Visual Separation
 

G'day all,

I have a couple of questions regarding the visual separation used in successive visual approaches (ICAO 4444 PANS-RAC Part IV).

How comfortable are you guys (APP) in using them? - in fact, do you use them? I know in the States they're in fairly common use, but what about Europe (assuming VMC)? Are you able to keep a full picture when responsibility is delegated?

Also, are there any 'common' safety issues you've seen (e.g. misidentification of target aircraft)

My purpose in asking this is because I'm involved in designing the ops procedures for ASAS spacing (i.e. the same as visual approaches, but using a cockpit display to find the traffic - ASAS being airborne separation assurance system).... and I'm not esp. happy at it being based on visual approaches which, although being more efficient than controller vectoring most of the time, can have safety implications. Incidentally, under this new procedure, controllers would still keep separation assurance (unlike visual approaches, where it transfers to the flight deck), and it would be an instruction, not a clearance (and the flightcrew would have to accept the spacing task except in emergency). It's currently being trialed in the US and Europe, so may come into practice by 2004/5.

RevStar,

A few observations. I have no problem at all with visual approaches, BUT when the aircraft in question are operating under IFR, responsibility for separation, (in the UK anyway), still rests with ATC. It is easy to maintain a full picture, but some degree of control is lost when a pilot decides to go too far downwind and the sequence is adversely affected. There should be no associated safety issues; successive visual approaches are conducted subject to following only one aircraft ahead, so correct identification should not be a problem.

Visual approaches being more efficient than radar vectoring? I don't think so! Would pilots be happy to sequence themselves 2.5 miles apart on final approach? I think not. And how would they know what speeds the aircraft ahead are flying to be able to maintain separation acccurately? Would pilots know the appropriate vortex wake categories and spacing requirements? What is the difference between an instruction and a clearance in this context?

Would flight crew really want to take on the spacing responsibility after a long haul flight? Is their workload not high enough in the approach phase? And you would be reducing ATC to merely a monitoring function - is this wise?

There is no doubt in my mind that these proposals would impact significantly on capacity issues. What is the driver behind their introduction?

ATCO2 - thanks for your reply.

As I understand it, the definitions of clearance and instruction vary in that through a "clearance", a pilot is authorised to perform something under conditions specified by ATC... through an "instruction", he is told to do something specific. (I'll look it up in the PANS-RAC if necessary).

Regarding visual approaches - I accept that they are not beneficial in every situation (EGLL with the FAST tool is an obvious example)... however, the fact that they are used at all displays an acceptance that they do help sometimes. It may be that the controller's workload decreases, leading to a greater capacity (less buffers added to minimum spacings).

The point about the controllers maintaining responsibility if the flight is IFR should be correct (after all, it is the ICAO mandate). However, it would seem that this is not the case if a visual clearance is given... how can the pilot not have the responsibility for separation if he is maintaining that separation, and has been cleared to do so by the controller? - obviously, there are no ICAO rules to transfer the responsibility for separation assurance, but that is effectively what has been done.

Misidentification of visual traffic has been the subject of a NASA report in the States - I assume it's a common enough problem there (since funding has been given to complete this study) - could any of the ATCOs in the US comment?

Many airports DO use this at minimum spacings (and below min separation - totally unsafe in my opinion)... because the transfer of responsibility has occurred, it is no longer the controller's responsibility to ensure that 2.5NM is kept. This, I agree, makes the controller's role pretty nonsensical, and negates the ATCOs effectiveness (e.g. how do they deal with any failure recovery?). All these problems would be exacerbated by successive visual approaches (i.e. a stream of aircraft maintaining separation form the one in front), which is also PANS-RAC'd.

The point of the ASAS spacing instruction is to alleviate controller workload, increase situtational awareness for the flightdeck and lead to a more efficient arrival sequence. It can be done with merging traffic (by using algorithms to determine future spacing at a merging point) or with in-trail traffic. We're not just talking about IAF onwards here... the procedure would start before the ToD. Because this is before the final approach, the SPACING (not separation) is much higher than 2.5NM/3NM. An initial figure of 8NM has been proposed... this may be less for certain Extended TMAs.

Also, the fact that this is an instruction means that the flightcrew would have no choice about taking this on - it may actually be preferable for them (in the trials so far, it has vastly reduced the number of time-critical speed/heading instructions on approach).

You say that this procedure will reduce ATC to a 'mere' monitoring function - this is exactly what it would do (awaiting flaming for that one!!:D), under certain constraints and conditions. The idea being to decrease controller workload, thereby allowing for more aircraft in the ETMA (obviously, if runway capacity is an issue, this will not actually lead to increased capacity).

Incidentally, this is only one of many initiatives looking to transfer tasks to the flightdeck using ASAS (and ADS-B), and therefore have controller monitor the situation - you ask whether this is wise - I don't know the answer... however, it IS what will happen worldwide in the next few years (2005 ->).

Err.. we don't have a "FAST" tool (well, not at work anyway!!!). The FAST tool has been, shall we say, "put on hold" due to lack of cash. In any event, it had a some way to go before it could be used fully operationally at Heathrow. Our spacing is done using our heads and no machine yet has been able to beat us!

I'm not sure I understand the benefits of ASAS - does it have any in a busy environment such as a major airport?

RevStar - The term "visual approach" clearly means different things to different people.

In the UK it is a means by which a flight may be expedited when the pilot reports that he can maintain visual reference to the surface. As ATCO2 said, ATC remains responsible for the separation from other traffic.

Up until a few years ago we were allowed to authorise a pilot to maintain "VMC and own separation" from other traffic, but this was withdrawn by the regulator on the grounds of being an inappropriate delegation of ATC responsibility.

The only method we have of allowing pilots to provide their own visual separation is by using one of the permissible"reduced separations in the vicinity of aerodromes".

This is used to good effect at Heathrow in suitable weather conditions when both parallel runways are being utilised during arrival peaks.

I cannot honestly see how transferring responsibility for separation to the flight deck could sustain landing rates at airports where High Intensity Runway Operations are performed.


:confused: :confused:

I don't see how we can "instruct" a pilot to follow another aircraft on a visual approach when there are so many variables, such as speed differentials, aircraft weight category, even changing weather conditions. We would have to specify conditions, e.g, no further East than 6 DME, or pilots would go on a sightseeing tour, and seriously disrupt the sequence.

Visual approaches can be more expeditious, and pilots welcome the opportunity to actually hand fly their aeroplanes, but in a busy environment they are just not possible. The controllers'workload would not be decreased because he/she would still have to monitor the visual approach situation and still set up the basic sequence. Not being able to accurately plan on a set landing rate would have serious repercussions for capacity. There would be no constant spacing criteria to plan around - even more critical for single runway airports - what final approach spacing would be provided - this HAS to be within the control of the ATCO. "Less buffers added to minimum spacings" - sorry I don't understand this phrase.

We are not giving a "visual clearance" to IFR aircraft! We would only do this if a pilot cancelled his IFR flight plan and went VFR - not very likely with heavy metal in a control zone, and impossible in a Class A control zone. As stated above - a visual approach is STILL an IFR clearance.

"How can the pilot not have the responsibility for separation if he is maintaining that separation, and has been cleared to do so by the controller?" Are we talking a clearance or an instruction here, or are you confused yourself? ATC RTF phraseology for visual approaches is along the lines of, "report if you get the airport in sight and wish to continue visually." The pilot can accept or decline this offer. If he accepts a visual approach, then he is cleared for a visual approach, usually with a condition to report turning base leg or final. He is then normally instructed to contact the tower. (Approach control still retain responsibility for separation). If a subsequent pilot reports he wishes to make a visual approach, he is given traffic information on the aircraft ahead but not actually cleared for the approach until it is safe to do so with regard to vortex wake and speed differentials. In this way ATC retain control of the separation responsibility. I would never allow a second aircraft to make a visual approach unless I was 100% sure he had the correct aircraft ahead in sight. Although not its primary function, pilots are tending to use TCAS to assist them in this respect.

Are you saying that visual approaches are carried out where pilots maintain 2.5 miles from the aircraft ahead? If this is possible with a continuous sequence of aircraft, then perhaps ATCOs are superfluous - I don't think we are at that stage yet. Neither will we be in 2005!

"A more efficient arrival sequence" - does ASAS look at the traffic mix and decide the best sequence of H, M, UM, S, L aircraft, as currently happens? How are the instructions given? Is the pilot given headings to fly, speed instructions and descents by a "magic box" on the flight deck? Or just left to follow another aircraft from a predetermined range from touchdown? Are the approaches flown totally visually, or by the use of an ILS or another precision or non-precision aid? What about light aircraft and helicopters also operating in the zone? 8 nm separation within the TMA is extremely wasteful and would immediately impact upon holding delays. Would the pilots be able to refuse an ASAS clearance in marginal weather? How about ATCOs maintaining competence in vectoring if the pilots are doing it all the time? What about short notice priority approaches - how long would it take to sort out the sequence effectively? What do pilots think abot the increased cockpit workload? There would be vastly differing levels of competence amongst pilots using these procedures.

I would love to put my feet up and "monitor" the situation, but from what you have told me I am far from convinced of the advisability of this initiative. ICAO have lost a lot of credibility with UK ATCOs recently, cf. "taxi to the holding position (runway)" - which caused no end of confusion; anything they are suggesting should be very closely scrutinised. Let's hear some more opinions about this.

Interesting thread going... However, we find that visual approaches with visual separation works just fine over here. You do get folks to follow others and join the conga line. <shrug> I know that for some who are not used to such procedures it seems the wrong way to do things, but come on over and watch a busy push on a nice VFR day. It works very nicely. When the weather goes down a bit we then just go right back to vectoring to the ILS and not a problem. Sometimes when it is VFR but is a bit hazy we just stick them on the ILS to make it easier on everyone...

As for FAST and PFAST... We appear to be getting away from using it. We have found that it wasn't doing much more than a controller was already doing.

regards

Hmmm... thank you all for your opinions and replies.

Firstly (and foremost), my apologies to the boys and girls at LATCC - I did not mean to imply you guys weren't **** hot at what you do - I'm well aware that you nail the 2.5NM final approach spacing day-in day-out... what we're not trying to do is upset that natural skill. Saying that, I didn't realise you'd discontinued use of the FAST tool - shows what happens when controllers aren't consulted right at the beginning of a project's development (i.e. they won't use it if it doesn't help).

There seems to be (from initial views) a dichotomy between the States and UK(/Europe?)... I was aware this existed, but not how pronounced it was. I think a lot of the ICAO PANS-RACs are developed with the US in mind, hence the problems with them in ECAC airspace. (and the difference in the perceived definitions of visual approach/separation, as mentioned by the 'Duke' above).

A question regarding Scott's post - how do you resolve the IFR / transfer of separation responsibility problems, as mentioned by ATCO2.... in theory, you can't have an aircraft operating IFR, yet maintaining own separation.

I'm NOT advocating visual approaches... I think they've been brought in without any regard for several important issues (responsibility, safety criticality etc).

My original question has pretty much been answered as far as the UK is concerned - any controllers from the rest of Europe care to comment?

ATCO2

Thanks for taking the time to reply to my points - its appreciated. I'll try and answer all your queries...

First of all, ASAS spacing is NOT a visual approach - its merely using that procedure (the US version) as a starting point in the analogy. I asked the original question because I'm not happy with that analogy.

ASAS spacing deals with ToD through to IAF/FAF. It is also an instrument procedure. Obviously, we specify conditions on the ASAS procedure - the only variable being transferred to the flightdeck here is the implementing and maintaining of a set SPACING behind a target aircraft... both aircraft would have to be following the same STAR, or direct WPT clearance. Also, both aircraft would have to have similar approach performance profiles (not too difficult - most a/c on approach are able to maintain similar speeds - the odd Dash-8 or whatever could be slotted in as necessary).

At all times, the controller will be responsible for designing the solution (to the sequencing problem)... the pilot will only be responsible for implementing it. For example, there may be converging aircraft from 2 different STARs at a single IAF... the ASAS spacing instruction would allow a pilot to identify the target aircraft (possibly 50NM away at this point), input the desired spacing at the IAF passed by the ATCO, and then fly the aircraft (manual or autothrottle) to get to the IAF that distance behind the target a/c (algorithms will help in calculations here - this will be hooked up to the FMS). This would then negate the need for vectoring or speed instructions by the controller for that aircraft (trials HAVE been done - it does virtually negate these types of communication for a busy Extended TMA).

"Less buffers" means less additional spacing used in case of garbled comms / busy VHF etc. There may be a number of time critical instructions that have to be given in the current airspace - this will decrease if the controller takes advantage of this tool.

It's important to note that this is a controller tool - you guys could use it as and when you like. We're not trying to design something here that will threaten you - just help you.

Hopefully, Scott's reply shows where the difference in terminology lies - and the different attitudes towards visual clearances (i.e. successive visual approaches used in the States).

I meant for visual clearances, not for the ASAS spacing instruction. They are two very separate entities - possibly I've confused things by mentioning the two together. ASAS spacing will build on the usefulness of visual approaches, but use IFR and no transfer of separation responsiblity.

To answer some of the mountain of questions in the penultimate paragraph (!)
- the controller does the sequencing.
- instructions will be passed at first by VHF voice (with information on a/c being uplinked by ADS-B, a new datalink technology in use worldwide). It is hoped that eventually something very similar to CPDLC will be used.
- in marginal (read bad!) wx, the controller would not offer the procedure... since it is an IFR procedure, it can be done in IMC though (doesn't rely on visual acquisition of the traffic).
- ATCOs maintaining competence is one of the issues that human factors people are currently looking at... since it is only a tool, vectoring can still be used if the ATCO wants.
- pilots and controllers would be trained in this procedure. On the tech-log forum, there is a worrying post about CPDLC - that some airlines train their pilots in using it, and some don't. This shouldn't happen.
- ICAO aren't suggesting it... amongst others, virtually all the European CAAs, EUROCONTROL, the FAA, NASA etc etc are working at developing ASAS procedures to be implemented circa 2005.

I hope this has helped in the understanding... carry on replying everyone - its something that WILL happen sooner rather than later.

There´s a big difference in how things are done in Europe and in the US. As Mr. Voigt points out maintaining own visual separation works fine and is very efficient, especially, in my opinion, when very different types of aircraft are involved, such as a small commuter following a heavy. The little guy, without ATC mandated separation, can stay higher and faster much longer, and touch down at a point well beyond the heavy and thereby avoid any wake. Go to any major U.S. airport with a variety of traffic and watch the guys plant their wheels the moment the aircraft ahead clears the runway.
In Europe this is almost unheard of, and most pilots are loath to accept the added responsibilty (a culture thing?).
The only airport in Europe where this wouldn´t add any benefits is LHR, as the controllers there are allready keeping spacing to a minimum (well done).

I think that another factor in the US is the use of a dedicated arrival runway at most busy fields; in such situations arrival spacing is often (not always, though) of little relevance.

EDI is obviously a far quieter field than LHR, but for us visual approaches are a useful tool - they reduce the controller's workload (in my experience) and actually increase our capacity as almost all pilots will fly a tighter pattern than we would have given them under vectors. Under the banner of "reduced separation in the vicinity of the aerodrome" we would routinely clear the first two in a sequence for visual approaches (always in response to a pilot request). Mis-ident of traffic ahead by crews is (again, in my limited experience) not a factor. I can remember it happening once, but (in fairness to the crews) the controller made a very poor call on that occasion.

As you could tell by looking at our handbook, we kind of look at ICAO doc 4444 as a very nice guide <G>. We don't use it much nor do we use a lot of other ICAO procedures. We look at what works here and use those. There are those who want to make our book look much more like ICAO's but there is indeed resistance to that for many reason...

As to handing over separation responsiblity, we do that when the aircraft reports the other aircraft in sight and we tell him to follow him. It is the ALL the pilots responsibility to follow and avoid...

regards

Scott,

you forget that one of the biggest differences between you yanks and many of the rest of us is that we can't give a landing clearance while the runway is occupied (or there are another 8 ahead!). Yes I know that there are special procedures for the Heathrows of the world but not for all of us.

Needs to be remembered sometimes.

Cheers.

;)

Guy;

I don't forget that at all <G>. In fact, that is one thing that we have been doing here for a little while and it is one of the things that I am not fond of. <shrug> However, it is legal and probably most of our folks use it...

regards