and went around. Later reconstructions described the separation in phrases that make pilots uncomfortable 'several hundred feet apart.' Nobody on board either jet was hurt. Yet the event instantly joined a short list of incursions that remind professionals how fast a normal day can bend toward catastrophe.
To understand why Boston keeps appearing in those stories, you have to start with what a runway incursion actually is. Regulators define it as any incorrect presence of an aircraft, vehicle or person on a runway that is supposed to be clear for landing or takeoff. That definition covers everything from a baggage cart that strays over a hold line to two airliners moving toward the same intersection. Not all incursions are equal.
The most serious ones category A events in FAA parlance are incidents where a collision was only narrowly avoided. One level down sit category B events, where the potential for collision was substantial and it took timely action by controllers or crews to restore separation. Below those, categories C and D cover lower risk deviations and technical breaches that still matter for trend analysis.
By mid 2026 official data show only a handful of serious A and B incursions across the entire United States, compared with more than twenty in 2023. On the surface that is a success story among millions of movements, only a few high-tier incursions, the graph bending in the right direction. Yet within the industry there is a different unease. Airline chiefs have testified that there are hundreds of near misses every day that never reach public databases, events that might not meet the formal definition of serious incursion but still reflect a system under strain. Boston Logan sits near the centre of this tension between numbers and experience.
The airport’s geometry alone raises the baseline risk. Built on a peninsula, Logan relies on a web of intersecting runways to handle shifting coastal winds. Unlike airports with long parallel strips, Boston’s layout forces pilots, controllers and ground vehicles to share more crossing points. Each intersection is a place where a misheard call sign, a moment’s distraction in the tower or a slightly late runway vacation can set up the kind of conflict that played out between Delta and American.
Add to that a heavy schedule across the day, a mix of long haul heavies and regional traffic, and frequent bouts of fog, rain and snow, and you have an operational environment that demands constant vigilance. In that environment the go around is the last clean lever pilots can pull to reset the tape. A go around is not an admission of failure. It is a planned manoeuvre drilled into crews from early training. The procedure is simple advance thrust to go around power, pitch to a target attitude, check positive climb, retract flaps on schedule, and follow the missed approach path programmed into the flight management system.
In practice it is the move that prevents an unstable or unsafe approach from hardening into an accident sequence. At Boston, Flight 2351’s go around took a converging geometry and turned it into two diverging ones. The aircraft climbed away, rejoined the pattern, and landed safely a few minutes later. Passengers felt an unexpected surge and climb, then another circuit, and perhaps some nervous laughter.
For the crew it was just the system working as intended under stress. Critically, the go around sits inside layers of defence. Surface movement radar and electronic flight strips help controllers keep track of who has been cleared where. Standard phraseology reduces ambiguity. Pilots are taught to read back every clearance and to challenge instructions that do not fit their mental picture of the traffic. Safety managers review tapes and data after each event to refine procedures. But no system built around human beings can be frictionless. Fatigue, workload and staffing shortages in towers and cockpits all play their part.
That is why a single incident in Boston can trigger broader questions. Are controllers working more positions than they should at peak periods. Are pilots under pressure to accept tight sequences to keep schedules intact. Are older airfields with crossing runways getting the same investment in ground radar, lighting and signage as gleaming new hubs with simple layouts. Compared with almost any other mode of transport, flying is still remarkably safe.