It's good someone was awake.
 

On 22/12 there was a breakdown in separation between a Cathay Pacific A330 (B-HLV) (HKG-MEL) and a Virgin Blue B737 (VH-VUJ) (MEL-DRW) at FL370 somewhere over NT.

from the ATSB
Merry Christmas
DJ737

Non-standard levels are a risk especially outside radar coverage. If I ever end up at a non-standard level (rare), I fly a strategic lateral offset. The accuracy of modern nav systems is too good........

What irks me even more are people who ask for block levels covering about 4,000'. In turbulence maybe a 2,000' block can be justified, but some carriers are just plain inconsiderate when regularly requesting block levels.

I noticed in the last ammendment in the ATC section of Jepps it now states that any request for a non standard level has to include the phrase "due to operational requirement".
I would not consider winds or travelling closer to optimum level an operational requirement unless fuel critical. Maybe ATC needs to be less generous with non standard level approvals unless pilots start using this phrase and they have a genuine need for a non standard level.

If it's not available ATC don't grant it. If having been granted it the clearance will subsequently become a problem ATC modify or cancel it. I'm not sure what the issue is for you if someone else gets a block clearance whether it's 2000' or 10,000'. :confused:

And further to Kegs post,

I'm not sure what relevance "operational requirements" should have on whether or not a non standard level should be requested or given.

Like Keg says, either its available or its not. If you're at a non standard level, you're the first one to get shifted if it doesn't suit anyway (in my experience).

Yeah, but you are talking about error catching. In the ME, we don't use N/S levels for cruise, and only ever (rarely) as a confilct fix. Just removes another 'hole'.

Given the standard of ATC & level of traffic congestion in some parts of the ME, especially India, I wouldn't be requesting non standard anything !! Thankfully OZ is a totally different case IMHO.

All that is absolutely fair enough and I agree, but we're talking an incident in Australian airspace and a new requirement (AFAIK) in the Australian ATC Jepp section (page AU-805).

If you're in the middle of nowhere and don't feature on a screen somewhere then fair enough, but up and down the east coast of Aussie? Surely we have the capability and skill to do non standard levels quite safely. Surely??


Worlds best pract..............and all that.

Any comment from a controller?

Removing holes is generally regarded as a good thing in ATM.

If they can't be removed or plugged, then some kind of operating practice/tips/auto-alert thingy provided to mitigate a perceived high risk situation.

This has been an evolving practice in a lot of ANS providers since day one, hugely aided (and at the same time made more complex) more recently by technology.

Unfortunately it seems to me that in mitigating against a lot of these "gotchas", the exposure of the human element in dealing with unusual or non-standard procedures has become very limited.

In ancient times the "non-standard" was almost the norm, most were well practiced at dealing with it and the tools used were regularly dusted off and given a work out. There were other "gotchas" that could catch the unwary out, but I believe the operating environment "way back then" cultivated and encouraged a different order of situational awareness and maintain-a-good-scan type mentality.

With the more systems-based technology we are now using it is more likely in my opinion that doing things in a non-standard manner, or approving a non-standard operation rarely is more likely to be risky than approving it (space and procedure permitting) either more regularly, or not at all. Ever.

It's my view that it's best to not try and come up with rarely used procedures aimed at mitigating risk for infrequent situations. The procedure will be forgotten unless called on regularly. If it is in a QRH, one has to remember and know there is a procedure in the QRH, then look for it. And at the same time the OS still requires monitoring, there is often less time and attention available for dealing with ancillary tasks.

Not a thing to want to be doing (if applicable) at 2.52 in the morning.

Either (1) make the procedure routine enough, and train for it, so it is almost instinctive to carry out safely, or (2) never use the procedure. I think ATM generally is moving toward option (2).
Whether that is the correct direction to be moving, I don't know. In my experience, sooner or later, there will always be a legitimate and sometimes pressing requirement to handle something non-standard.
I say thank goodness to an alert flight crew that had the situational awareness, and the luxury of a common language, to query it.
Without wanting to second-guess the result of the inquiry, that was possibly the second last hole in the cheese.
That's where we're all heading. ;)

[edit to add:] My comments, in case it is not obvious, are concerning occurrences of this type, particularly where a non-standard element is involved, generally. I have no inside knowledge of this one, nor of the way area control is performed in Australia.
I think it is high time a holistic overview of training/technology is entered into.

Direct no speed it's the same with GA, while ever there are people prepared to work those hours for that pay under that pressure then it's only going to get worse. Does knowing those conditions are dangerous and still working breach your duty of care legally? You have to watch those legal folks always out for income maximumisation. Isn't stating an opinion of such on a public forum an admission of problems? Can'o'worms stuff.

No supervisor awake though. I find it odd that there is no night shift supervisor in that part of the world seeing as it is the peak of complexity and traffic around 3am.

You can expect a standard knee jerk reaction to this incident...the use of non-standard levels will probably be henceforth banned.

There are probably multiple factors to consider here.
First is the "can do" culture that has sprung up in Oz over the past 20 years.
There is also the push from management that the controllers have to facilitate their "customers" as much as possible, often to the detriment of safety.

I have no idea of the circumstances of this incident other than what has been mentioned here, but making a stab at it anyway, IMHO controllers in Oz are trained these days to pretty much say "yes" to most requests...in the old days some were trained to think "no" and then look for a reason to justify it. If no reason could be found, he got the level/routing etc.
There has to be a happy medium there somewhere.

The "back of the clock", disengaged workforce argument I do not think is the reason (although I hasten to add, I do not work in Oz anymore).
We have had two nose to nose STCA (Short Term Conflict Alert) sorted incidents in my patch in Europe recently, in full radar coverage, with a fully engaged and well rested workforce.

Sh!t happens, humans make mistakes, and happily all the holes in the cheese didn't line up.
Good work from the pilots for having good situational awareness.

ferris, I was responding to someone's comments about large block clearances rather than operations at non standard levels.

I don't mind using non standard levels* if I need to but I make sure that I'm more tuned in with the traffic around me if I do.

*(current operations limit this to Continental Australian Airspace. I'd never do it in Indonesia, Phillipine, Thai, Indian, Pakistani, or Burmese airspace.....and probably a few others that I haven't included there also).

Havent most of QFs incidents in Indonesian/Asian airspace occurred when we were operating at standard levels?

Quite frankly i don't think it matters if you re at standard or non standard levels... you can still be at risk. (Think of jets changing levels through yours - deviating around weather without a clearance - loss of comms through indo /Philippine airspace with crossing traffic)

The introduction of RVSM in Australian airspace has produced a parcel of "convenient" levels above F290 that were not an option in prior times. The natural tendency has been to make use of these levels as desired, consigning the 'non-standard' aspect to the 'doesn't matter' bin.
Over the North Atlantic, and similar, RVSM has a real needed purpose. In Australia it is a luxury. (but, dare I say, world's best practice?)

It's kind of funny to reflect that we fly along designated air routes (railway tracks in the sky) at set altitudes within +/- 50 feet vertically and +/- 100 metres laterally to such an extent that two a/c pass each other head-on at 1800 km/h in the middle of absolutely nowhere on a regular basis with only 1000' of separation. The much reduced probability of this happening if aircraft used random tracks and altitudes in remote areas (no ATC) would seem to be a far better option sometimes.

Someone with a brilliant theoretical, but practical mind will solve this problem sometime in the future and we will all sit back and say, why didn't I think of that?

Keg, I wasn't responding to you either- the quote was taken from waren9's post. I was talking about (from an ATC perspective) the use or not of n/s levels.

In oz, although being a legitimate sep standard, we never used assigned rates of climb/descent. In the international environment, people do it all the time. N/S levels used for cruise in oz, not elsewhere (M.E.).

Could go into a big post about risk and the GAFA, but no time. Agree with DD that there will prob be a knee-jerk rule made. It's the way modern management operate.

FR: the world is moving away from opp direction tracks, which is the only place where offsetting helps. But where opp dir. tracks still exist- yes, offset should be mandated (in the RPT world).

David Gunson summed it up beautifully... "We force them down narrow corridoors, therby greatly increasing the risk of collision, while at the same time justifying the job of the Air traffic controllers to keep them apart."

Flying on North Atlantic routes, pilots now have the approved option of flying 2nm off track (to the right!) this beautifully solves the head on case and the difference in speeds of following traffic should really make it a non-event.

The problem associated with the non-standard level debate is that unfortunately they are sometimes required to facilitate an expeditious flow of traffic. During the night over central australia there can sometimes be up to 30+ aircraft in a congo line heading to YSSY/YBBN/YMML... Without non-standard levels we would have people cruising at FL280 ect ect... However it must be said that whenever a controller assigns a clearance then seperation becomes HIS/HER responsibility. We can say we are all pissed off/fatigued ect ect but once we plug in and assume control then we are bound by law/duty to perform our role.

Absolutely.
So, having, say, chosen to approve a non-standard level at, say, 3 in the morning (or any time, really) what are you personally going to do to mitigate against the possibility that you might not notice someone coming the other way?
Are there any tricks you use? Does the unit give you any guidance, or a procedure to follow? Are there any automated warning systems? Or is it just down to you keeping your eye on the ball? (Lets face it, that's what is often the case.) There are a handful of controllers around the world who probably relied on that last method, sometimes in adverse conditions, (Like eqp outages, or poor display design) who probably have a few serious regrets.
Not to get too maudlin, here, but really, what do you do?

Unfortunately alot of it has to do with keeping your "situational awareness" ect ect, what I personally do is set a timer (an audible alarm) for 20 minutes before the 2 contacts meet. That gives me enough time to issue the requirement to ensure separation before 10 minutes to time of passing. However at 3 in the morning, the caffeine has worn off, you are on your 9th shift in a row and your neighbours garage band was practicing all arvo and you have had 2 hours sleep.... well, then anything is possible. The scary thing is that in procedural airspace the first alarm we receive is you guys responding to a TCAS RA.

Some areas of the media have reported this as a "Near miss" which is typical of them.

Does anyone know if if was actually a near miss (as reported) or a breakdown of sep and by how much ?

I feel for the individual who was plugged in, a sep breakdown is never a good feeling but well done to the aircrew's SA.

Second hand information tells me the required separation standard was not infringed. This was not due to positive control, the aircraft decided to move themselves off track.

In large pieces of airspace such as this, an unfortunate happpening is that aircraft are at the same level and are known that in the future they will be in conflict. This can be known for hours in advance. There are memory prompts that can be used to remind of who is going to hit who but often the aircraft are only in the airspace that they will break down in for a small time and these memory prompts can not be used. What happens then is that a situation you are on purpose forgetting, get forgotten. As a contrast in a small scale radar environment, a conflict is seen and pretty much immediately fixed.

For the record I do not work on the airspace this occured in and am just summarising a rumour I heard.

In Radar environments in Oz there is Short Term Conflict Alert - 90 second look ahead and not really about collision risk, focussed upon break down of radar separation.

GAFA (no surveillance) no alert yet available. Flight Plan Conflict Probe (utilising pilot reporting) is functional but not yet operationally validated and implemented. It will be as good as the data fed into it. Awaiting system upgrades prior to roll out.

Regardless of how good the alerts might or might not be, the best defence is alert controllers and pilots. Either can make a mistake and what's important is that something is done to fix the mistake when recognised - TCAS, STCA, GPWS etc. are all last ditch defences. The systems should be robust enough that you don't get to the last throw of the dice.

Random tracks work on the big sky theory but navigational tolerances on some aircraft can be huge. Proving separation when not on fixed route structures without significant system improvements will push controller workload in some airspaces so high that there will be an elevated risk of operator errors. Reciprocal tracks at standard levels help, plenty of track crossings where standard levels put you at the same level. A place for everything.

20 years is quite a long time for something to "spring up". The guys and girls that do this stuff are very good. That being said we are all human, yes even those with 4 bars on their shoulders. Operators within Australian administered airspace expect more flexibility, optimum route and levels than they would consider safe in many other parts of the world (indeed in much of the world), they can't have it both ways. Either accept the additional fuel burn or look again at the risk modelling and let us know what you want.

No doubt this incident will attract very close scrutiny. It's not the first time that it's happened, nor will it be the last. The reporting will always strive to sensationalise. I look forward to reading the final report on the investigation. What we don't need is a knee jerk trying to make someone a scapegoat, rather the simple "just culture" premise that we need to focus upon preventing repeat occurrences and treat those involved fairly.

DNC

I very much like simple timers as an aid to situational awareness. They have the advantage of being relatively foolproof, and easy to set.
One thing I used to do in such a situation (when I was an area controller) is also let the pilot know the future plan. I'm a big fan of keeping the crew in the picture.
I don't know the airspace, but is this good old fashioned procedural area control at its best? No (or limited) radar coverage, not many navaids, lots of reciprocal/converging/crossing tracks? (Primary separation tool the good old T10? Or D15 if you're lucky?)
If that's the case it might actually be time to start saying "unavailable" when at all busy, or below par due to time of day etc. I don't know. There is nothing wrong at all with a "can do" culture- I think most of us want to do the best we can for the flight crews, but it needs to be backed-up by a "can do" attitude from management in regard to tech upgrades. Time for radar (or similar), perhaps? The powers that are capable of making decisions about expensive upgrades often can only act or press for funding for same when enough operators complain about it loud enough.
I sincerely hope that's how it pans out. (The "just culture" thing, not the "scapegoat" thing.) For the sake of your workforce, and for future safety in the region.

Yeah, a big "gotcha", that one.
What do you do?
Maybe say "unavailable, one coming the other way at that level"?
Sounds like maybe it needs a bit of work from Aus's finest minds.

Tarq57,

Yep, good old fashioned procedural (now called non-radar) but in a plan-view display scenario. Bugger all navaids and dependent upon aircraft nav capability anything from the suite up to 10 minutes.

I have said it before - and now I'll say it again;

In a modern Nav environment two way routes are a dangerous and stupid anachronism.

I takes only a little imagination to establish race tracks where possible - and where there isn't room for race tracks, formalise the strategic lateral offset.

The large navigation tolerances ATC apply to aircraft are largely fantasy in this day and age - if two are going to hit in the cruise, it will happen on track, not 14NM either side of it.

I agree but this all goes out the window when weather is involved. I have had aircraft up to 300NM off track. Whilst this is extreme, it's not uncommon for a/c to be 50NM left and right at this time of year in northern Aust.

We are told ADSB will fix this and 5NM between a/c can be used. Whilst almost all of Aust has ADSB coverage, unfortunately not all a/c have the gear, so not much use at the moment, except for crossings or change of levels. The gear is to be mandatory by 2013 for all a/c above F290, however I'm sure there will be exceptions just like RVSM.

7, 14, 30, or 50 cross track might be fantasy but ATCs don't get to choose. ICAO sets the standards, all we can do is apply them.

It happened near Tindal and was in radar coverage - 2 acft nose to nose, same level - seems most likely to me the controller f..ked up - but, these days, instead of admitting it, we try to find every other excuse under the sun. :rolleyes::rolleyes::ugh::ugh::ugh:

Oh dear
 

direct.no.speed,
Now Now.Surely none of your mentioned items would ever be thought of as being part of the causal factor ? Surely not ? We are talking safety here,never would a government put cost saving and penny pinching before safety,surely ?

With this incident I note that it was a couple of days after it started being discussed on PPRUNE. :suspect: The incident you mentioned didn't rate a mention on PPRUNE (that I recall) and thus hasn't appeared in the papers.

ADSB takes the pilot out of the loop. I don't know how many times I've been flying along and then look at my TCAS and wonder "who the f%^k is that". Turns we have both been on ADSB on the same track for hours and didn't know each other existed.

We need offset tracks right now!

direct no speed
No, I'm fed up with controllers that won't take responsibility for their own actions. It's become more and more prevalent in the last 10 years - and you provided a list of 7+ other excuses for a f..kup.

Why do you think that might be the case?

Good point.........from someone whose nuts are on the line TWICE :eek:

Your seriously joking i hope.

How many pilots get stood down when ATC may have made an error that they may have caused?
I can tell you about 2 incidents that have happened to me where formal counselling and stand down happened because of Pilots mistakes (im an ATC):ugh:

Amen. I'm over pricks who won't stick there hand up. Get out if you can't handle the responsibility.

I believe the sectors involved are very short of staff. Is this true?

This probably wouldn't even be reported in europe.

Ok, so would be you be happy to be paid $72,000 a year and take the same responsibility?

well said mikk, not everyone is on the top bracket.

Having said that, if you don't like the responsibility, there are plenty of jobs out there where you can turn up, f**k up, and go home. This isn't one of them.