The difficulty/confusion was the DIRECTION OF TURN to get there. Left? Or right? Left is towards traffic they have just been told STOP CLIMB! for and Right is towards high ground - and they've been levelled off.

The controller has told us to turn LEFT then RIGHT then LEFT. Which way does she bloody-well want us to turn to get southbound? The most recent direction was left - does she mean go that way? Or the more direct way which is right?

When faced with two diametrically-opposed options, the human reaction is often to do nothing. Human Factors stuff. You need to clarify, and do so quickly.

Can you blokes truly not put yourself in that cockpit and picture this confusion?

I have 15,000 hours of 744/767/737 experience and I can picture this - how one insidious mistake is compounded by further factors and the whole thing can go pearshaped in a matter of minutes.

I really do wonder about the real world flying experience of some of the Monday morning quarterbacks around here who assert that this was primarily the crew's fault. Walk a mile in his shoes, etc etc

Then ask the bloody question but don't continue in a direction with a known huge accumulation of big Cumulu Granitus and in opposite direction to the Course given by ATC, FFS.


How did they end up with Course 0 instead of 180 in the first place? And that was long before Left, Right...

henra:

She immensely compounded her incorrect instruction by focusing on separation of EVA from Air Canada. She did this to exclusion of terrain clearance, which she permits to become a near fatal event. KLAX is only 125 feet above sea level. It is only 23 miles to terrain and obstacles nearly 7,000 feet above sea level. It is only 18 miles to the point where the minimum vectoring altitude (MVA) is higher than EVA's achieved altitude. Keep in mind he had ask for and been granted a "high speed climb" (which is not an approved ATC procedure in the U.S.)

In any case, she became fixated on airplane-to-airplane separation to the exclusion of the pending rising terrain disaster. Once EVA entered the MVA area of 7,700 that surrounds Mt. Wilson, the controller became a passenger. She may have well been in the lunch room for all the good she could do by that point. It's a no-mans land when she has an airplane some 2,000 feet below MVA. Yes, she has an emergency map called the Emergency Obstacle Video Map (EOVM), which provides only 300 feet of obstacle clearance, but which does not have the resolution or fidelity to avoid a ridge line or a massive array of tall antennas.

Her blunders, compounded by a much more serous blunder, did not result in a catastrophic aviation disaster essentially only because the dice rolled in her favor.

I've worked with issues of instrument flight procedures and associated terrain clearance issues for many years. MVAs in mountainous areas have always been a very weak link in the safety chain with the FAA. In the year 2016 it is as bad as it was 50 years ago when you factor in a weak air traffic controller and a crew not familiar with the area.

This woman should find another line of work. But, the FAA will simply decertify her, send her to Oklahoma City for remedial training, then she will be back at the scopes.

And, the "cops" will have by then finished their investigation of themselves. (No conflict of interest there.)

I absolutely agree. She seemed to be not up to the task from a mental strength perspective.
She quickly collapsed with a situation that started with a small lapse. From tiny Error to full panic mode within a few seconds and with no immediate danger present. Someone working in ATC at an Airport like LAX needs to be much much more resilient and stress resistant. Something which can't be fixed with a little bit more training.



Hopefully not. Maybe she gets the hint by herself that this profession isn't for her.

I absolutely don't agree; she did not turn up at work to do a bad job.
She wouldn't be there in the first place if she weren't a first class controller anyway.
But I agree that she did mess up big time!
So I would like to repeat my post of 5 days ago on this thread:

There would be a lot more smoking holes if we were in the cockpit alone..
Where was her backup? :=

Forty plus years ATC here, Sydney, Dubai, Kai Tak, Chek Lap Kok, and it never ceased to amaze me how many of my ATC colleagues from North America that I worked with, somehow managed to get things confused. Culturally many of them could not flatten down the Americanisation of their personalities to Keep It Simple Stupid. Now before I'm jumped on, I did use the word "Many" and I did not use the word "all". I'll also add that the "good" North Americans were very good. That sheila was bloody hopeless.

'North Americans' is often a code phrase for Canadians in my experience. It doesn't mean Mexicans. Beware cheap imitations from the Great White North. And never get between a Canadian and a Walmart in the lower 48. ;)

Kinda like when someone says they are a 'pilot with a major airline', it usually means a commuter or freight dog FO. And when they are a 'consultant', it often means that they are between flying jobs. :)

But, mentioning Hong Kong, you can see vast cultural differences in say, the departures out of HKG and those out of LAX, both with significant terrain in the area. We Americans like simplicity in aviation. But I do agree that we need to do much better with our ATC phraseology in the international realm.

She certainly won't be fired. And, if it is like some cases over on the pilot side, she may be able to get a settlement and be paid not to come back to work.

What does the controller really want
 

The cockpit crew was understandably confused as to what the controller wanted, especially when the initial compliance to instructions produced a confusing series of contradictory instructions from the controller.

Even native English speaking pilots can have difficulties when a controller pulls something unusual out of the hat - as I have had in my own country.

We don't know yet if terrain considerations were addressed in the cockpit, but it seems they were concerned about traffic .

It would be interesting to throw random selections of PPRuNers into a sim with the ATC tape and score the damage to the antenna farm along with Snitch alarms.

I guess we are all using the same Real ATC recording from the 1st post on this thread.

From the recording it's unclear why EVA 15 made the left turn northbound, but lets assume for sake of argument that the controller did indeed make an error and issue the left turn the EVA crew confirmed. However, as soon as the controller saw them turn left she issued an unequivocal correction:

SoCal Departure: „..turn right, turn heading 180.„
EVA 15: „Copied, right heading 180, EVA 15 heavy.„
SoCal Departure: „..please expedite your right turn.„
EVA 15: „EVA 15 heavy, roger just passing heading 010, continue right turn heading.„

Pretty clear to me.

But for whatever reason, they didnt continue the turn to 180 as instructed but appear to have maintained their 010 heading setting up the conflict with Air Canada 788 and getting ever closer to terrain.

Why are we blaming the controller?

They didn't continue a turn right, COFlyer, because they were suddenly told "STOP YOUR CLIMB!", closely followed by "TURN LEFT HEADING 29 CORRECTION 270."

I too hope this lady gets back to work soon, as well as the crew if it's true that they were stood down too. I have no doubt they are all proud professionals.

Even the very best people make mistakes. These days we don't simply discard such individuals; we evaluate errors, work out improvements, do retraining and get people back in the saddle if they're up to it. Besides, the system is more than just one individual: they need to look at the what supervision was happening in SOCAL that day, what oversight of "the big picture" was going on. That's more than this one lady.

We are all on the same team; just got to get the team working together and running down the field in the same direction.

I don't trust the ADS-B plot to perfectly reflect turns. Ian W mentioned that it could be up to 30 sec. (Worst case) before a turn is seen on the radar scope. But it's something like 21 nm, mesaured on Skyvector, from the point where they're heading for Mt. Wilson to Mt. Wilson. What happened during those 21 nm?? Even at 360 kts over the ground, it's 3,5 min. No sign of a turn.

I can agree it would be better by the controller to keep direction of turn to the same direction, as the autopilot (which i am sure was never disconnected) can be quite sluggish.

This whole " incident " is a fiasco and the controller is not the only one that should be under scrutiny. I think the front end crew didn't have a clue. Why in hell would they allow the speed to be in the 330 to 360 range as they head toward the high ground ( mountain ) .....I seriously doubt they even knew of the high ground until the GPWS started going off....

Hmm, dunno. It is not so much the error made (that was a rather minor one -can happen) that shocks me but the total collapse of structure on both sides following that small hickup.
The pilots seemed to have frozen and merrily continued towards the Mountains on opposite direction to the one given by ATC even after repeated requests to turn even if erratic in detail. All this in a relatively benign (at least initially - they let it become critical by no structured action for several minutes) situation.
The controller got into panic mode a few seconds after she made a minor mistake.
Both showed personality traits that are profoundly worrying in their respective profession. And this cannot simply be fixed by a few hours training. I would not want to think about what all of them would do in a direct emergency situation. Resilience should be a top personality criterion in both professions: Airline Pilot and ATC.

172 driver:

No doubt about that. But, it generally shows what happened. The FAA has great data, though, but we will never see it because, alas, the NTSB didn't get involved.

The SoCal TRACON, like all FAA mega-TRACONs has Fusion radar; i.e. all the radars they are using are in a matrix. Their radar recordings are impressive, no doubt. But, since the FAA is investigating itself, those data will not see the light of day.

ADS-B is not subject to any appreciable delay.

If all the packets it transmits are captured, the resulting plot of the turn will be far more accurate than radar, and as near to real-time as makes no practicable difference.

Ok, makes it even stranger that an airplane is plotted tracking north for 20 odd miles while repeatedly instructed to turn south.

172 driver:

An Asian crew in disbelief (understandable) and a controller overloaded with only 2 or 3 airplanes in normally quiet airspace at that early morning except for the fact that LAX is east at the time, which requires a better skill set on the part of the controller for the unusual.

FR24 health warning
 

No argument there, although I wasn't aware that FR24 used any kind of estimating algorithm other than crude extrapolation and dumb joining-the-dots.

That's why a track on the map is only useful given a sufficient plot density and, preferably, timestamps for each point.

Ironically, given the current context, the worst example I've seen was where FR24 turned a lazy 270 into an instantaneous 90 in the opposite direction ...

Plot from WebTrak, showing both the EVA and Air Canada:

http://www.avgen.com/EVA015.jpg