We have a hard, simple lesson from the January collision near Reagan National: see-and-avoid is not a complete safety system when it is asked to do the heavy lifting in dense, mixed traffic. The loss of 67 souls was the result of a chain of small, human and procedural compromises that converged on a single, catastrophic outcome.

What happened at the end of that final approach is straightforward in sequence and complicated in cause. Tower controllers approved visual separation between an Army Black Hawk and an arriving regional jet. The helicopter crew reported the jet in sight and asked for visual separation. Seconds later the two aircraft struck. Live ATC audio shows the tower asking the helicopter if it had the jet and then issuing an instruction to pass behind the jet, yet neither the visual scan nor the procedural safeguards prevented the collision. Visual separation shifted the primary avoidance responsibility to pilots who were operating in marginal conditions for that technique.

Operationally, three factors made see-and-avoid a brittle last line of defense. First, the physical geometry of the area gives almost no vertical margin between a helicopter at the published maximum altitude and an airliner on final. The NTSB found that a helicopter at the 200 foot route ceiling could have as little as roughly 75 feet of vertical separation from an airplane on the Runway 33 approach. That is tiny. In real flight, with sink rates, instrument differences, and visual illusion over water at night, the illusion of safety can evaporate in a handful of seconds.

Second, surveillance and conspicuity gaps erode every pilot’s ability to find and track traffic. The accident helicopter was not broadcasting ADS-B out at the time of the accident, a capability that would have added an independent layer of awareness for controllers and other equipped aircraft. Military exemptions for surveillance equipment have real-world consequences when those aircraft transit airspace next to commercial final approaches. The FAA moved quickly to restrict operations and to require broader ADS-B use near the field, but the lesson is clear: relying on naked visual lookout in an environment where some actors are invisible electronically is asking too much.

Third, human factors in the tower and in the cockpits created brittle communication paths. Investigators uncovered that positions were combined and controllers were coping with heavy traffic demand. Helicopter and local control operate on different frequencies, so crews could not monitor each others’ transmissions directly; they were dependent on the controller to coordinate and to pass traffic information. When you compress time and then layer communication friction on top of marginal visual conditions, the workload spikes and the probability of missed information rises. The NTSB and FAA testimony makes clear that the controller should have made a more definitive call to the airliner about the helicopter and should not have relied on visual separation as a routine traffic management tool in that environment.

From a pilot and operator perspective, the mitigation path is procedural and technical and must be enforced from the top down. Operationally viable steps are these:

  • Stop routine mixed fixed-wing and helicopter operations where approach paths converge. If a helicopter route brings it within feet of a final approach, that route needs to be suspended during arrivals and departures on those runways. The NTSB has made that recommendation and the FAA has moved to restrict the problematic route segment.

  • Eliminate routine use of visual separation inside busy terminal areas. Visual separation is a tool, not a default. When traffic density approaches the point where controllers rely on pilots to see and dodge, the system is being asked to operate beyond its design limits. Controllers should use electronic separation or procedural buffers in lieu of routinely delegating separation to pilots.

  • Require electronic conspicuity. ADS-B or equivalent broadcast capability should be standard for all aircraft that transit near controlled airspace serving commercial traffic, with strictly limited operational exceptions for actual tactical missions. The absence of a broadcast signal should trigger an automatic reroute or ground hold. The FAA already accelerated restrictions and ADS-B requirements around the field after the accident; that direction should become policy wherever mixed traffic is frequent.

  • Improve tower staffing and procedural redundancy. Combining positions to cope with staffing shortfalls may keep traffic moving, but it increases single-point failure risk. Towers under heavy load need redundant oversight and explicit rules that reduce discretionary practices like visual separation until traffic density is manageable.

  • Fix the small things that matter to a scan. For helicopters using night vision devices, ensure altimetry and barometric settings are crosschecked and validated for the operating environment. Night over-water approaches create illusions. If a helicopter operating on a low corridor cannot reliably reference its barometric altitude or radio altitude against a trusted standard, it should not be in that corridor during busy arrival flows. The accident review called out unreliable pressure altitude recordings and NVG-related issues as investigatory items.

Last, the cultural change. Pilots, military leaders, operators, and regulators must stop treating see-and-avoid as an acceptable primary separation method in constrained airspace. It is a contingency measure. When airspace is crowded and fast moving, you build redundancy into the system: surveillance, procedures, communications, and conservative traffic flow limits. Anything less leaves lives on an improvised lookout.

We can, and must, make those changes. They are operationally practical, and they are easy to justify after a night like that on the Potomac. The families who lost loved ones deserve a system that does not ask a pilot’s eyeballs to be the final firewall against catastrophe. The answer is not to ban helicopters everywhere. It is to align routes, equip aircraft, shore up tower procedures, and stop delegating separation to visual lookout when the math and the evidence show the margins are too small.