Another runway incursion - Reagan National 03/14/23

[GarageYears]

https://www.cnn.com/2023/03/14/us/washington-dc-airport-runway-close-call/index.html

I wish it wasn't a CNN report but it seems sadly to be headline news these days

- GY

[chuboy]

FAA needs to tighten things up. US ATC leaves much to be desired but most troublingly there appears to be little insight into its problems, or even outright denial thereof. Slackness with phraseology is endemic in the culture. This causes incidents through loss of SA, but also indirectly because it reinforces a complacent attitude to the job.

[BFSGrad]

Not a good sign when the NTSB and the FAA can't get their stories straight as to what actually happened.

[ATC Watcher]

I think those incidents make the news as the media starts to look for them , and of course if you report them all in succession it makes catchy headlines. It would be interesting to compare the numbers of such inbcidents in previous years and see if the tendency is drastically up, but my gut feeling is that they are not.
The main issues in the US large numbers of runway incursions are well known, complex airports taxi ways set up , crossing runways in use simulteanously, unique procedures like LAHSO , and very lax use of non-standard phraseology to name only a few. Combine this with very high traffic levels at times, and understaffed towers and you have the picture. But since is not new, it's been like this for years.

[roger4]

I thought that Phil Washington said yesterday in his Senate nomination hearing, that there had been 7 runway near-misses in the US at major airports so far this year, and that the normal range was 4 to 10 such events in a full year, i.e. that the current numbers were a significant increase that merited FAA investigation.

[netstruggler]

I thought that Phil Washington said yesterday in his Senate nomination hearing, that there had been 7 runway near-misses in the US at major airports so far this year, and that the normal range was 4 to 10 such events in a full year, i.e. that the current numbers were a significant increase that merited FAA investigation.

You can search the FAA database here (https://www.faa.gov/airports/runway_safety). Looks to me like there were something like 150 in 2019, but not sure how to restrict the search to major airports.

[BFSGrad]

Incident occurred on 3/7 at 13:33Z. KDCA was using runways 1 and 33 for departures.

Republic E75L was cleared by ground controller to “cross 4 at J, monitor tower.” ADS-B data shows that the E75L actually turned left from J onto F and proceeded across runways 1 and 4, heading to runway 33. The original taxi instruction would have taken it to runway 1. The E75L eventually departed runway 33.

The United A319 on runway 1 was issued “cancel takeoff clearance” immediately after completing its takeoff clearance read back and exited at A. ADS-B data shows the United A319 never accelerated beyond taxi speed. The United A319 initiated a successful takeoff on runway 1 about 5 minutes later.

At the time of the rejected takeoff the distance between the A319 and the E75L was at least 2000 feet.

[ATC Watcher]

Thanks for the explanation BFSGrad, looking at the airport diagram , was the initial taxi clearance and pilot read back for Runway 1 or 33 ? If it was for holding point (bay) Runway 1 and it was read back as such , then it is another Pilot error picked up ny ATC, similar to the JFK one . . not really the same as Austin ., and not really ATC related except once again the use of expedited non-standard phraseology (if the "cross 4 at J " was the only instruction issued ) which seems to be one of the common contributing factor in all these

netstruggler You can search the FAA database here (https://www.faa.gov/airports/runway_safety). Looks to me like there were something like 150 in 2019, but not sure how to restrict the search to major airports. . Thanks for that link , well the stats do differentiate between national and regional airports , and can be separated in months and comparing the last 5 months of 22/23 with 2020/21 shows very similar numbers, even on the decrease . No sure what the PD and VPD categories mean though ,

[GarageYears]

I just looked at the FAA database and for Cat A and B incidents I get:

2022: 18 category A+B incidents
2021: 17 category A+B incidents
2020: 16 category A+B incidents

To date in 2023 (not even 1/4 of the year in), there have been 7 incidents reported publicly (the database only covers part of January for 2023), so we're about 200% up on prior years at this point.

- GY

[roger4]

. No sure what the PD and VPD categories mean though ,

PD - Pilot Deviations (Pilot's error leading to the runway incursion)
VPD= Vehicle or Pedestrian Deviations (entering the active runway without authorisation)

and for completeness, the other entries in the table are

OI = Operational Incidents (ATC's error leading to the runway incursion)
Other = not one of the above three.

Not sure what to make of "other" - things like stray animals?

[BFSGrad]

Thanks for the explanation BFSGrad, looking at the airport diagram , was the initial taxi clearance and pilot read back for Runway 1 or 33 ? If it was for holding point (bay) Runway 1 and it was read back as such , then it is another Pilot error picked up ny ATC, similar to the JFK one . . not really the same as Austin ., and not really ATC related except once again the use of expedited non-standard phraseology (if the "cross 4 at J " was the only instruction issued ) which seems to be one of the common contributing factor in all theseThe plot thickens…

After push, the E75L (Brickyard 4736) queries ground about assigned runway. Ground essentially says (marginal audio) “if they said runway 33, expect it.” E75L requests taxi and ground instructs “runway 33, N, J, F, hold short of 1” (letters phonetically spoken). E75L correctly reads back taxi instructions.

Just prior to this instruction, ground incorrectly assigns runway 1 to GoJet (Lindbergh 4430) but uses runway 33 routing (1, K, F, hold short 1). GoJet queries this inconsistency and ground issues the correct taxi instructions (33, K, F, hold short 1).

The E75L subsequently is instructed “cross 4 at J, monitor tower,” which indicates ground is now incorrectly routing the E75L to runway 1. Appears that the E75L expecting to hear “cross 1 and 4 at F” executes this incorrect ground taxi instruction as part of its original routing to runway 33.

About 5 minutes after the incident, a new ground controller voice directs all aircraft to hold all push requests as a controller change is in progress. Planned controller changes are usually seamless with the only indication being a new voice.

[tdracer]

About 5 minutes after the incident, a new ground controller voice directs all aircraft to hold all push requests as a controller change is in progress. Planned controller changes are usually seamless with the only indication being a new voice.

Quite possible that the controller was so shaken that they needed a break - probably to get themselves back into the right frame of mind for a very demanding job.

[WillowRun 6-3]

Knowingly risking a post in "poor form" here . . . but could someone (probably not an SLF like me) pls post a current airport diagram for the airside of this airport? - yes sure I could Google it but I'd much prefer the certainty of having a definitely correct and current version. Thnx.

(I watched the morning segment of FAA's "Call to Action" Safety Summit -- some useful insights yet also much platitude-driving and mere taking up of verbal space. Hopefully the "break-out" sessions this afternoon (not livestreamed) may yield actionable inputs, and FAA said it would provide 'read-outs.')

(Just real real quickly, NTSB Chair Homendy gave intense remarks; and the Acting Administrator showed yet again why he has earned the top job if only the WH would see the value in moving on from its nominee, whose confirmation is stalled and appears quite doubtful of success in the Senate.)

[Concours77]

Incident occurred on 3/7 at 13:33Z. KDCA was using runways 1 and 33 for departures. Republic E75L was cleared by ground controller to “cross 4 at J, monitor tower.” ADS-B data shows that the E75L actually turned left from J onto F and proceeded across runways 1 and 4, heading to runway 33. The original taxi instruction would have taken it to runway 1. The E75L eventually departed runway 33. The United A319 on runway 1 was issued “cancel takeoff clearance” immediately after completing its takeoff clearance read back and exited at A. ADS-B data shows the United A319 never accelerated beyond taxi speed. The United A319 initiated a successful takeoff on runway 1 about 5 minutes later. At the time of the rejected takeoff the distance between the A319 and the E75L was at least 2000 feet.

Fair enough, but imagine how easy it could have been 1000, or 200?

[BFSGrad]

Knowingly risking a post in "poor form" here . . . but could someone (probably not an SLF like me) pls post a current airport diagram for the airside of this airport? - yes sure I could Google it but I'd much prefer the certainty of having a definitely correct and current version. Thnx.
Go to the link below and select "KDCA Airport Diagram" at the left margin for a PDF.

KDCA (https://skyvector.com/airport/DCA/Ronald-Reagan-Washington-Ntl-Airport)

[Equivocal]

About 5 minutes after the incident, a new ground controller voice directs all aircraft to hold all push requests as a controller change is in progress. Planned controller changes are usually seamless with the only indication being a new voice.
Quite possible that the controller was so shaken that they needed a break - probably to get themselves back into the right frame of mind for a very demanding job.
A mature ATC organisation will always relieve a controller who is involved in an incident as soon as possible. This is for two reasons, as tdracer rightly points out, the controller may be shaken up by the event and someone needs to assess whether they are in a suitable state of mind to return to operational work, but also to enable an assessment of the event and the controller's actions to ensure that there is nothing to suggest that the controller's competence could be in question.

[fdr]

A curious incident at an airport that is designed to cause confusion. When determining hot spots, the majority of the airport would be included. Add any poor visibility, fogged windows, snow on ground, standing water etc and it is not a great place to not have a chart showing your location on it.

If ICAO, there would have been a need to note as an amended instruction, can't see that under the FAA 7110 orders or FARAIM.

Seems like the ATC confused the depart way, and the crew had an expectation bias to the taxi instructions, and ambled onwards to 33, along a taxiway that was not part of the last instruction, which had inadvertently been changed to taxi for another runway. Both sides dropped the ball, and the airport sets the scene for this type of event to happen.

6. When assigned a takeoff runway, ATC will
first specify the runway, issue taxi instructions, and
state any hold short instructions or runway crossing
clearances if the taxi route will cross a runway. This
does not authorize the aircraft to “enter” or “cross”
the assigned departure runway at any point. In order
to preclude misunderstandings in radio communica-
tions, ATC will not use the word “cleared” in
conjunction with authorization for aircraft to taxi.
7. When issuing taxi instructions to any point
other than an assigned takeoff runway, ATC will
specify the point to taxi to, issue taxi instructions, and
state any hold short instructions or runway crossing
clearances if the taxi route will cross a runway.
NOTE−
ATC is required to obtain a readback from the pilot of all
runway hold short instructions.
8. If a pilot is expected to hold short of a
runway approach/departure (Runway XX APPCH/
Runway XX DEP) hold area or ILS holding position
( s e e FIG 2−3−15, Taxiways Located in Runway
Approach Area), ATC will issue instructions.
9. When taxi instructions are received from the
controller, pilots should always read back

[ATC Watcher]

I just looked at the FAA database and for Cat A and B incidents I get:
To date in 2023 (not even 1/4 of the year in), there have been 7 incidents reported publicly (the database only covers part of January for 2023), so we're about 200% up on prior years at this point.
- GY
I think we are looking at different data bases , my numbers are coming from here : National airports alone ( thre are additionla pages for regional airports) https://www.faa.gov/airports/runway_safety/statistics/year/?fy1=2023&fy2=2022
@ fdr : A curious incident at an airport that is designed to cause confusion.
Indeed , crazy airport layout to sart with .and once again , a you note , the use on non-standard ICAO phraseology prevented the second cheese layer to do its job. Hopefully one day someone at the FAA will address the phraseology issue . but not to hopeful there.

[BFSGrad]

A curious incident at an airport that is designed to cause confusion. When determining hot spots, the majority of the airport would be included. Add any poor visibility, fogged windows, snow on ground, standing water etc and it is not a great place to not have a chart showing your location on it.

If ICAO, there would have been a need to note as an amended instruction, can't see that under the FAA 7110 orders or FARAIM.An alternate viewpoint is that the air side of KDCA was designed for a different era, a different mission, and a different capacity. The ground side has been wonderfully modernized and expanded, which has unfortunately exacerbated the capacity problems of the air side. Like KSFO, KDCA is stuck on a water-locked parcel of land with no room for expansion or reconfiguration of the air side. Other than a slightly lengthened main runway and some EMAS, the air side looks much as it did 60 years ago.

Regarding your ICAO comment, for a note to be added as an amended instruction, the person amending the instruction would have to first realize that they’re issuing an amended instruction. In this incident, I think the ground controller forgot where the E75L was headed and lumped it in with the rest of the stampede heading for runway 1. In his mind, his “cross 4 at J” instruction was just a continuation of previous taxi instructions rather than an amendment.

KDCA has to be one of the top (bottom?) U.S. airports for comms at a blistering pace during crunch time. There is just no dead air time. Poor United 2003 (RTO A319) had to call 3 times for push and another 3 times for taxi.

[punkalouver]

The plot thickens…

After push, the E75L (Brickyard 4736) queries ground about assigned runway. Ground essentially says (marginal audio) “if they said runway 33, expect it.” E75L requests taxi and ground instructs “runway 33, N, J, F, hold short of 1” (letters phonetically spoken). E75L correctly reads back taxi instructions.

Just prior to this instruction, ground incorrectly assigns runway 1 to GoJet (Lindbergh 4430) but uses runway 33 routing (1, K, F, hold short 1). GoJet queries this inconsistency and ground issues the correct taxi instructions (33, K, F, hold short 1).

The E75L subsequently is instructed “cross 4 at J, monitor tower,” which indicates ground is now incorrectly routing the E75L to runway 1. Appears that the E75L expecting to hear “cross 1 and 4 at F” executes this incorrect ground taxi instruction as part of its original routing to runway 33.

About 5 minutes after the incident, a new ground controller voice directs all aircraft to hold all push requests as a controller change is in progress. Planned controller changes are usually seamless with the only indication being a new voice.
Once again, the pilots crossed a runway that they were not cleared to cross. If they had used the technique that I suggested on the thread for the JFK incident, it would prevent this.

ATC: “Brickyard 4736 cross 4 at J, monitor tower”.
E75 PM: Transmits "Cross 4 at J, monitor tower Brickyard 4736"
E75 PF says: "Cross 4 at J, monitor tower" in order to have the runway number clearly in his mind.
E75 PF then looks at the red runway sign clearly in view and starts to say "runway 4" when he notices that the sign says runway 1 which does not match what he just said. He then refuses to move the aircraft further forward until there is clarification from ATC in order to get a clearance to cross a runway with a number that matches the sign. There may be some confusion until all clarification worked out but no runway incursion.

The problem is.....pilots are not looking at runway signs and making sure the number on the sign matches the runway they were cleared to cross. They are hearing a clearance to cross a runway and always assuming that it is the runway they expect to cross. Of course, 99.9% of the time, the clearance and the expectation align.

AA 106 @ JFK (13 Jan 23) - Page 9 - PPRuNe Forums (https://www.pprune.org/accidents-close-calls/650801-aa-106-jfk-13-jan-23-a-9.html#post11399287)

[GarageYears]

I think we are looking at different data bases , my numbers are coming from here : National airports alone ( thre are additionla pages for regional airports) https://www.faa.gov/airports/runway_safety/statistics/year/?fy1=2023&fy2=2022
@ fdr :
Indeed , crazy airport layout to sart with .and once again , a you note , the use on non-standard ICAO phraseology prevented the second cheese layer to do its job. Hopefully one day someone at the FAA will address the phraseology issue . but not to hopeful there.

I am looking here: https://www.asias.faa.gov/apex/f?p=100:28:::NO

This allows certain filters to be applied. I simply set the category filter to cat A and then cat B and counted. The cat C/D/E incidents are vague 'maybe' and 'sort of' events. Those, although not desirable aren't really what we should be focused on.

A category A or B incidents nearly crunched metal and hurt folks.

-GY

[ATC Watcher]

I am looking here: https://www.asias.faa.gov/apex/f?p=100:28:::NO
This allows certain filters to be applied. I simply set the category filter to cat A and then cat B and counted. The cat C/D/E incidents are vague 'maybe' and 'sort of' events. Those, although not desirable aren't really what we should be focused on.
A category A or B incidents nearly crunched metal and hurt folks.-GY
Thanks for that site. Interesting. I know quite well the classification having been on airprox qualification meetings. Very subjective depending on which seat you're on but definitively no, B incidents are not at all what you describe .
For instance the latest incident where the controller cancelled the Take off clarance to an aircraft that has not started to move yet will definitively not be included.in A or B. In Europe at least

But in order to learn from errors made one has to look equally at all incidents as it is not the higher incident category that delivers the higest benefits. When I was giving lectures I used the Tenerife analogy, Had the Pan Am 747 vacated the runway at the point it was instructed to do by ATC , or even later had the freqency not been stepped on /blocked , it would have been a minor incident (taking off without clear instructions) that, at that time (1970s) would most probably not have been reported, not by the crews and not by ATC. ., and the major phraseology changes that followed never implemented. We do things very differently today.
.

[Hartington]

I just looked at the FAA database and for Cat A and B incidents I get:

2022: 18 category A+B incidents
2021: 17 category A+B incidents
2020: 16 category A+B incidents

To date in 2023 (not even 1/4 of the year in), there have been 7 incidents reported publicly (the database only covers part of January for 2023), so we're about 200% up on prior years at this point.

- GY

The better comparison would be how many incidents up to the same point in the year.. It's quite possible (but unlikely?) that there will be no more incidents in 2023.

[netstruggler]

I just looked at the FAA database and for Cat A and B incidents I get:

2022: 18 category A+B incidents
2021: 17 category A+B incidents
2020: 16 category A+B incidents

To date in 2023 (not even 1/4 of the year in), there have been 7 incidents reported publicly (the database only covers part of January for 2023), so we're about 200% up on prior years at this point.

- GY

2020 and, to a lesser extent 2021, are not typical years for any ongoing air travel statistic.

[ATC Watcher]

" Never trust s statistic that you did not manipulated yourself " or something similar , said Winston Chruchill
To prove you have a higher rate of incidents for a given situation e.g following wrong taxi leading to runway incursions involving Commercial airlines at majot airports , you would need to separate those from the mass. Then find if there is a common denominator.
Takes a lot of efforts .
We still have a lot to learn from the efects of the COVID pandemic,, especially on remaining current, , trainng and cerifying new people during very low traffic , and deviations from normality to accept the sudden raise of traffic .
But based on what I see in my area , there is certainly not a doubling of incidents, but more mediatised ones , yes. , as with FR24 and LiveATC R/T mixed with Twitter we have a lot of amateurs incidents investigators out there..