display | more...

It's a little difficult not to feel a bit fraudulent, spending time noding subjects that are perfunctorily covered in less than an hour during one's air traffic control training, and applied with little thought by tens of thousands of people, millions of times every day, worldwide. Still, as I sit here, listening to my remixed 1990s computer music, I find all I can think about are vertical separation standards.

No, I haven't had sex in a little while now - why do you ask?

archiewood clears his throat

Vertical separation is the bread and butter of air traffic control. It is always, always, always the separation method we go for first. Every other separation method in existence is a fallback for an inability to provide vertical separation. Why? Because the controller in the pilot-controller relationship is like the male of any male-female relationship: they just want a quiet life. I wonder if the fact that 80-90% of controllers are male has any relevance here.

A "quiet life" for an air traffic controller is just what it implies - the necessity for as few instructions as possible. The less time the controller has to spend talking to one aircraft, the longer they can spend talking to others. More aircraft can be accommodated so more aircraft can fit in the sky; that always makes the bean-counters happy, and if thousands of tons of formated, hurtling metal can be put to use keeping a controller's mind peaked, that is also a bonus.

If a controller has vertically separated all of their aircraft, that means they don't have to use the dreaded headings - aircraft can just fly wherever they like, which as we've established before can only be a good thing. They know which direction they're going, so why should the controller care? They're vertically separated! Go where you like! Almost!

Though relatively simple, there are a few provisos for the, er, provision of vertical separation.

First of all — and this is a no-brainer, really — the aircraft involved must be using the same altimeter setting, otherwise the altitude readings the controller sees on the radar may not be correct. Without getting into all sorts of nonsense like transition levels, transition altitudes or flight levels, we're basically saying both aircraft need to be using the same reference point to calculate their altitude.

Now, what of the standards themselves? I won't go into where they're applied, because I've covered that elsewhere. Starting simple, we've got the oft-quoted (by me, anyway) rule that all aircraft not horizontally separated must be separated vertically by one thousand feet.

One thousand feet really isn't that much. A statute mile is 5,280ft. The typical horizontal separation requirement is five nautical miles.

The difference between the minimums for horizontal and vertical separation reflects the difference between the potential horizontal and vertical speeds of aircraft. Civil aircraft may fly at anything up to ~500 knots (575mph), but even the best of those (including ATC favourites of the Learjet series, and pretty much every jet made by Gulfstream Aerospace) won't typically climb faster than 4-5,000 feet per minute, and even those will not reach that climb rate right away. So, 1,000ft is reasonable for vertical separation; considering the time it would take to erode that separation to a dangerous point, we've got a few systems that can help us out in time to reestablish it.

Of course, this wouldn't be an ATC writeup without a quivering mass of provisos, clauses, regional variations and exceptions. Here's the first of them: 1,000ft vertical separation only applies up to FL290 (flight level 290 - about 29,000ft) in the UK. Why? First, because aircraft at or above this level tend to be flying faster, meaning any altitude deviation will erode separation more quickly. Secondly, altimeter inaccuracies 'stack', to use the nomenclature of Dungeons and Dragons, EVE and other such roleplay nonsense. In other words, the higher your aircraft is, the greater the deviation from your desired altitude that may be caused by a given inaccuracy in your altimeter.

Above FL290, aircraft are required to be separated vertically by at least 2,000ft.

Now we get to the provisos of the provisos. From 1997, a standard called Reduced Vertical Separation Minima (RVSM) has gradually been adopted, to the point where as of 2005, for most purposes it applies worldwide. As aircraft instrumentation became more accurate - and this applied particularly to airliners flying long distances with comprehensive equipment sets - 2,000ft separation started to look a bit conservative. So, the separation minimum was reduced to 1,000ft after much legal wrangling, tentative agreements, back-biting, red tape, assassinations, bureaucratic foot-dragging and blackmail. Possibly.

To take advantage of this reduced separation, the altimeter and autopilot carried by an aircraft must be certified as sufficiently accurate. Several height-monitoring units exist to verify that aircraft are complying with their RVSM requirements; the UK has one of these, located by the Strumble VOR (which is the crossing point for the majority of the traffic to and from the North Atlantic). For aircraft that aren't RVSM-capable, the vertical separation minimum remains 2,000ft. RVSM airspace stretches up to FL410 - about 41,000ft. Above this, the 2,000ft separation minimum is universal. More on that lie shortly.

Although it's not that common, non-RVSM aircraft do occasionally nose into RVSM airspace, and this is rather a nuisance for controllers (that is, beyond the nuisance posed by aircraft in general). Controllers working aircraft in RVSM airspace get used to separating them vertically by 1,000ft, and if a non-RVSM aircraft comes along it must be separated from aircraft above and below it by 2,000ft. This requirement blocks off two levels from use, reducing the capacity of a 5,000-foot band of airspace from five aircraft to three. It's also a pain to remember, since it's only indicated by a single character on the aircraft's flight progress strip. Certain radar systems do also indicate it with a symbol which appears next to the appropriate blip, though.

Supersonic Flights

Although this provision is virtually redundant now (even military aircraft usually stay out of civil airspace, and no SSTs are presently operating) I'm including it for completeness. And hopefulness, I suppose.

The easiest way to summarise the provisions for supersonic flights is to say that if an aircraft is flying at supersonic speed, the vertical separation requirements for that aircraft essentially double compared to what has been mentioned so far. Furthermore, supersonic flights cannot take advantage of RVSM.

Up to FL450 (about 45,000ft), if an aircraft is flying at supersonic speed, it must be separated from others (supersonic or not) by at least 2,000ft. Above that, it must be separated by at least 4,000ft.

There's a further provision here, which requires a little preamble; if you've ever seen a controller's radar screen, you may have noticed most or all of the 'blips' have groups of numbers next to them. This information is provided by transponder for that particular aircraft; the transponder 'replies' to interrogations by a ground-based antenna, and equipment in the radar system ties that information to the appropriate blip on the screen. One such bit of information (if the aircraft's transponder is suitably capable) is the aircraft's altitude or level: the altitude-reporting facility of a transponder is called 'Mode C' or 'Mode Charlie'. Mode C information is accepted to be accurate to within 200ft of the aircraft's true altitude.

As an aside, if you take this to the extreme, it's totally possible that two aircraft which appear to be separated by 1,000ft are actually only separated by 600ft; one aircraft reading 200ft above its actual altitude, and the other reading 200ft below its actual altitude. Nice.

moving swiftly on...

The UK Manual of Air Traffic Services (MATS) Part 1 goes on to say that since the Mode C of some military aircraft is only accurate to the nearest 400ft during supersonic flight (i.e. half the typical accuracy), military controllers may have to apply 4,000ft separation at any level.

As another aside, the 400ft figure also pops up in rules for assessing an aircraft's level by using its Mode C readout. An aircraft that's climbing or descending can be considered to have passed a particular level when its Mode C readout indicates 400ft above or below that level (as appropriate) "and is continuing in the anticipated direction." This means that if two aircraft are only separated horizontally, one climbing and one descending, the controller must wait until their altitudes are showing a difference of 800ft before giving any instructions that would lose horizontal separation.

The quirk produced by the wording is that if one of the aircraft is in level flight and the other is climbing or descending, only 400ft of vertical separation is needed before horizontal separation can be lost! This being the opposite to what makes logical sense.

In practice, these parts of the rules are applied fairly loosely; this and certain other deviations or bending of the rules are quite justifiable. For instance, if two aircraft separated by 1,000ft have just passed each other flying in opposite directions, the rules state that five miles of horizontal separation must exist before vertical separation is eroded. In practice, the controller will rarely hesitate to lose vertical separation before five miles has been reached. Short of spontaneous teleportation, there's absolutely no way the aircraft could possibly hit each other and added to that, radar information is delayed by several seconds so considering the speeds involved, the aircraft will already be well past each other by the time they appear to have passed.

The Introduction of Section 1, Chapter 1 of MATS Part 1 says, and I'm paraphrasing, "a controller can do what they like as long as they can justify it." For what it's worth, you do generally have to demonstrate you understand and can apply the rules, to the letter, before you get to bend them in this way.

All the nausea above is summarised in this handy diagram:

          Both aircraft        Either aircraft 
            subsonic:            supersonic:

      |                    |                    |
      |                    |       4,000ft      |
      |      2,000ft       |                    |
      |                    +--------------------+ FL450
      |                    |                    |
FL410 +---------+----------+                    |
      |  RVSM   | Non-RVSM |                    |
      |         |          |                    |
      | 1,000ft | 2,000ft  |       2,000ft      |
FL290 +---------+----------+                    |
      |                    |                    |
      |      1,000ft       |                    |
      |                    |                    |

Source: CAP 493

As the number of pipelinks in an ATC writeup that link to my own writeups increases, the probability of an ATC metanode approaches one. You have been warned.

Log in or register to write something here or to contact authors.