The short answer
At 8:47 p.m. on January 29, 2025, an Army UH-60L Black Hawk with the call sign PAT25 and PSA Airlines flight 5342, a CRJ700 flying as American Airlines from Wichita, collided over the Potomac River beside Ronald Reagan Washington National Airport (DCA). All 67 people aboard died: 64 on the airplane and the helicopter’s three crewmembers. NTSB final report, p. 1
The two aircraft were talking to the same tower controller. About 15 seconds before the collision, the helicopter’s instructor pilot reported “aircraft in sight” and asked to keep clear of the jet visually, and the controller approved it. NTSB final report, Appendix C, p. 375 The tower’s conflict alert had gone off about 26 seconds before impact. NTSB final report, p. 7
The NTSB’s probable cause names several parties. It starts with the Federal Aviation Administration (FAA), for placing a helicopter route close to a runway approach, failing to review the routes and act on recommendations, and relying too heavily on pilots seeing each other. It adds the tower team’s degraded performance under a combined workload, the Army’s failure to teach pilots how far their altimeters could read low, and the helicopter crew’s ineffective visual separation. NTSB final report, p. 299
Watch the episode
The film follows both aircraft on 3D maps and cross-sections drawn to the NTSB’s numbers, so you can see where the helicopter route met the turn to runway 33 and how little height separated them. It uses the NTSB’s own animations, the camera on the riverbank near the airport, the tower’s radar replay from the investigation’s records, and puts the report’s exact words on screen as the narrator reads them. The footage and records are real. The 3D maps and cross-sections are illustrations built from the NTSB’s figures, with heights exaggerated so they read. AI-generated images appear only as plain backgrounds behind the quotations and act titles, and none of them shows the events.
Two flight paths beside Reagan National
PAT25 was on an annual standardization evaluation flight at night, with the crew wearing night vision goggles. It was flying south along the Potomac on the published helicopter routes, Route 1 and then Route 4, back toward its base at Davison Army Airfield. NTSB final report, pp. 1-4
Flight 5342 was approaching from the south, cleared for a visual approach to runway 1. At 8:43 p.m. the tower asked the crew whether they could take runway 33 instead. Moving an arrival to the other runway made room for departures. The crew checked that the airplane had enough runway, accepted, and was cleared for a circling approach to runway 33. NTSB final report, pp. 4-5 The NTSB describes the runway change as an established spacing technique. The turn to line up with runway 33 took the jet across the helicopter route along the river’s eastern bank.
A route 75 feet below the approach
Along that stretch of the river, Route 4 had a published maximum altitude of 200 feet above sea level. The NTSB worked out what that left: a helicopter at 200 feet over the eastern shoreline would be only about 75 feet below an airliner following the runway 33 guidance lights down their 3-degree glidepath, and farther west over the water the margin got smaller. NTSB final report, p. 263; NTSB urgent recommendation report AIR-25-01, p. 6 That is a calculated case for aircraft flying exactly as charted, not a measurement from the night of the crash.
The FAA, which published the routes, told investigators that “helicopter routes were not specifically designed to provide separation between helicopters and fixed-wing traffic.” NTSB final report, p. 266 The chart said to follow the eastern bank but defined no width for the route, and it mixed “maximum” and “recommended” altitude wording. NTSB final report, pp. 44-46, 265-266
Three Army pilots from the helicopter’s battalion told investigators they thought staying at the route altitudes kept them clear of airline traffic. The airline approach charts and PSA’s information for DCA didn’t show the helicopter routes, and the PSA pilots interviewed knew about them to varying degrees. What the accident crew knew could not be determined. NTSB final report, pp. 266-267
What “traffic in sight” meant
The route alone didn’t separate the two aircraft, so the tower used a procedure called pilot-applied visual separation. Once the helicopter crew reported the airplane in sight and the controller approved it, the crew took on the job of keeping it in view and maneuvering to stay clear. At the time, the rules set no distance for that. In the NTSB’s words, “FAA guidance does not prescribe minimum distances by which aircraft must remain separated when operating under visual separation.” NTSB final report, p. 85
The controller kept duties of his own. He had to tell the other aircraft about the helicopter and that visual separation was in use, advise both crews when their radar targets looked likely to merge, and issue a safety alert if he recognized they were dangerously close. NTSB final report, pp. 231-232
Flight 5342 never got that traffic advisory about the helicopter. The controller told investigators he meant to go back to it, but the report says that “because he was attending to other priority tasks, he did not return to the airplane before the conflict alert activated about 1 1/2 minutes later.” NTSB final report, p. 231
What each cockpit heard
The helicopter and the airliner were on different radio frequencies with the same controller. Each crew could hear the controller, but not the other crew’s replies. So the helicopter crew didn’t hear flight 5342 accept runway 33, and the airline crew didn’t hear the helicopter’s position reports. NTSB final report, pp. 2, 256
At 8:46:01 p.m., about two minutes before the collision, the controller told PAT25 about the jet: “traffic just south of the Wilson Bridge is a C-R-J at one thousand two hundred feet circling runway three three.” NTSB final report, Appendix C, p. 370 The airplane was about 6.5 nautical miles away, among several sets of approaching lights. The instructor, who was working the radio while the other pilot flew, reported the traffic in sight and asked for visual separation, and the controller approved it. NTSB final report, p. 6
The word “circling” never reached the helicopter: the tower’s recording has it, and the helicopter’s cockpit voice recorder doesn’t. The NTSB found that “due to degraded radio reception, the crew did not receive information about the airplane’s circling approach to a different runway, which reinforced an incorrect expectation that the airplane did not pose a conflict.” NTSB final report, p. xxiii It concluded that the instructor had not positively identified flight 5342 and had likely picked the wrong target. The record doesn’t show which aircraft he was watching. NTSB final report, pp. 237-239
278 feet, not 200
The NTSB put the collision at about 278 feet above sea level, above Route 4’s 200-foot maximum. NTSB final report, pp. 9, 118 The crew had talked about descending to 200 feet a little over two minutes earlier. NTSB final report, p. 5
The report’s explanation is the helicopter’s altimeters, which measure altitude from air pressure. It concluded the pilots likely saw a reading about 100 feet below their true altitude and believed they were under the limit. Tests on other UH-60L helicopters showed how errors each allowed by the tolerances, in the instruments and in the airflow around the helicopter, could add up. The exact error on the accident helicopter could not be determined. NTSB final report, pp. 117, 120-121, 242-244 The NTSB did not find that the crew knowingly flew above the limit, or that both altimeters failed.
The last 26 seconds
At 8:47:33 p.m., the conflict alert on the tower’s radar display activated. NTSB final report, p. 7 The controller asked PAT25 whether it had the CRJ in sight and told it to pass behind. That instruction never fully arrived: “one of the helicopter pilots pressed the radio push-to-talk switch for 0.8 seconds while the controller was speaking, and this brief radio transmission blocked the helicopter crew from receiving the ‘pass behind’ portion of the controller’s instruction.” NTSB final report, p. xxii
At 8:47:44 p.m. the instructor again reported “aircraft in sight” and requested visual separation. The controller replied “vis separation.” NTSB final report, Appendix C, p. 375 Neither crew got a standard safety alert telling them where the other aircraft was and to turn, climb or descend. NTSB final report, pp. 230-232
The controller was working both the airplane and helicopter positions. He had 7 to 12 aircraft under his responsibility during the last 20 minutes and made 29 transmissions in the final two. An assistant controller and a supervisor were also in the tower. The NTSB found the two positions should have been staffed separately. NTSB final report, pp. 64, 67, 223-230
In the airliner, the collision avoidance system, TCAS, announced “Traffic, Traffic” about 20 seconds before impact. It worked as designed. During descent below 900 feet above the ground, its commands to climb or descend are switched off, and the jet was already lower than that, so it gave only the traffic warning, which doesn’t say where to look. NTSB final report, pp. 31-33, 268-269 PSA told its pilots to look for the traffic and not to maneuver on that warning alone. With the crew busy flying the circling approach and the helicopter hard to pick out against the background lights, the NTSB concluded that “the airplane crew did not detect the helicopter until about 1 second before impact.” NTSB final report, pp. xxiii, 246-248
The helicopter was not broadcasting its position by ADS-B Out because of configuration errors. The NTSB found that didn’t contribute: the tower’s radar still tracked it, and flight 5342 had no ADS-B In equipment to receive the signal. NTSB final report, pp. 244-245, 267-268
Warned after a 2013 near miss
After a near midair collision in 2013, DCA controllers proposed moving part of the helicopter route away from the runway 33 approach. Controllers told investigators they had proposed moving it east, over Interstate 295. The route stayed beside the airport. NTSB final report, pp. 48-49, 264
Later, the tower’s helicopter working group proposed adding “three ‘hotspots’ to the helicopter route chart, which were areas that the HWG had identified as posing an increased risk of midair collision.” NTSB final report, p. 49 The FAA made some boundary changes but left the hotspots off. Its charting manager said such markings were reserved for hazards on the ground. The NTSB could not obtain the proposals or the formal replies, and found no documented annual reviews of the routes. NTSB final report, pp. 277-284
The report’s summary says: “Although midair collision risk was apparent in multiple data sources and concerns had been repeatedly raised by air traffic control (ATC) personnel, the FAA failed to act on recommendations or available information to mitigate the identified risk.” NTSB final report, p. xxii
What the NTSB found
The NTSB adopted its final report on January 27, 2026. Its probable cause names:
- the FAA’s “placement of a helicopter route in close proximity to a runway approach path; their failure to regularly review and evaluate helicopter routes and available data, and their failure to act on recommendations to mitigate the risk of a midair collision,” and the air traffic system’s “overreliance on visual separation”;
- the tower team’s loss of situation awareness and degraded performance “due to the high workload of the combined helicopter and local control positions”;
- the Army’s failure to make sure pilots knew how error tolerances affect their barometric altimeters, which led the crew to fly above the route’s maximum altitude;
- “the lack of effective pilot-applied visual separation by the helicopter crew.”
The report also found that the Army had no program to monitor routine helicopter flight data, which could have shown how often its helicopters flew above the route limits, and that its safety reporting was poorly used. NTSB final report, pp. 284-289
Where the Army disagreed
The Army ran its own administrative inquiry under its AR 15-6 rules, and its findings are dated May 14, 2026. It drew on much of the same evidence but weighed it differently: “The most salient contributing factor was the two broken radio transmissions resulting in the PAT 25 crew not receiving critical details from DCA Air Traffic Control (ATC) that could have enabled them to positively identify AA5342 circling to land on runway 33.” Army findings, p. 8
The Army agreed the crew expected traffic landing on runway 1, but it disputed the NTSB’s finding of expectation bias. Its investigators wrote that they “did not find evidence of information received to be consciously or unconsciously ignored.” The Army still found that requesting and approving visual separation, together with flight 5342 accepting the approach to runway 33, “directly contributed to the accident.” Army findings, p. 8 The two investigations also differ on one word: the Army heard the instructor’s last report as “aircrafts,” plural, where the NTSB’s transcript has “aircraft.” Army findings, pp. 6, 10 We read both reports, not the original cockpit audio, and can’t settle it.
What changed
On March 14, 2025, about six weeks after the crash, the FAA announced it had closed Route 4 between Hains Point and the Wilson Bridge, and the segment came off the helicopter chart. FAA, March 14, 2025 In January 2026 the FAA said it had cut DCA’s arrival rate from 36 to 26 an hour, later raised it to 30, and added staff. FAA statements An interim final rule that took effect on January 23, 2026, restricts helicopter flights in a defined area around the Potomac near the airport, with exceptions for essential operations. Federal Register, 91 FR 2840
Visual separation changed too. From March 18, 2026, an FAA notice barred it between helicopters and the airplanes whose arrival or departure paths they cross, in the Class B, Class C and terminal radar service areas the notice covers, the controlled airspace around larger airports: “visual separation must cease between helicopter or powered-lift aircraft crossing the arrival or departure paths of these aircraft.” FAA GENOT N JO 7110.801 From October 28, 2026, another notice prohibits letting targets merge with less than 500 feet of vertical separation when pilots are keeping themselves apart visually. FAA N JO 7110.804
The NTSB’s final report also calls for traffic awareness integrated into helicopter cockpits, better collision avoidance and alerting, and changes to safety oversight. NTSB final report, pp. 300-306 NTSB recommendations are voluntary, and a recommendation is not proof that anything has been done. We checked the FAA records above between November 11 and 14, 2026, and did not check the status of each NTSB recommendation.
So who was supposed to stop it?
On the night, the helicopter crew had the job of keeping clear once it reported the jet in sight, and the NTSB found that its sighting didn’t establish it had found the right airplane. The tower was supposed to tell the airline crew about the helicopter and to warn both crews when their targets merged, and the controller handling two positions didn’t. Behind both was a route the FAA had drawn about 75 feet under an approach path and kept there after controllers asked to move it, and altimeters the Army hadn’t taught its pilots to doubt. The NTSB named all of them as causes.
Primary sources
- National Transportation Safety Board, Midair Collision Over the Potomac River Near Ronald Reagan Washington National Airport, Arlington, Virginia, January 29, 2025, Aircraft Accident Report NTSB/AIR-26-02, adopted January 27, 2026, corrected February 25, 2026 (printed page numbers cited above; the links open the matching PDF page). The minute-by-minute record of radio calls is Appendix C. Investigation page and animations.
- National Transportation Safety Board, Aviation Investigation Preliminary Report and Urgent Safety Recommendations, AIR-25-01, March 2025 (Figure 3, the 75-foot shoreline case).
- U.S. Army, AR 15-6 administrative investigation findings, May 14, 2026, released by the Army (Army FOIA public records).
- FAA, statement on the NTSB’s DCA recommendations, March 14, 2025, and statements on the midair collision.
- FAA, GENOT N JO 7110.801, Interim Helicopter Separation Procedures, effective March 18, 2026, and N JO 7110.804, Visual Separation, effective October 28, 2026.
- FAA, interim final rule on helicopter operations near DCA, 91 FR 2840-2847, January 23, 2026.
- NTSB, board meeting overview animation, January 27, 2026, with the riverside airport camera (the hero image).
