JetBlue A320 tail strike on takeoff, B350 in opposite direction
avherald.com
avherald.com
And a good visual representation of the evasive action performed: https://www.youtube.com/watch?v=YQLOwje4U9Y
There was ILS as well, though only CAT 1.
Technically part of Europe.
There’s nothing inherently dangerous about an untowered airport, or anything inherently safer about towered ones. In the US and Canada, untowered airports account for a majority of takeoffs and landings.
Also keep in mind that untowered doesn’t mean not talking to ATC.
Someone here fucked up severely, but keep in mind that the deadliest aviation incident ever happened in Europe at a towered field (Tenerife).
It's technically true but when saying something doesn't exist in America, I think raising Hawaii as an exception is still valid or at least interesting.
The reference to "Europe" was likely more about political Europe than geographical Europe.
The citizens of St. Barts are French. Are you claiming they aren't European? Tell me, do they become European when they visit the Metropole? Does this quality vanish when they return to the Caribbean? I'd love an explanation!
People born there? French kids of French parents.
If St Barth had different demographics, there might be a more meaningful debate to be had about this. It’s not exactly the melting pot you might expect from an old colony.
This entire thread is a response to a claim about Europeans, and I took pains to draw out that distinction. Yet here you are talking about Europe. Why?
No. Not by this European and it's almost universal to be specific when you talk about these things. That some would say "Europe" but really mean "the set of countries which are part of the European Union" is so rare that I can't recall anyone doing that (except people from outside Europe. That does happen sometimes).
"Europe" to Europeans means the geographical area called Europe.
I'd say it's even rare in France to say "Union Européenne" instead of just Europe, unless it's in a quite formal context.
I have heard it used on a context where the speaker only means Western and Central Europe, and not Eastern Europe.
I'm just saying the word "Europe" is in practice very often used, within the EU, as a synonym for European Union and I certainly didn't think that would be controversial. On the other hand, in my experience I've never heard "Europe" being used to exclude Eastern Europe, although I've certainly heard it used to exclude the UK (by British persons most often).
And Saint Barthélémy or Réunion island being "in Europe" certainly doesn't mean in any way that Switzerland isn't, I don't think my comment had even the slightest hint of that.
You were responding to a comment that claimed ""America" is used by US people as a synonym for the country, not the continent." (rightfully so, from my 6 years experience in the US). You responded by making a point that, similarly, Europe is used as a synonym for EU. Which many Europeans claimed it is absolutely not typical (rightfully so, as EU citizen). And now you're saying that the term "Europe" actually does include other, non-EU countries... Quite confusing, to be honest, what your point here is.
> I'm just saying the word "Europe" is in practice very often used, within the EU, as a synonym for European Union
Absolutely disagree, it's not common at all. People say EU when they think EU. It is quite common to refer to oneself as European, though, not Europunionan or whatever, when one is thinking of EU citizenship. And perhaps this is what you had in mind.
I don't even understand clearly what we are arguing about. The term "Europe" has different meanings depending on context and speaker, as this thread seems to evidence clearly. All I'm saying at the core is that it's really not exceptional to consider overseas France as part of Europe, and this is actually done all the time here in France.
Regarding the use of "EU" vs. "Europe" our experience clearly differs, but I have to take offense at your claim that my experience is wrong, having lived most of my life in several countries of the EU myself.
Just as another datapoint, look at Wikipedia's disambiguation page for Europe[0]:
Europe may also refer to [...] European Union, a European political supranational entity that excludes Russia and several other eastern European countries.
I've heard people try to use Europe in the context of EU, but typically it requires a clarification because someone misinterprets what they say or it's actually incorrect due to the hold-outs of Switzerland, Norway, UK etc. I heard it more a few years ago than I do now, these days people tend to specifically refer to EU, and Europe more refers to Schengen or the geographical region.
If someone says a European holiday, or it was designed in Europe, or someone is from Europe, the listener doesn't expect Reunion or Serbia, they expect somewhere roughly between Czechia and Spain. Including others under the Europe umbrella would be purposefully deceitful from the speaker's side, or trying to make a political point.
Then again, perhaps this is a central European perspective.
Hawaii is more like Malta in EU, a full member state.
People in Saint Barthélémy are just regular French and EU citizens who live far away from the mainland. The territory does have some degree of autonomy (more than other overseas départements) but still less so than a US state relative to the US federal government.
In that, most French territories would have a status maybe more comparable to a county within a US state (afaiu the US system) than to a US state within the US. The whole comparison is a bit sketchy but France isn't a federal state and only New Caledonia and Polynesia have local legislation, all the rest of the country is very unitary in political and administrative aspects.
For instances, when there was all the talk about Barack Obama's birth certificate, it wouldn't have disqualified him as a candidate to claim his birthplace was in Puerto Rico. (See also John McCain, who was born in the Panama Canal Zone, which had a similar status to Puerty Rico, but distinct from American Samoa.) It would have disqualified his candidacy if they were able to prove he was born in Kenya or American Samoa, which would have made him not a natural born citizen, even though his mother is a US citizen, which would have Barack Obama a US citizen at birth, regardless of where he was born.
St Bart's, like New Caledonia and Polynesia, is an EU Overseas Country/Territory, and a French Overseas Collectivity. St Bart's isn't as close politically to France as French Guiana or Martinique, which are both French Overseas Regions, and EU Outermost Regions.
So I would compare places like French Guiana or Martinique to places like Puerto Rico, Guam, and the US Virgin Islands, and places like St Bart's, New Caledonia, and French Polynesia to places like American Samoa.
https://en.wikipedia.org/wiki/Special_member_state_territori...
https://en.wikipedia.org/wiki/Hawaii: “It is the only state outside North America”
« Europe » by default means the EU where I'm from. There's no standard.
And the majority of those takeoffs and landings are probably tiny planes, which I presume is easier to do adhoc coordination for
Ad hoc coordination does not mean “swerve out of the way if someone shows up on the runway”. It means announcing intent to land on the runway well in advance and coordinating with other planes using the airport over the radio.
If you have a bunch of GA active, odds are that 95% of the pilots are only SEP rated and PPLs. I think US pilots are generally a lot more experienced. GA is so expensive here and GA is often banned from busy areas so there's not many ways to get this experience. I saw some vids of people just flying up to LAX in their 172.. In Europe this would be unthinkable. They'll refuse and reroute you to a local small airport.
In our area they were really strict on announcing all intent to traffic, but it was probably because we did have a tower at certain hours so that kept it on our minds.
- it was over 40 years ago (1977).
- it occured in exceptional circumstances (the Tenerife airport was overloaded because Gran Canaria airport was closed due to a terrorist incident).
- having a tower didn't help a lot because of thick fog. They didn't have ground radar either, but that probably wasn't unusual for a "provincial" airport in 1977.
Little bit of trivia, but on Sully’s famous flight he didn’t quite read back his takeoff correctly. You need to state your identifier AND runway and, IIRC, he only read back “cactus 1549 cleared for takeoff”, omitting the runway. Totally unrelated to the flight, just a little detail which the movie got right (I checked the liveATC audio).
A major cause of Tenerife was that the KLM captain was under the impression that clearance had been granted and the Pan-Am had left the active runway. And that his fellow crew who had been listening were afraid to speak up. A lot of improvements have resulted from that terrible accident.
We reviewed it in flight class, they were always using accidents to explain why rules are so important.
A tower only allows more efficient usage of the runway without a bunch of distributed coordination over a general radio frequency.
There are many airports in the US, Canada, and elsewhere that receive a couple of commercial flights a week and it's not really feasible to have a control tower.
Air traffic control (at least in the US) has the purpose of ensuring separation of air traffic. That is, to help ensure that aircraft do not hit each other, by coordinating when there could be conflicts. They do not primarily exist for (or were not at least created for the purpose of) helping aircraft fundamentally navigate and or avoid non-aircraft obstacles. Or controlling an aircraft's every permissible action in the air.
So, if an airfield has little traffic, in general or such as in off-hours, there may be no tower, or the tower may be unmanned. Aircraft then are responsible for coordinating themselves by announcing their activities, and taking the observation of the airport conditions into their own responsibility.
It may be you believe that because European airspace is a bit denser, and in general the airports with scheduled commercial service are busier because of the density. But as others noted, there are uncontrolled/untowered airports just as in the US.
In fact even when a private jet would come in after hours the tower would be active. It was really for all kinds of commercial activity.
I'm not saying that Europe is better at this than the US but I do think it makes sense to have someone watching especially to ensure separation of traffic as you say. Especially with something as big as an A320.
In this case especially a tower assigning an active runway would have prevented the Beechcraft approaching from the opposite direction.
I have to say also that GA is not as big as in the US and that makes GA pilots in general less vigilant and experienced with proximity to airline traffic so that's another factor.
If you read German, you can check out the magazine Pilot und Flugzeug, it reports frequently about rather insane regulation in Europe (for general aviation).
Part of this is because of the strict regulation. Most pilots only do VFR because getting IFR rated is so expensive and not really interesting from a hobby perspective (no nice views, you can get the same experience in a simulator). Some would do special VFR but it's generally frowned upon.
Also, GA aircraft are generally not welcome at major airports unlike in the US where airport access seems to be viewed more as a right than a favour. So the processes around them tend to not be understood very well. This is another reason nobody opts for IFR ratings, there's not much you can actually do with it anyway, they're not going to let you land at Amsterdam with a Cessna 172 :P They definitely wouldn't carry avgas either.
So, basically we have a whole continent full of "sunday pilots" in terms of GA. And I think the strong legislation and thus costs are part of the cause for that.
PS I've never flown in or even visited the US so perhaps my understanding of the GA landscape there is wrong. But AFAIK it's really used for personal transportation there, which is kinda cool IMO.
I’m curious what it’s like from a passenger POV immediately following such an event.
EDIT: Ohh. Never mind. The tail strike wasn’t the issue. The issue was the gigantic plane coming at them from the other end of the runway. So the passengers had no idea they all almost died.
Kudos to the pilots for getting the plane airborne with nothing more than a tail strike.
Something is off, though: the other plane was landing, not taking off. This plane was taking off. Why was there a tail strike to avoid the other plane?
It would make sense if the other plane was already on the runway. Then you’d have to pull up as hard as possible, resulting in a tail strike. But in this case…
Hmm. Maybe I’m thinking of it the wrong way. If the other plane is landing, your goal is still to “get above it”, as if it were on the runway. So maybe they realized what was happening and immediately made the plane go boing out of surprise.
Also the debate is fascinating in the linked article’s comment section. Someone claimed that the photo was of a different aircraft, and somehow that kicked off a multi day discussion of paint patterns.
Or just to get off the ground so you can get out of the way, can’t really turn until you’re airborne
One other question that might have an obvious answer: how did the pilots become aware of the other plane?
It would be interesting if it was due to radio comms. It implies that this was a shining example of multiple failsafes -- the other plane was coming in for a landing, had already reported that they were landing, but reported it again, and this pilot heard it went "Oh shit."
It would be equally fascinating if the pilot saw the other plane out the window. It implies that if this had happened at night, everyone could have died.
You can see the light easily even during the day when they are flying towards you. It's like cars using their high beam lights - and looking at them on the same straight road from the opposite direction. You are directly in the main cone of where that light is focused.
These days even bikes with their LED headlights on during the day can be quite annoying to people having to look straight at them on the same path even when they still are a hundred meters away. That light on the airplane, the landing light, is meant to illuminate the runway when landing - small airplanes may have to do a night landing on barely illuminated runways. I did some of those, you could only see the outline of the runway and everything else was a blackhole, the surroundings and the runway itself. The runway itself only became visible when I got within a few meters above it. I'm just saying this to show that those lights are not just decoration or position indicator but actually quite strong.
It would be hard to miss that light in the sky (unless you take off towards the sun of course) when it's flying towards where you are and you look straight at it. Fortunately I only ever saw it from such a position waiting next to the runway, not standing right on it (now that would be a panic attack moment). Also, usually takeoff and landing use the same direction, into the wind, in both cases to reduce the amount of runway needed (speed in relation to ground is less than speed in relation to air in that case).
Speculation: maybe they both got TCAS advisories [0].
[0]: https://en.wikipedia.org/wiki/Traffic_collision_avoidance_sy...
I assume they saw it.
Your goal is get out of way. On the runway you can't go lower, you can't maintain level (you never can be sure how far and how fast the incoming plane), you can't go to the side (rudders are an option but you need to be airborne and be sure you still have the time, see level) so your only option is to go up and then go to the side.
> out of surprise
Less of surprise, more of need to take an immediate action, again you can't judge the speed and distance of the airplane of unknown (and often even known) type. To see how hard it is search Youtube for aircraft carrier landings videos.
Also, that the B350 crew was able to report it is also suspect. I'm guessing they thought they had right away as they were descending (I'm not a pilot, but someone here definitely had priority I'd assume) but if you can see a plane taking off at you well enough to see the tail hit the ground, surely you should go around?
Seems like the classic 'both parties were in the wrong' situation. Not reporting such a harsh tailstrike does seem rather wild though. Mistakes happen, ignoring them on the other hand is something else.
I have no idea why the crew in the back didn't let the pilots know.. you'd think there would be a culture of overcommunication when it comes to airline safety. Esp. as the pilots kept ascending..
But on the other hand, we also have evidence that the power gradient on a plane is fairly strong and this means lower-ranked crew are less likely to report problems, as much as they try to create a culture where they would.
It's actually "right of way".
I must be watching a different video, because that one shows a plane landing hard, not taking off.
TCAS exists and can provide collision warnings and some degree of information on where other aircraft are using secondary radar data, but it's very much a last line of defense.
Both planes are expected to see each other, and they're also expected to hear each other on the radio.
Would an A320 not have a similar system? Likewise, both pilots physically moved their heads to scan both directions of the runway for hazards. I’m just confused why the big boys wouldn’t be more sophisticated than this.
Yes, it's called TCAS / TCAS II (mentioned). Or portable systems called PCAS for the smallest planes.
Alerts are generally disabled on the ground, and .. there's a lot of technical factors that limit its usefulness in this scenario.
> Likewise, both pilots physically moved their heads to scan both directions of the runway for hazards. I’m just confused why the big boys wouldn’t be more sophisticated than this.
A careful, visual scan is essential. But when you have oncoming fast aircraft, convergence rates are fast. They were 8-9nm apart, at night, when the A320 started its takeoff roll.
There's basically 3 defenses against this scenario at an untowered airport:
- Weakest: hopefully both aircraft pick the same runway direction. There's a "calm wind" runway for this reason. But when winds are 0 and not quite calm, do they agree?
- Stronger, but problematic-- unicom: aircraft are supposed to announce their intents and listen on a common frequency. If someone has the wrong frequency tuned, that won't work.
- Strongest: visual avoidance. That happened late here, and triggered an overreaction in this case.
Lots of GA aircraft have TCAS-like systems, where they listen to secondary radar responses from other transponders. So, when airliners or ATC interrogates planes, these other planes hear the responses. And, of course, ADS-B makes all of this easier and more precise.
But in any case, TCAS is not designed to prevent a runway conflict. It doesn't know about flight intentions: that the Airbus is going to be climbing a bunch soon.
Additionally, there are guidelines for how to enter the airspace of an uncontrolled airport to reduce congestion and the chance of a collision. There are a few exceptions, one of them being an approach during poor weather, which appears to have been the case in the incident mentioned here. In those situations, the approach controller usually coordinates arriving and departing aircraft to use appropriate runways. I'm not sure why that didn't happen in this case, but I haven't gone and looked for the ATC archives.
EDIT Just read it was an uncontrolled airport as i posted this, so i guess ATC is out of the picture.
An interesting tidbit from wikipedia: "Straight-in approaches are discouraged at non-towered airports in the United States"
[1] https://en.wikipedia.org/wiki/Traffic_collision_avoidance_sy...
Quite obviously, switching units is complicated, but the current patchwork is a nightmare, and has lead to accidents (like the famous Gimli glider, where kg/pound confusion lead to insufficient fuel).
While it’s true that the Hayden airport is uncontrolled, it is in an area in which there is class “E” airspace all the way down to the surface. This means that aircraft must be able to fly at least 500 feet below the lowest cloud to use the airport under “figure it out yourself” visual flight rules (VFR). Just one minute before the accident, the airport reported an overcast ceiling at 500 feet above the ground, so there was not sufficient weather to operate VFR, and therefore all flights would have been Instrument flight rules (IFR).
At an uncontrolled airfield in IFR conditions, it’s true that aircraft have to coordinate with one another for ground operations (who’s taking which taxiway, etc), but must have an IFR clearance from ATC to approach and depart. Since the class E airspace goes all the way to the ground around Hayden, this means that the airport is effectively controlled, except for driving around the surface.
What this means is that every arriving aircraft needs an ATC clearance before flying the approach and must inform ATC that they’ve arrived before ATC can clear another aircraft. Similarly, every departing aircraft needs a clearance before they leave and ATC should not issue a conflicting clearance until after the aircraft is inflight and has checked in. Of course, a takeoff may not be possible, so a departure clearance is issued with a “valid time” and a “void if not off by” time. What this means is that you will receive a clearance which has a start and end time, if you cannot take off within that time window then you need to call ATC back and get another clearance.
At many remote airports, ATC would be contacted literally via cell phone (by pay phone if that doesn’t work – better run back to the plane through the snow and get your engines started real fast before your clearance expires if you only have the pay phone option – I’ve been there). At Hayden, however, there is a remote transmitter (RCAG). This means that Denver Center, the overlying air traffic control facility that controls a huge swath of the Rockies, has a transmitter located nearby so pilots can talk directly to them on the radio. As a result, taking off from an airport like Hayden in IFR is very similar to taking off from a controlled airport, the only real difference is that the actual movement around the taxiways is at the pilots own recognizance.
At airports like this without a local tower, ATC generally has to operate with a “one in, one out” procedure, meaning, essentially, that only one airplane can use the airspace at a time under IFR flight rules. If you’re on the approach, no one else can fly an approach or departure, and vice versa. You typically talk to ATC until they “cut you loose” to talk on the local advisory frequency.
In terms of runways, Hayden is in high, mountainous terrain overall, and although departure procedures are published to avoid terrain, they are very simple, and indeed, I will summarize them as climb to a specified altitude and then off you go. In addition, there is no specified runway to use in calm conditions and the weather report at the time of the incident reported no wind. This means there is no particular reason to prefer one runway over another, and in fact it would have be perfectly reasonable to select either runway on this day.
One interesting feature of Hayden is that the only approach to Runway 28 that was suitable for the weather (RNAV (GPS) 28) has a nearly 20 degree offset from the runway. This means that as the Beech was approaching the runway, it would have been coming from the South and making a final course correction just moment before touchdown, after breaking out of the clouds at 500 feet. In fact, the minimum altitude on this approach is a fairly high ~400 feet above the ground. This means that except for a short window while the Beech made it’s last 500 feet of descent (and a low 20 degree turn to line up with the runway), it would have been been in the clouds, and thus it is unlikely either aircraft could possibly have been in visual contact until very close together.
In terms of the weather, the weather report indicates it was cold, with a 500 foot ceiling. This was indicated as an overcast ceiling, meaning it fully covered the sky. In addition, the weather reported that snow ceased to fall just 5 minutes before the incident. Thus, the cloud cover has to be fairly dense to have just generate falling snow. As a result, these clouds would likely have fully obscured the aircraft from one another.
Hayden has two interesting notes in the airport directory: first, it indicates that all aircraft should announce their departure or arrival intentions on the advisory frequency 10 minutes before takeoff or landing. It also notes that departing on Runway 28 (that is to say, departing in the same direction as the Beech was flying) requires aircraft to make a turnout as “soon as safety permits” to avoid noise over the town of Hayden.
Here cease facts and begin commentary.
Based on weather, their flight patterns and operating rules for the JetBlue flight as well as where the Beech had been flying, both aircraft should have been (and presumably were) on IFR flight plans. As the Beech was arriving from the East and the weather allowed it, it was perfectly reasonable to request the approach to runway 28 (facing to the West) in order save time circling around. Similarly, the JetBlue flight wanted to head East, so a takeoff on runway 10 was reasonable. Furthermore, I would tend to see a note requesting a turn “as soon as safety permits” as shorthand for “there’s a nosy neighbor who will call the FAA and complain about your tail unless you make a very low altitude turnout”, something I would prefer not to do in the clouds while going through all the rest of the rigamarole of an IFR departure in the snow. No thanks, I’ll take 10, please.
My personal expectation flying into this type of airport is that the controller would not cut me loose to talk to the advisory frequency until just a few minutes before landing. Of course both of these airplanes had multiple radios and it should have been possible to listen and talk on the advisory frequency while still talking to center. Nonetheless, especially in certain IFR conditions like this where everyone should be on an IFR clearance, communications on the advisory frequency can often be deprioritized to focus on communications with ATC.
In any case, while talking on the advisory frequency should have been a way for these aircraft to become aware of one another on their own, they were both operating in an IFR environment, meaning that only one of them should have been in a position to be flying through that airspace at once.
This is a very scary event because it was a very near miss that should not have happened and suggests a real breakdown in the system of IFR flying at that time and place.
One small speculation: I wonder if the fact that the Beech was on the approach with a 20 degree course offset saved things? Although 20 degrees is not a huge angle, it may have been just enough to prevent disaster.
I assume the jet was flying IFR, the question is whether the Beech was on IFR and whether/when they cancelled. Let's see what the report says.
For the entire site. Not using a proxy, VPN service, Tor, etc!
I saw someone mention https://mullvad.net/en/check/ here on HN. It was helpful, I found the IP my ISP had me using was on some block lists.
There wasn’t much I could do but it’s good info to have. Browsing blocked sites from cellular or by tethering works. Ironically, a VPN or a service like iCloud private relay (the kind of services that you’d think would cause these issues) can help, too.
I guess contacting lists or site operators can help, though idk how effective it is.
If you do a Google image search, you’ll see lots of cases of them happening.
i thought the whole idea was that the computer maintains safety margins and clamps pilot input if it would result in a stall, collision, etc
regardless of whether fly by wire was implemented (which eventually came for boeing) there were two competing philosophies. boeing believed that when in manual control, the pilot has the final say. airbus built something new that looked more like a modern drone. if the pilot issued a control input and the aircraft thought it was dangerous, it would override it. similar to how it ranges from impossible to hard to flip a drone.
in the early 80s, a demonstration flight of the new airbus system at an airshow in toulouse ended in a fireball when a pilot attempted a low fly by and the plane switched into landing mode, ignoring his inputs to ascend and landing the aircraft in trees.
of course, official story was that the computer (and the multi billion dollar bet behind it) was fine and the pilot and crew were declared responsible. (i think they survived, only to be jailed)
I'd recommend reading this article about it: https://admiralcloudberg.medium.com/fly-by-wire-the-crash-of... :
> This camp believed that the A320’s computers had detected that it was in a landing configuration at low speed approaching the ground and had entered landing mode, preventing Asseline from going around. This wasn’t how the landing mode worked, (landing mode is not a flight envelope protection and can be easily overridden by the pilots using the TOGA switches) and investigators had conducted live flight tests to show that the computers wouldn’t go into landing mode anyway
The pilot was flying low and slow, over a different runway than they had originally planned, at 30ft instead of 100ft, without knowledge of the trees at the end of the runway, with a flight full of passengers, right at the edge of the flight envelope. It was incredibly irresponsible and dangerous, and that's why they crashed.
Another choice quote:
> the A320 went on to be the second most popular airliner ever built, and not one has crashed due to a failure of the fly-by-wire system or an erroneous activation of the flight envelope protections
> Davis specifically alleged that four seconds were missing from the final moments of the flight — enough to put the engine response outside the certification requirements. His main evidence was an apparent disagreement between the time stamps on the ATC transcript and the flight data recorder. The FDR had a “radio transmit” parameter, which showed up in the data four seconds after the final air traffic control transmission. But this was actually a really basic misunderstanding of how the FDR works: the “radio transmit” parameter is only recorded when one of the pilots makes an outgoing transmission, not when an incoming transmission is received. The data point corresponded to Mazières’ reply to the transmission, not the transmission itself, and matched up perfectly with the official timeline.
live and learn.
https://www.youtube.com/watch?v=-0wh2pZHSjc
TL;DR: different flight laws (modes/parameters) are loaded until airfact has been airborne for 5 seconds and this could have allowed it.
As long as the traffic has a transponder, TCAS can see it.
AFAIK, they do, but it's weather radar (to detect things like precipitation and clouds).
There could indeed be other factors, but I'm guessing this is the main one.