The National Transportation Safety Board (NTSB) during a public board meeting on Tuesday adopted a probable cause and approved 50 new safety recommendations stemming from its investigation into the Jan. 29, 2025, midair collision near Ronald Reagan Washington National Airport (DCA) involving a PSA Airlines CRJ700 and a U.S. Army UH-60 Black Hawk helicopter. The accident, which occurred over the Potomac River, killed all 67 people aboard both aircraft.
Probable Cause Centers on Systemic Issues
The board concluded the crash resulted primarily from the placement of a helicopter route in close proximity to a runway approach path without procedural mitigations, combined with failures by multiple organizations to identify and address known risks.
As adopted, the board found the probable cause was “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 near Ronald Reagan Washington National Airport.”
The board also cited “the air traffic system’s over-reliance on visual separation in order to promote efficient traffic flow without consideration for the limitations of the ‘see and avoid’ concept,” along with degraded controller performance due to high workload and combined control positions.
Additional causal factors included the Army’s failure to ensure pilots were aware of barometric altimeter tolerances, which resulted in the helicopter flying above the published route altitude.
NTSB Chair Jennifer Homendy said the investigation reinforced long-standing concerns about relying on visual acquisition alone.
“What we refer to as human error is, in reality, the last event in the causal chain immediately preceding a crash,” Homendy said during the meeting, adding that such errors are “a consequence, not a cause.”
Broad Set of Recommendations Approved
The board approved a sweeping set of recommendations aimed at the FAA, the U.S. Army, the Department of Transportation, the Department of War, and industry groups. Of the new recommendations, most were directed to the FAA, including changes to helicopter route design, controller training, conflict alert systems, staffing practices, and airport arrival-rate management at DCA.
Among the actions adopted were requirements to reassess helicopter routes near DCA, to integrate helicopter route information into approach procedures, to enhance controller training on visual separation, and to implement time-based flow management to reduce traffic surges.
The board also approved recommendations calling for expanded use and improved performance of airborne collision avoidance systems and ADS-B equipage across both civil and military aircraft.
Homendy emphasized the urgency of implementation.
“This was 100% preventable,” Homendy said. “We’ve issued recommendations in the past that were applicable here. We have talked about see and avoid for well over five decades.”
Next Steps
The board adopted the final report Tuesday evening, pending technical revisions, and indicated that several members intend to file concurring statements. An executive summary of the findings and recommendations is expected to be released by the NTSB following the meeting, with the full final report to be published in the coming weeks.
Homendy said the board would continue to press for action.
“This is just the first step,” Homendy said. “We must relentlessly and vigorously pursue safety change to ensure this never happens again.”
Commenting as a retired tower operator, my first impression was that having the airliner land straight-in on Rwy 1, rather than switch to Rwy33 for faster taxi time to the gate, would have prevented the collision.
The biggest problem was the design of the helicopter routes and approach paths around DCA. They should never have been that close. Second biggest problem was relying on the helicopter to visually identify the potentially conflicting aircraft when surrounded by ground lights and other traffic at night. Radar guidance would have avoided this. I’m not sure that ADS-B would have prevented the collision. I’ve had it act weird sometimes with false proximity alarms. Third biggest problem was the helicopter flying above the maximum altitude. The Blackhawk’s two barometric altimeters were inaccurate by 80 to 130 feet and Army personnel were aware of this but apparently the flight crew was not. The solution would have been to replace the altimeters so they were accurate. The fourth biggest problem was the volume of traffic around DCA. It should have been restricted by the FAA long ago but political pressure from Congress and Army leadership overrode what the FAA knew was a dangerous situation. It took an accident like this to give the FAA enough leverage to overcome those pressures. A lot of people are pointing the finger at short tower staffing and overworked controllers and that certainly had an effect but it wasn’t the biggest problem in the accident chain.
49 CFR 830.5 (a)(10) requires immediate notification “Airborne Collision and Avoidance System (ACAS) resolution advisories issued when an aircraft is being operated on an instrument flight rules flight plan and compliance with the advisory is necessary to avert a substantial risk of collision between two or more aircraft.” And one of the NTSB reports says there were about 15 RA’s a month in DC airspace.
Did the Board meeting mention if NTSB received these RA reports? If so, how dispositioned? Anyone know?
I did the math on this early-on and with a normal 3:1 approach path to that runway, the helicopter “corridor” was within 100 feet altitude-wise of the approach path, the NTSB found it to be ~75 feet. When you think about the stuff hanging down and up on an airplane (landing gear, vertical tail, etc.), this is basically an intersecting path. Allowing helicopters on the route at the same time as aircraft conducting approaches to the airport was absolutely insane. As opposed to a 200’ helicopter route that exists below the FAF or something, still probably not a good idea to operate helicopters at the same exact time laterally in that case, but the fact that the two routes basically intercepted and they allowed them to both be in use at the same time…crazy.
I think there’s a lot of ulterior motives behind a lot of these recommendations. If the FAA had simply had SOPs in place that required conflicting traffic at intersections of these helicopter and fixed wing routes to managed like airport surface ops, the accident wouldn’t have happened, e.g. holding the helicopter at Hains Point until the CRJ was no longer a collision risk. Everything else was superfluous.
Discussions about this accident seem to assume the helicopter can’t hover or significantly slow its forward progress. Granted hovering at low altitude adds a little increased risk for the helicopter, that pales in comparison to the alternative.
TCAS/ACAS has always had too many false alerts in near-airport operations, which is why RAs are inhibited below 1,100’ AGL and TAs below 600’. I’m not sure that ADS-B in/out would change that, at least without intent info included in position reports and integrating additional logic. I’m all for driving ADS-B in equipage for safety, efficiency and resiliency purposes. That includes utilizing it for self-separation, which is a function the specs assumed. However, I’m not convinced it would have prevented this accident, at least not in its current form.
Hindsight:
“Helicopter PAT25, immediate left turn heading 120. Emergency.”
Why that didn’t happen:
The controller had task saturation. In the previous couple minutes, the controller had just lined up a half dozen airplanes, coming from various headings, into a neat lineup for the runway. That demands focus, and it doesn’t allow the controller to spend a lot of time thinking about any one aircraft.
So, the controller trusted the helicopter pilot.
All the helicopter pilot could see was points of light in the sky. From my own work studying nighttime bicycle crashes, I’ve learned from psychologists how one’s brain can “paint” a picture from points of light that is inaccurate. Depth perception from those points of light is poor.
•••••••
Andrew M commented that the barometric altimeters on the Blackhawk helicopter were 80 to 130 feet inaccurate. Wow! Why didn’t that ground the aircraft?
This was a tragic accident, that does seem largely preventable. Still, I think of the late great aviation writer Ernest Gann’s book “Fate is the Hunter”. As an Aero Engineer and instrument rated pilot, I was interested to know what amount of time those two aircraft occupied the same space. I did a simple 2D AutoCAD study with two circles representing the greatest area of each. The wingspan for the CRJ and rotor diameter for the Blackhawk. The two aircraft were closing on each other on about a 30 degree angle at a closing speed of around 300 mph (working off memory now). I did not complicate my study with altitude changes. Of course the CRJ was descending and the Blackhawk slightly climbing, but this would have made it a 3D problem to solve. In any case, those two circle only intersect for 1/4 of a second! That can be seen by how quickly the images in the simulations grow larger. I think of how if one baggage handler had been slower or faster or the Blackhawk crew the same by a mere 1/4 second all those souls would still be alive.
As a former airline pilot who was based at DCA and flew the exact accident aircraft many times, The reason to for circle to 33 was always FAA encouraged. To the level our POI would talk to us in training once a year and nearly every time the discussion was why that circle was safe (or rather what conditions to circle are safe) and that by accepting the circle. That we are helping DCA get in some number of additional aircraft per hour by accepting the circle. I dont remember the exact number, but it was like 4-8 more take-off/ landings every hour. It all stems from pressure for the airlines that operate there, and the FAA attempting to accommodating more and more operations.
Very interesting! I based my comment on the very first report I heard on TV that mentioned the circle to 33, but I never heard it again on any reportl.
As I have said before, DCA needs to be closed, period. That is the only way to make this crazy complicated airspace safe. Of course I know Congress would never allow that to happen to their “ personal” airport. Surprised the NTSB hasn’t made any mention of Congress interfering with safety in DC airspace.
What does it take to get the military to operate in military operations areas (MOAs) only? One to several exist in every US state. Military helicopters in commercial airspace? Why? Very unlawful by itself!!! MOAs have been around over 50 years. And the military uses them as specialty airspace; which is a not, not, not. No more military in commercial airspace; MOAs only for the military!!! Tell the FAA and the US Congress. The military needs to use their MOAs as multiple use at different times. This is a no brainer people. Why is this situation being ignored by so many people?
Mark, you are assuming that these two flights occurred only once. Commercial traffic and military traffic crossed paths (and didn’t collide) in that airspace many times. Given a large number of opportunities, very low probability events will eventual occur, you can bet on it. It is an illusion that fate had a role. It only seems that way.
Fred, you might re-read my post. I didnt state it was fate. I didnt even say I believe in fate. I simply referenced Gann’s book with that title. He had a pretty interesting flying career, including flying the hump during WW2. Ever read that book or hear of Gann? What I did say is that those aircraft occupied the same airspace for only 1/4 of a second. And that slight changes of event timings would have yielded a different result. I also didnt assume “that these two flights occurred only once”. I understand similar flights have happened many times. The Concord flew many many times until one time it ran over a shed engine cowl part. I’m not saying that was fate either. But in both cases it got a lot of people thinking of how to prevent similar events.
Hi Mark, I read your post more than once and I know who Gann is. Think of the converse, all those mid air collisions that didn’t happen for the reasons you mentioned - a baggage handler delayed by a second, etc. All of those tiny variations work in both directions - sometimes they result in a near miss and sometimes (rarely), they put two aircraft in the same place at the same time (catastrophically). We can ask the same question of near misses - if the baggage handler was a second earlier - or later - planes that missed each other would have collided. Hope that clarifies my point.
I’ve heard no one discuss night illusions with light sources. As the two converged, the lights of the jet would hold station (although growing larger) in the pilot’s field of view. We know this means convergence.
Additionally, the departing plane in the background further muddied the waters. There was a departing plane that may have made the landing plane’s lights appear to be moving away since the only motion discernable was theirs’. I believe the arriving and departing plane’s lights were perceived as a single aircraft by the chopper pilot in night vision goggles..
Fred, I understand variations work in both directions. But no one cares about the near misses, for the most part, except briefly on the prime-time news. So no, those questions never get asked for near misses. I didn’t expect to get into a discussion on fate, it was just a comment in my first post.
MOAs are not exclusive to military use. Civilians are free to fly through them without permission and do so all the time, even though it’s unwise. If the military needed to use military-exclusive (restricted) airspace, one of two things would need to happen:
The amount of restricted airspace would increase by 1000x and massively limit civilian ops, or
The military would accomplish virtually zero training due to lack of airspace.
The reality is that civilian and military aircraft need to share the airspace.