When the fire truck you sent becomes part of the incident. Built within hours of the event — facts incomplete, investigation ongoing.
At approximately 11:40 PM Eastern on Sunday, March 23, 2026, Air Canada Flight 8646 — a Jazz Aviation Bombardier CRJ-900 operating from Montreal — touched down on Runway 4 at LaGuardia Airport. The aircraft was slowing from landing speed when it struck a Port Authority Aircraft Rescue and Firefighting vehicle that had been cleared to cross the runway.
The fire truck was responding to a separate emergency: United Airlines Flight 2384 had aborted its takeoff on the other side of the airport after an anti-ice warning light activated. The crew reported an odor in the cabin; flight attendants felt ill. The ARFF vehicle was cleared to cross Runway 4 at Taxiway Delta to reach that aircraft.
Air traffic control audio captures the sequence: the truck was cleared to cross, then seconds later the same controller called urgently — "Truck One, stop, stop, stop!" — before confirming a collision on the field. The CRJ-900 struck the ARFF vehicle at an estimated 93 to 105 miles per hour. The impact destroyed the cockpit section of the aircraft. Both the pilot and first officer were killed.
What the comm center saw, and when. Color coding indicates the operational dimension.
The vehicle destroyed in this collision was an Aircraft Rescue and Firefighting truck — the exact resource that would have been a primary responder to this crash. When it was struck, the Port Authority lost two officers and a purpose-built emergency vehicle simultaneously with the emergency those resources were needed for. This exercise asks you to sit with that: what happens when the people and equipment you send become part of the incident?
The remaining 72 passengers and two flight attendants survived, though the collision tore open the forward fuselage and one flight attendant still in her jump seat was ejected through the opening onto the tarmac. With both pilots dead and the flight deck destroyed, the aircraft stopped on the runway. There was no cockpit crew to lead an evacuation. Passengers self-rescued — deploying emergency exits and evacuating onto the tarmac — before responding units arrived to assist.
Forty-one people were transported to area hospitals, including 39 from the aircraft and two Port Authority officers from the ARFF truck. By Monday morning, 32 had been released; nine remained with serious injuries including at least one reported brain bleed. LaGuardia closed for approximately 14 hours.
This exercise was built within hours of the incident. Key facts including final casualty counts, NTSB preliminary findings, ATC sequence details, and staffing context are incomplete or unconfirmed. Use known facts; flag gaps for discussion. The exercise will be updated as the investigation develops.
The ARFF paradox. When the primary response resource is destroyed in the initiating event, you have a casualty incident plus a resource deficit. The first minute requires immediate identification of the next-available ARFF capability — not just calling for "more units." Most airport MCI plans assume ARFF resources are available. The scenario where they aren't is less commonly exercised.
No cockpit, no crew leadership, no information. Standard aircraft emergencies assume crew-directed evacuation. When the flight deck is gone, the comm center has zero information about conditions inside the aircraft until responders board it. First-arriving units need to immediately establish a perimeter to contain self-evacuating passengers away from active taxiway and fuel hazard areas.
Personnel statements on recorded channels are now evidence. "I messed up" on a public ATC frequency is a human response to a catastrophic event — and also part of an NTSB investigation and public record. Centers need a protocol for the moment after a critical incident: who relieves the affected dispatcher, who initiates the welfare check, who contacts union representation, and how fast does that happen at 11:40 PM on a Sunday?
Concurrent emergencies create the failure window. The most dangerous moment is often not the major incident but the second, less urgent call that arrives while you're managing the first. ATC was simultaneously managing the United odor complaint and the Air Canada arrival. Cognitive load, task saturation, and divided attention are real. Supervisors should monitor workload and redistribute before a dispatcher reaches capacity, not after.
Notification cascade is pre-incident planning, not real-time improvisation. Comm center supervisor → airport operations → agency leadership → regional hospitals → FAA → NTSB → PIO → family assistance → mutual aid. If your jurisdiction's airport MCI plan lives "somewhere in a binder," that's the gap to close before the next event.
Surface surveillance technology raises the floor; it doesn't eliminate human error. LaGuardia has advanced surface surveillance designed to alert controllers to runway conflicts. The collision happened anyway. The parallel for comm centers: CAD alerts that get acknowledged and dismissed, enhanced data displays that don't get read, protocols that assume the technology caught what a human missed. An alert a dispatcher doesn't act on isn't a safety layer — it's noise.
No right answers. Tap a question to expand the analysis. Use one or all — whatever fits your time.
You just lost your ARFF unit at the same moment you received an aircraft-with-casualties incident that requires ARFF response. Simultaneously, the two officers on that truck are now patients. Your ARFF capability is degraded or gone at the moment of peak demand for it.
First minutes: immediately determine what additional ARFF resources are available at the airport, what mutual aid ARFF resources exist in your response plan, and whether FDNY aviation assets can fill the gap. You are not just dispatching to a crash — you are dispatching to a crash while managing a simultaneous resource deficit.
The ARFF truck that rolled up to Jazz 646 was not a bonus resource; it was the intended primary responder. What is your second line? This is a resourcing discussion your center should have before an event, not during one.
Documentation accessibility matters at 11:40 PM. Does your agency or airport have a written protocol for ARFF asset loss during an active ARFF-required event? Who knows where that document is on a Sunday night?
In a standard aircraft emergency, you expect crew-directed evacuation. Flight crew are trained in emergency procedures, know their aircraft, and provide organized egress. When that leadership is gone — dead, incapacitated, or removed from the aircraft — you have 72 passengers acting on instinct in darkness on a runway, potentially near fuel, in a partially destroyed aircraft.
What does your comm center know, and when? The self-evacuation almost certainly happened before first responders arrived on scene. By the time units reported passengers on the tarmac, some may have been injured further by the evacuation itself, or wandered into active taxiway areas.
Does your first-arriving unit know to immediately establish a perimeter to contain evacuees? Are you tracking an unaccompanied minor, who in this case was eventually located and reunited with family?
The absence of crew command structure is also a communication gap: nobody on the aircraft is coordinating with the comm center. Your information about conditions inside the aircraft is zero until responders board it. Plan for that ambiguity from dispatch.
This is not a question about whether the statement was accurate or appropriate. It is a question about what your center's protocol is for the moment after a critical incident when a dispatcher or controller makes a statement on a recorded channel that goes beyond operational necessity.
"I messed up" is a human response to a catastrophic event. It is also now evidence in an NTSB investigation and public record on LiveATC. Most centers have protocols about post-incident communications — relieving personnel, securing recordings, restricting non-operational transmissions on active channels.
The question for your supervisors: at what point after a critical incident do you relieve the affected dispatcher, and who makes that call?
This is also a wellness conversation. The controller worked a simultaneous emergency, made a clearance decision, and watched the result in real time. That person needs support, not just procedure. Does your center have a protocol for immediate welfare check on involved personnel?
The United incident was relatively routine — an aborted takeoff, odor complaint, flight attendants feeling ill. It was serious enough to require ARFF response, but it was not a mass casualty event. It was exactly the kind of call that gets handled a hundred times a year.
The problem is that handling it required crossing an active runway, at night, with an inbound aircraft in the final seconds of its approach.
The parallel for 911 comm centers: the most dangerous moment is not always the major incident. Sometimes it is the second, less urgent call that arrives while you are managing the first. The clearance that cleared the truck to cross Runway 4 was given in the context of managing a separate active emergency. Cognitive load, task saturation, and divided attention are real.
The question for your shift: what is your center's protocol for recognizing when a dispatcher is at capacity, and how do you redistribute load in real time? Does your supervisor monitor for task saturation, or do they wait for a dispatcher to ask for help?
This is a pre-incident planning question as much as an operational one. In the LaGuardia event, the Port Authority activated emergency response protocols, the FAA issued a ground stop, the NTSB dispatched a go-team, and dozens of arriving flights were diverted — all within the first 30 minutes. Kathryn Garcia held a press conference before most of the city was awake.
For a major airport MCI, the notification cascade runs roughly: comm center supervisor → airport operations center → agency leadership → regional hospitals (trauma activation) → FAA operations center → NTSB duty officer → public information officer → family assistance center activation → mutual aid partners.
The airline also has its own notification chain running in parallel. Air Canada stood up a family assistance line within hours.
The discussion prompt for your center: pull your airport MCI plan. When was it last exercised? Does your comm center have a laminated notification card for an aircraft mass casualty? If the answer is "it is somewhere in a binder," that is the gap to close before the next event.
Advanced surface surveillance systems use radar and transponder data to alert controllers to potential runway conflicts. LaGuardia has this system. The collision happened anyway. The reasons why are for the NTSB to determine.
The lesson for dispatch professionals is consistent with every prior technology-failure incident in this library: technology raises the floor but does not eliminate human error. It supplements judgment; it does not replace it.
In your center, the parallel might be: CAD alerts that get acknowledged and dismissed, enhanced data displays that do not get read, protocols that assume the technology caught what a human missed.
The question for your training staff is whether your dispatchers understand both what their technology does and what it does not do. An alert that a dispatcher does not act on is not a safety layer. It is noise.
Five questions. Answer, then submit for inline feedback. Progress saves locally.