Runway End Safety Areas (RESA)

[Dagger Dirk]

One of the contentious matters to emerge from the Garuda 737 crash in Yogyakarta is the fact that the existing RESA of 60m is far short of the ICAO minimum. That brings up the matter of just how they should arrive at a specific RESA that's appropriate for a particular runway.

Questions:

a. What is the ICAO minimum?

b. What is/where can I locate the exact reference? (not having a copy of Doc 9197)

c. How is it linked to the aircraft type and runway length?

d. Does ICAO itself specify any alternate means of compliance such as EMAS? i.e. short of reducing declared distances (TODA/ASDA/TORA etc)

[OverRun]

a. What is the ICAO minimum?
ICAO upgraded the requirement for a 90m runway end safety area (RESA) from a recommendation to a requirement (in 1999). There is a further recommendation for a 240m RESA, but this is not yet a requirement. It will be when the country, that manufactures EMAS systems and flies jets into ridiculously short runways in blinding storms, has its way. The RESA shall be 90m long and twice the runway width.

b. What is/where can I locate the exact reference? (not having a copy of Doc 9197)
ICAO Annex 14 – or for Australia, MOS139, section 6.2.25.

c. How is it linked to the aircraft type and runway length?
ICAO say

:
A runway end safety area shall be provided at each end of a runway strip where:
— the code number is 3 or 4; [this is the aircraft code – typical 3 is a B737, 4 is a B747]
— the code number is 1 or 2 and the runway is an instrument one.

The Australians (and possibly others) are less stringent and link upgrades partly to international flights.

MOS 6.2.25.2 The new RESA standard shall apply to all new runways and existing runways when it is lengthened. Operators of existing code 4 runways used by air transport jet aeroplanes conducting international operations must make provision to comply with the new RESA standards within five years of the promulgation of CASR Part 139.

d. Does ICAO itself specify any alternate means of compliance such as EMAS? i.e. short of reducing declared distances (TODA/ASDA/TORA etc)
Not in Annex 14, yet. Well gosh and shucks it sure enough will come one day. ICAO say in Attachment A-8 to Annex 14:

9.2 Where provision of a runway end safety area may involve encroachment in areas where it would be particularly prohibitive to implement, and the appropriate authority considers a runway end safety area essential, consideration may have to be given to reducing some of the declared distances.

RESAs have impacted on airports across the world, and the responses depend on circumstances. At some airports, there was sufficient land at the runway end so the RESA could easily be constructed. I'm presently building RESAs at a number of such airports in Africa, along with general runway upgrades. At other airports, the runway declared distances (not the physical runway itself) are reduced in a paper exercise to provide for the RESA. I've done this at other airports where there is no land available for extension. Umm, and those who have looked at EMAS systems have come away shaking their heads in unhappiness. Lovely to see, lousy to own.

[hvogt]

a. What is the ICAO minimum?

ICAO Annex 14, volume I, chapter 3, paragraph 3.5:

3.5.1 A runway end safety area shall be provided at each end of a runway strip where:
— the code number is 3 or 4; and
— the code number is 1 or 2 and the runway is an instrument one.

3.5.2 A runway end safety area shall extend from the end of a runway strip to a distance of at least 90 m.

3.5.3 Recommendation.- A runway end safety area should, as far as practicable, extend from the end of a runway strip to a distance of at least:
- 240 m where the code number is 3 or 4; and
- 120 m where the code number is 1 or 2.

3.5.4 The width of a runway end safety area shall be at least twice that of the associated runway.

3.5.5 Recommendation.- The width of a runway end safety area should, wherever practicable, be equal to that of the graded portion of the associated runway strip.

b. What is/where can I locate the exact reference? (not having a copy of Doc 9197)

See above.

c. How is it linked to the aircraft type and runway length?

ICAO Annex 14, volume I, chapter 1, paragraph 1.6:

1.6.1 An aerodrome reference code — code number and letter — which is selected for aerodrome planning purposes shall be determined in accordance with the characteristics of the aeroplane for which an aerodrome facility is intended.

1.6.2 The aerodrome reference code numbers and letters shall have the meanings assigned to them in Table 1-1.

1.6.3 The code number for element 1 shall be determined from Table 1-1, column 1, selecting the code number corresponding to the highest value of the aeroplane reference field lengths of the aeroplanes for which the runway is intended.

Table 1-1 (excerpt)

Code number 1: Aerodrome reference field length less than 800 m
Code number 2: Aerodrome reference field length 800 m up to but not including 1200 m
Code number 3: Aerodrome reference field length 1200 m up to but not including 1800 m
Code number 4: Aerodrome reference field length 1800 m ond over

[OzExpat]

Just remember that nothing in any ICAO Annex means anything unless it's implemented by the State in question. It is quite legitimate for a State to file a difference with any requirement in any Annex, so you should check the AIP for specific State's requirements and, particularly, its differences.

[old,not bold]

OverRun's post is absolutely right, and he/she does indeed sound like a Professional Airport Engineer.

The point he does not make is that a longer RESA does NOT necessarily mean a safer runway.

Going back to basics, the RESA is there mostly for the accelerate/stop case, from V1. (Yes, there are other reasons, but that's the crucial one).

In that particular case, without the jargon, you work out the weight you can start the T/o roll at, so that when V1 comes up you've still got enough room to stop safely within the remaining paved runway plus the RESA, if you decide not to take-off (using the TODA), and to stop instead.

If you extend the RESA from 90m to 240 by going further out into the country from the existing runway, you simply move the point at which V1 is reached another 150m along the paved surface from start of roll. Now, if you reject the take-off at V1, you hit the RESA surface with 240m of your stopping distance left, instead of 90m, and therefore at a much higher speed and in a heavier aircraft than you would have with the 90m RESA. (Other things being equal etc etc).

As OverRun says, many airports cannot physically extend a RESA, and therefore shorten the T/o Run Available by ending the take-off run 150m before the end of the existing paved runway, thereby reducing operating weight and speed at the point where the paved surface ends. That's OK, and safe.

There's another factor working here. The good book says that RESA's must be able to support the heaviest aircraft using the airport without damage to the aircraft. The whole philosophy of having a RESA depends on that being the case. You, in your fully laden A380, B747 or whatever, should hardly notice as you go off the paved surface and on to the RESA. Your braking action should be unaffected by the surface, and the wheels should not sink in by a centimetre, because doing so would stress and damage the undercarriage..

In your dreams. Few if any RESA's are even built to these standards, let alone tested. If they were, the airport operator would probably simply extend the paved surface with the appropriate bearing strength, because that's not much more expensive.

If you want to check this out at an airport of your choice, try and find, beneath the layers of shopkeepers and accountants, the person who understands RESAs and aircraft safety, and ask him or her what the RESA bearing strength is, when it was last tested, wet and dry, and who by. If they are expecting the A380, this is a very relevant enquiry. At many small regional airports in UK and Europe (to my knowledge, perhaps also elsewhere in the world?) you would get a blank look.

For new airports, one of which I was recently involved with, all the planners are doing is looking at the maximum accelerate/stop distance they can use/need, and then paving that less 240m at each end.

The beancounters love it; no performance penalty and much cheaper.

With ultra-heavy aircraft now in normal use, and an even heavier one on the way, it seems to me that the philosophies that worked for DC7s need a review.

[Musket90]

old not bold - are you not confusing the RESA with Stopway when referring to V1. Runway declared distances which have an accelerated stop distance available (ASDA) length more than the TORA means that the runway has a stopway which has different characteristics to a RESA. The RESA does not form part of the runway declared distance and therefore should not be considered when calculating V1. It's the TORA or ASDA figure that should be used.

[safetypee]

I recall that there is advisory information which recommends that operators make allowance in their landing distance calculations where a runway safety area does not meet the ICAO standards. Can anyone point me to a reference.

[Old Smokey]

I'm in agreement with Musket90 that old not bold may be confusing RESA with Stopway. Every inch of the Stopway must be able to support the weight of the aircraft, although it's surface MAY be damaged by the aircraft.

RESA poses a physical constraint on how much of the Stopway may be used for in determining the accelerate-stop performance of the aircraft. As an example, for the airports that I work with, although the Stopway may be, for example, 150M in length, not more than TORA+60M may be used for the Accelerate-stop performance calculation.

Regards,

Old Smokey

[john_tullamarine]

OverRun's post is absolutely right, and he/she does indeed sound like a Professional Airport Engineer

That he most certainly is .. which is why I flick any runway engineering questions his way ...

[old,not bold]

It's sackcloth and ashes time.....

I was indeed confusing RESA and Stopway, and I've just finished banging my head against a wall to try and erase the embarrassment, without success. I'm supposed to have airport manager, inter alia, on my CV, for God's sake.

What I was getting at was not the RESA, ie the subject of this thread, and the longer the RESA the better, of course, and of course this has no efffect whatsoever on V1.

I was getting at the stopway, and at the risk of drifting the thread, my point is that the trading of pavement, as "an economical substitute", is only valid if the stopway can indeed "support the aeroplanes it is intended to serve". My personal view is that it should do a lot more than that if the ASDA, balanced field length or whatever, includes it, and I do not believe that many stopways even fulfill the "support the aeroplanes..." need.

A Stopway "need not have the same bearing or wearing qualities as the runway with which it is associated". I have always thought that this is a fudge. Why not? It is there, says the book, to be "a suitable area in which an aeroplane can be stopped in the event of an abandoned take-off".

Strangely the book does not say "safely", but we can assume that it is implied. A very heavy aircraft, at its maximum permitted mass, takes some stopping from V1, especially if the reverse thrust is asymetric, or would be if used at all. (Is another fudge the manufacturer's figures for stopping from V1? I don't know, but I often wonder.)

If the airport owner is minimising TORA while maximising ASDA, ie the "economical substitute", the aircraft will hit the stopway at some speed, while carrying out full braking. Even dry, on many stopways the nose and main gear will sink a little. After heavy rain it could be much worse.

So I stand-by the assertion that at many airports it is only "in your dreams" that a big, heavy aircraft can be brought safely to a halt without a risk of undercarriage collapse at least, within the stopway, after abandoning a take-off at V1, especially if the aircraft is a heavy twin with a failed engine.

But I apologise to all for the drivel that was my previous post on this subject.

[Sir George Cayley]

Time for me to state the obvious..

..when landing, the amount of runway behind you is of no consequence.

And talking about ASDAs - what about the runways that offer an intersection take-off? How many times do you see a/c tight on slot take an intersection departure - have done the calcs? or is it opportune?

In the UK CAP168 Chapter 3 deals with physical characteristics and largely follows Annex 14. Interestingly over the past year my Jepps UK airfield updates have had quite a number of shrinking runways published.

Nothing to do with ICAO coming see how the Belgrano works later this year?

Sir George Cayley

[mutt]

especially if the reverse thrust is asymetric, or would be if used at all.

Not used as per FAR25.

the aircraft will hit the stopway at some speed,

Not sure if i agree, the maximum stopway is half the takeoff flare distance, so i doubt that its long enough to say that an aircraft enters it as some speed!

Sir George Cayley

We publish performance data for all regular intersections, otherwise takeoff is not permitted.


Mutt

[Mad (Flt) Scientist]

Not sure if i agree, the maximum stopway is half the takeoff flare distance, so i doubt that its long enough to say that an aircraft enters it as some speed!

Well, if you assumed a 160m stopway and 1/2'g' decel (reasonable for a max effort stop) then assuming you are stop-limited, you'd enter the last 160m of the decel at about 40m/s - something like 75 knots. That's pretty fast.

Granted the constant decel is a bit of an approximation, but it won't be far off. Even if the decel were halved, the entry speed would only drop to about 50 knots or so.

[OverRun]

These aircraft speeds seem to be pretty close to the RESA designer assumption. The FAA (in their AC150/5220-22A on EMAS systems) call for the design entry speed to be 70 knots.

To the maximum extent possible, the EMAS must be designed to decelerate the design aircraft expected to use the runway at exit speeds of 70 knots (approach category C and D aircraft) without imposing loads that exceed the aircraft’s design limits, causing major structural damage to the aircraft or imposing excessive forces on its occupants . . . . Standard design conditions are no reverse thrust and poor braking (0.25 braking friction coefficient).

Turning to the pavement discussion, the stopway and RESA can be pretty light/thin pavements. Mainly because they are only trafficked occasionally, and also because they can be weak such that aircraft sink into them so long as they do not rip off the landing gear. The RESA however is required to take the weight of fire engines, which are in the F50/F100 aircraft weight class, and that means some sort of pavement is usually needed.

For the unpaved or dirt stopway case, the designer (hopefully) has considered the wet and the dry case. It is likely that the stopway has been surfaced with enough gravel to give it some bearing capacity (and to make sure it doesn’t erode with the wind and rain), and typically the gravel has strength even when wet.

For the paved stopway case, the designer will quite properly use a thinner pavement than the runway, although it is usually a designed structure in its own right. I can illustrate this with the concrete design open on my desk at the moment (and the principles also hold true for flexible pavements). The design principle for concrete (rigid) pavements is to control fatigue from bending. The busier the runway, the lower the allowable working stress for the concrete in the slab. Analogous to aircraft structural design and fatigue, I suspect. Stopways can be thinner than runways because they have (hopefully) much less traffic.

For this runway designed for the MD-11 as the critical aircraft (which is more demanding than a B744), the allowable working stress is 348 psi (sorry about the American units) and the slab thickness required for 150,000 coverages (about 36,000 annual departures) over a B subgrade is 400mm (15.75 inches). For the stopway, assuming only a few excursions per year, the allowable working stress is 515 psi and the slab thickness required is 290mm (11.5 inches). So the stopway would be designed as a thinner pavement quite properly and would work well, yet it could easily be said that it doesn't have the same bearing or wearing qualities as the runway.

As old, not bold indicated, there aren't too many airport staff in any country that would know this level of detail. They typically refer such questions to their consulting engineers who usually have some pavements experience; and they in turn might refer the harder questions to a specialist pavement engineer. In Australia, there are less than half a dozen such specialists, and in Africa much the same.

[old,not bold]

Thanks, Overrun, and as always a tremendously informative post.

My beef is that there is a big difference between what should be and what is. Supine regulatory oversight, heavier and heavier aircraft, and the shopping mall model of airport management, may or may not contribute to this.

It I were running an airline these days with anything heavier than a large turboprop, I would not rely on the airport operator or the Regulator to ensure that the stopway is properly constructed. If the airport cannot supply well-documented, recent test results, I would have my own tests carried out and adjust the runway distances if necessary for my aircraft, either to avoid the stopway altogether or to reduce the speed at which it would be entered after a max weight rejected take-off at V1.

PS If an intersection is used, it would only be if the ASDA from the intersection is sufficient for that take-off, (plus all the other factors, eg obstructions, are not a problem with the displaced start of roll). So that's alright, then, except that if the full length is used there's lots of spare pavement to pull up in more calmly, probably without even entering the stopway, as opposed to hitting the stopway at "some" speed with full braking, and hoping like hell the undercarriage won't collapse. As someone said in this thread, runway behind you is useless when you're trying to stop.

PS. I have seen a landrover stuck up to its axles in mud, fairly recently, in the stopway of a Cat 6 runway in the UK, after a period of not unusually heavy rain. I rest my case!

[OverRun]

I can say with clarity that if old, not bold saw "a landrover stuck up to its axles in mud" on a designated stopway, then that stopway does not meet the requirements of ICAO Annex 14, and should not be used in the performance calculations whilst in that condition.

From a pavement engineering viewpoint, the stopway and RESA have to able to support a light vehicle (such as a landrover) without any rutting if they are to be strong enough not to cause major structural damage to an aircraft using them.

While each pavement needs its own testing and design, I can give rule-of-thumb guidance as to the sort of answer that you should get if you go asking questions (or that your airport inspector should be getting if they ask questions). And enough guidance to assess the validity of the answers, especially those given by shopkeepers and accountants.

The sort of numbers I’d be looking for as a minimum are pavement thickness of 150mm reasonable gravel (soaked CBR 45) on top of a sandy B subgrade (B being the ICAO strength rating). In the case of the airport where the landrover got stuck, I assume that the subgrade is a bit clayey and maybe a C rating, and so a pavement thickness of 125mm good gravel (CBR 45) on top of 125mm poor gravel (CBR 25) is the minimum. If it is a D rating, add another 100mm of poor gravel. And because of jet blast, wind and water erosion, the whole thing often needs to be surfaced with a light bituminous surfacing. If it is unsurfaced, add another 100mm of gravel to provide for the erosion. For the technocrats, I calculated this using the MePADs software based on the procedures used in the South African mechanistic – empirical design procedure, with design traffic from a US military spec, cross-checked by British Airways' unplanned research into bogging a B744 at Perth, and my own unplanned research into severe rutting and pavement damage using a 737-800 and a 767-200 of (I forget whom) at (I forget where).

If the stopway is in use, then the airport inspector of your country's regulator should be notified of the deviation from ICAO, and they can request that the airport produce evidence that the stopway complies. Standing above your regulator's head, and encouraging their genuine interest, is the new ICAO Safety Oversight Audit programme. When the audit occurs in your country, the auditors will be looking at pavement bearing strength and should require documented evidence of the regulatory requirements and effective compliance with these requirements. Don't forget to mention this to your regulator . . . . .

[enicalyth]

runway for zhh

I love bureaucracy. We still await an airport on St Helena. Up on Prosperous Bay Plain the available land designated as potentially suitable is constrained by ravines. the trouble with common sense is that it is none too common. The solution it would appear is to insist on 240m RESAs and a terribly short runway, declare the project unworkable and return to the days that if someone fell ill it was time to get out the spade. Very reminiscent too of that old joke about the short runway at Macrihanish. Very short indeed but awfully wide.

[OverRun]

240m RESA is not needed. Only 90m. Them is the rules.

There is a gaggle of doomsayers going around pushing 240m, as if any airport without that length cannot function. I am not one of them. Prudence dictates that 240m be considered where possible, but that is not the governing rule.

In fact, if the place desperately needs an airport, and you can't get 90m RESAs due to impassible constraints, then the airport can be done without RESAs. Legally, in accordance with ICAO, and able to be signed off by the ICAO audit team. I had coffee yesterday with the guy who did it recently.

[old,not bold]

Stopways........I wonder how many airports in UK, the rest of the EU, and world-wide have been asked by its regulator to produce evidence of tests? Perhaps airlines need to be more pro-active here. But I think the point's been made enough in this thread.

RESA's. As Overun says 90m is the minimum. But there's more to it of course. In the UK at least "licensees should not assume that 90m is sufficient" and they are required to review their RESAs once a year, taking into account a large number of factors including a change in the type of aircraft and traffic level, navaids, overrun history and so on. A 240m RESA is "recommended" for Cats 3 and 4, notwithstanding this review.

It seems to me, on consideration, that there is understandable confusion between the roles of the stopway and the RESA. The RESA serves several purposes, including when an aircraft "undershoots" (CAP 168's term, not mine).

But among those purposes, the RESA is there to "minimise risk to aircraft and their occupants when an aircraft overuns .. a runway". Hang about; the stopway is "an area of ground beyond the TORA which is...a suitable area in which an aeroplane can be stopped in the event of an abandoned take-off".

This appears to mean that a RESA is provided, in the over-run case at least, as a secondary stopway if the aircraft cannot be stopped within the ASDA. But why? the whole idea is that the aircraft an be stopped within the ASDA under all circumstances, and its take-off weight is adjusted, if neceesary, to achieve this.

Is this simply a pragmatic recognition that stopping distances achieved in manufacturers' tests are unrealistic in real life?

If we accept that, the problem then is that the RESA surface "should enhance the deceleration of aircraft in the event of an overrun"; it might even contain an arrester bed. Thus if the aircraft leaves the stopway heavy and still rolling with any speed, structural damage is predictable, with consequent higher risks to passengers and crew.

Surely there needs to be some clarity about this? If there is any risk whatsoever that an aircraft cannot be stopped safely and without risk of structural damage within the declared ASDA, that risk must be eliminated. It is a fudge to use the RESA instead of eliminating the risk of not stopping within the ASDA, because using the RESA as additional stopway simply creates more risks.

Surely the correct solution is to make sure that the declared ASDA is absolutely up to the required standard along its whole length (or if not to assess and use the real ASDA), and to review the way in which accelerate-stop performance is calculated so as to provide a 100% probability that an aircraft can be stopped within the ASDA under all circumstances (including an allowance for load sheet inaccuracy, among all the other adverse factors like weather, crew pissed off by security, etc etc).

I do realise that the response is likely to be, "Get real, it'll never happen. You'll have us reducing payloads next. And that's money." But there's no harm in being aware of the issues. And I still get nervous in the back on intersection take-offs, just to help the traffic flow or whatever.

[Stefanovitch]

Enicalyth

Would you please check your private messages.... Thanks..... SHL