Heathrow Tower

I have absolutely no idea! I'd guess there is no mitigation, it's just based on likelihood, but that is speculation on my part. I will try and find out

Aha, I've not read it through myself, but here you go, some bedtime reading!

https://www.faa.gov/documentLibrary/...AHSO_Order.pdf

chrismk

Great thread - I even managed to understand it!
Thank you for taking the time.

umbyboy

Cheers , very informative!

Shuttle_Endeavour

Are any operators using RNAV / GPS approaches routinely? And do these allow you to reduce spacing as the aircraft isn't using the ILS?

mad_rich

Thanks Heathrow Tower. Fascinating to read, and well written.

As the fog clears, only to be replaced by strong winds, how does that affect you? Tomorrow is forecast for very strong Southerlies up here in the North. Not so bad down at LHR, but how do you cope with Southerlies?

I'd also be interested in to what extent language barriers are a problem for you guys. Listening to some ATC videos online, it sometimes terrifies me how many misunderstandings there are.

ATCO2

I'll leave Heathrow Tower to explain TBS and how strong winds affect approaches.

As for the language barrier issues, it's really not that bad at all. All controllers/pilots must pass an ICAO English Assessment. Most videos online I assume are taken from America. Phraseology there is a lot more colloquial and 'off script', and often the delivery of clearances is overly quick. Obviously this doesn't help foreign pilots who might have a pretty limited grasp of English. The percentage of incorrect read-backs is pretty low, the most common problem is pilots not picking up their callsign when you're issuing an instruction, and having to keep repeating yourself.

Heathrow Tower

Yes, we have RNAV(GNSS) approaches at LHR which do not rely on ILS.

However, they are not approved for use in anything worse than CAT I conditions, so when it's foggy we can't use them.

For that, GBAS (Ground Based Augmentation System) is the future. However, that technology is again not yet approved for CAT II/III ops yet, but should be in a few years.

GBAS consists of a ground station (that doesn't have to be in line with a runway like ILS) that locally broadcasts error correction to the GPS signal so it is far more accurate than just relying on GPS. We're talking accuracies in the order of a 20cm.

One station can 'contain' 48 procedures, so one station could cover one whole airfield, rather than the four localisers and four glide paths that make up the ILS systems at LHR.

I can bore anyone with GBAS talk, I sit on various working groups and panels regarding its introduction.

I'll answer the wind question later this evening.

Nimrod1965

Purely professional interest, do you use the Mode 3/A SSR code to correlate on the ground radar or Mode S Downlinked Aircraft ID?

Heathrow Tower

Strong winds:

Right, there are two aspects to this; crosswinds and headwinds (or tailwinds I guess!), and they happen at the same time to a greater or lesser extent.

I'll take the easier one first.

Headwinds and their impact on arrivals.

Apart from LHR (since late April 2015), all other radar controlled airports in the world provide spacing between successive arriving aircraft in terms of distance.

This distance might be the required wake vortex separation...see this video to visualise wake vortex:

The distance might be the minimum radar separation, or perhaps a runway occupancy-based spacing to enable the second aircraft to receive a landing clearance (or perhaps on a mixed-mode runway, to permit a departure in the gap ahead as at Gatwick).

If we take the wake separations in the UK, it depends on which categories each aircraft of the pair is in, eg:
Super: A380
Heavy: 747, 777, 787, 767, A340, A330, A350
Medium: 757, 737, A319/20/21

Super then Heavy is 6nm, Super then Medium is 7nm, Heavy then Heavy is 4nm, Heavy then Medium is 5nm.

So Approach Radar will give heading, speed and descent instructions to arriving aircraft with the objective of achieving these distances. The problem is they stay the same whether there's a 5kt headwind or a 30kt headwind. If you imagine those stretches of motorway where you have to keep 2 chevrons between you and the car ahead. Hopefully you're all familiar?

Now imagine you're standing at the side of the road watching the cars pass by, always 2 chevrons apart. If every car travels at 70mph that means that over a period of time, you'll have a 'x' number of cars go by. Imagine now that they all slow down to 35mph, but still keep 2 chevrons apart; you'll now have '1/2x' number of cars passing you. They are the same distance apart, but they are travelling more slowly and therefore take longer to pass you.

This is what happens is strong headwinds. The aircraft travel more slowly relative to the ground, but keep the same distance apart, therefore fewer aircraft land. In 5kt headwind we might land 42 in an hour, in a 30kt headwind we might only land 36. Not good.

So this is where Time Based Separation comes in (introduced in April 2015). This system takes both the time equivalent of the distance separations mentioned above in a 5kt headwind (roughly 22-23 seconds per mile) and wind data from aircraft transponders (worked out from GPS by the aircraft), and then modifies the distance so that it corresponds to the required time equivalent.

So, let's say that a 4nm separation gives 90 seconds between arrivals under the old system on a 5kt headwind day. In 30kt headwinds, the tool will work out what distance would achieve a 90s gap, and arrives at 3.2nm instead of 4nm. Then the tool will draw a line on the radar display 3.2nm behind the lead aircraft, for the radar controller to use as a reference. At the moment we only compress the wake vortex spacing, but future developments will broaden the scope. On average, throughout the whole time of operation (including calm or tailwinds, when we actually increase the spacing over what was provided before), we have gained 0.8 arrivals per hour. In periods of strong headwinds it's not unknown to gain 3 arrivals per hour.

It's same as if our roads could change the distance between the chevrons when a variable speed limit had been applied: 100m apart at 70mph, 85m apart at 60mph, 60m apart at 50mph etc etc.

For anyone who wants to view the corporate stuff, have a look here:
http://www.nats.aero/tbs/

I've been working on it now for 8 years, and it's still being developed. Any other/further questions, please fire away. I'll come back to crosswinds at some point in the next few days!

Heathrow Tower

SSR code.

Nimrod1965

Thanks.

Heathrow Tower

I should have been a bit more verbose! The identity is generated by the SSR code, but the secondary position is generated by triangulating the Mode S transmissions.

Nimrod1965

Thanks again.

NoTiersForMe

Amazing thread Heathrow Tower. If one was to use an anatomical/dietary metaphor for your individual posts here one might say they have 0% body fat or are 100% lean information. I really can bore friends at the pub even more off the back of this.

I came to this thread via a referral from a current fog thread and am glad Prospero has given this thread its own icon on the stickies.

I was querying the other thread about impact on takeoffs during fog. You introduce concept of "free flow" in takeoff procedure at post #63 with the discussion of handover between ATC and radar in regular ops and how that might affect control of when you next give takeoff clearance to next aircraft. But in fog are we more likely to have go arounds of inbound aircraft and so that knocks on to when you handover for some [in this case] southbound departures crossing 27L. I guess I'm curious about any other fog/weather issues for takeoff. Or, simply, the delay of the inbound aircraft is biggest influence of the delay of the same plane making up the corresponding rotation outbound.

Heathrow Tower

I have written a post in the past about flow control and slot times, but I can't find it right this moment!

The following is a simplified description. Flow control (sometimes referred to by its acronym as ATFM or ATFCM -Air Traffic Flow (Control) Measures) which forms part of the overall Network Management, is an incredibly complex beast.

Firstly, let's talk about 'slot times'. There are two types, one is an airport slot or schedule slot. These are the rights given to an airline to fly from an airport within a certain time window. These are what LHR is short of, and LHR slots are often so valuable they are their own asset class, often changing hands amongst airlines for tens of millions of pounds.

Forget those, we are now going to talk about the ATC slot, or it's official title of Calculated Take Off Time (CTOT). This is a time, with a tolerance of -5/+10 minutes, within which a flight has to take off.

We'll consider a flight from FRA to LHR, all figures quoted hereafter are for illustration purposes only!

In Brussels there's the NMOC, the Network Managment Ops Centre. NMOC has a very large computer that effectively holds all the flight plans for any flight going on in Europe. So NMOC know airport of departure, estimated time of departure, route, speed, airport of arrival.

Also, each airport in Europe will have declared an arrivals capacity to NMOC, in terms of aircraft per hour.

Airspace is divided up into sectors (both vertically and horizontally) by each ATC organisation. Each sector will also have a maximum capacity in terms of aircraft per hour, again, this will have been declared to NMOC.

So, thinking of our flight's route from FRA to LHR, the West departure sector of Frankfurt may declare a capacity of 60, the NW Germany mid-level sector a capacity of 60, NW Maastricht Upper sector; 80, London Clacton sector 45, NE London mid-level 45, Heathrow arrivals 40.

The NMOC computer has all this information. It knows, at any given time, where all the aircraft should be, and it what sectors, and what the capacity of those sectors is. On a benign day, then NMOC might realise that for the hour in which our FRA-LHR flight is esitmating to land at LHR, there are another 45 aircraft estimating arrival. This makes 46, but the capacity is only 40.

NMOC will then start issuing CTOTs to move 6 of those aircraft into the next hour, so delaying the take-offs of the last 6 aircraft in the originial hour by a few minutes each. NMOC will probably also issue CTOTs to most of those in the original hour, but with zero delay, just to ensure they take off and use the slot, and don't depart late causing the same problem in the next hour. Let's imagine our FRA-LHR flight is one of those that is going to depart in the original hour, but there's a flight BRU-LHR which has been given a 10 minute delay on its CTOT to move its arrival to the next hour. If, for some reason (tech, baggage, slow boarding) our FRA flight misses it's slot, then the NMOC computer will start trying to find another canidate to use our place at LHR....BRU is closer to LHR, so it could take our place, so it gets a slot improvement and takes our place in the first hour's arrivals into LHR.

Luckily, the BRU flight was boarded on schedule and the flight/cabin crew and passengers were all ready to go when the improvement came through. In this situation, even they knew they had a delay, as soon as they were ready, the BRU flight crew would have asked ATC to send a 'Ready Message' which tells NMOC that they can take advantage of a slot improvement as in this case.

The above example shows the process with only one (LHR arrivals) constraint on a route. Now imagine the complications that arise when the numbers of flights means that all of the FRA-LHR's flight's sectors are at capacity, and moving a flight's arrival time into LHR back by 10 minutes might then cause another sector on its route to be over capacity, and thus the flight picks up even more delay. Now imagine a flight from Denmark/Norway to France/Spain crossing this route at right angles, or the flights from outside of Europe that have already been airborne for 10 hours, which need to be worked around.

And all this is on a benign day. Imagine the complications when there are thunderstorms around Brighton-Dover, and instead of just the flights from/to central and eastern Europe and London using this route, the flights that would normally go to/from France, Italy, Africa and London have to use the same route, thus doubling the amount of traffic flying through these already 'at capacity' sectors.

And in bad weather, sectors and airports will often reduce their declared capacity. At LHR we might reduce the arrivals capacity to 38, 36 or even 32 per hour. If you imagine lots of the sectors around northern Europe do this, then you can see very quickly hour delays of multiple hours quickly materialise.

CTOT delays, and significant CTOT changes, are common in such scenarios. In ATC we just get a time on each flight's data display in front of the controller. All the calculation is done by the supercomputer in Brussels. We don't just 'decide' who goes first. We have to abide by the CTOT.

On a benign day, then there is a very limited opportunity to do a bit of 'CTOT-swapping', but only on very simple situations within one airline. So for example, if the LHR-GLA flight has a 10 min delay, and the LHR-EDI has no delay, and the constraining sector was in the Midlands where both aircraft were going to fly through, and the EDI flight was not going to be ready for its CTOT, but the GLA was ready early, and the swap wasn't going to affect any other flight at all, then it might be done. You can see the number of factors that need to line up there!

Hope that's explained it a bit. Happy to answer any questions. Like I said, that's a very simple version of reality, but hopefully it gets the idea across. Mods, happy for you to separate out if you wish!