B737-800 aquaplaning at Christchurch New Zealand. Excellent report

[A37575]

https://www.atsb.gov.au/publications/investigation_reports/2015/aair/ao-2015-046/

Well worth reading. The crew did everything right but were caught by surprise by aquaplaning conditions during the latter part of the landing run.

[CurtainTwitcher]

The report hints at the reason, FAA SAFO 15009 (https://www.faa.gov/other_visit/aviation_industry/airline_operators/airline_safety/safo/all_safos/media/2006/safo06012.pdf) explains: Purpose: This SAFO warns airplane operators and pilots that the advisory data for wet runway landings may not provide a safe stopping margin under all conditions.

In a nutshell, the FAA SAFO are saying wet runway landing distance including factoring isn't sufficient to cover you behind. A training Captain once mused to me about a hypothetical exchange between a prosecutor and a Captain in the subsequent court of inquiry:
"So Captain, was the runway wet or dry"
"the runway was damp, therefore the manual says it could be therefore considered dry"
"Captain, If you sat on the runway, would your bottom be wet? No further questions your Honor"

[Capn Bloggs]

CT, you've linked to the wrong SAFO. In any case, I don't think the SAFO has anything to do with this.

The problem was, due to company policy, they used Dry figures on a Damp runway. The "Damp" runway chestnut rears it's ugly head again.

They should have used the "Good braking action" figures. ;)

[AerocatS2A]

The problem was also that the "damp" runway was largely wet.

I disagree that the 15% additional factor is not enough, it seems it is spot on. ;)

[B2N2]

The crew did everything right

Isn’t this where “pilot stuff” comes in?
Company changes the book.
Runway is “damp” and dry numbers say it’s tight but ok and wet numbers say no go.
You get paid to say no.

[PEI_3721]

The relevant SAFO is 15009, but linked here.
http://www.faa.gov/other_visit/aviation_industry/airline_operators/airline_safety/safo/all_safos/media/2015/SAFO15009.pdf

The FAA AC for pre landing runway assessment is here.
http://www.faa.gov/documentLibrary/media/Advisory_Circular/AC_25-32_Final.pdf

Aero, #4, re 15%, it depends which 15% is being considered.

If based on certificated Landing data; 1.15 dry distances for wet, then ‘this 15%’ may be insufficient in many situations. The FAA SAFO appears to relate to this, but caution because their ‘manufacturers actual’ distances (Boeing term) is not the same as AFM certificated distances.

Post TALPA and using Operational Landing Distances (OLD) which provide a more realistic view of achievable landing distance (actual); then manufactures and regulators (FAA AC) recommend a further 15% minimum addition, often published a Factored OLD (FOLD). This 15% should relates more closely to what happens in day to day operations on wet runways. However ‘how wet is wet’, there are some, many combinations of water depth, runway surface, tyre/brake condition, etc which warrant use of more than 15% minimum addition.

Some of the technical aspects are here
https://skybrary.aero/bookshelf/content/bookDetails.php?bookId=3841

[Intruder]

To me, "damp" = wet, since it is not as good as dry. How much less good? I don't know, so that's why I'll err to the safer side.

[Jwscud]

I have aquaplaned in a 737 NG and it is not pleasant. The crew were put in a difficult position with the workload pretty high.

I found the report frustrating in that it didn’t bother to mention the Runway length until quite late in the picture. Once I eventually found it was only 1700m, I was very surprised that they chose Autobrake 3 rather than Max as the NG is not a good stopper.

I also couldn't see see whether they had Carbon or Steel brakes?

[73qanda]

To me, "damp" = wet, since it is not as good as dry. How much less good? I don't know, so that's why I'll err to the safer side. Sounds like a sensible approach to runway performance. I like that you’re not burrowing down into various documents to justify your approach.
A good working knowledge of the terminology, phraseology, factoring techniques, data sources and SOP’s was much more difficult to maintain in 2015 when this occurred, than it has been historically, because of
A/ ambiguity between source documents and B/ large scale change of methodology
We are probably starting to get to grips with TALPA a bit more now but still, a conservative, sensible, and simple approach from individual pilots as displayed by Intruder above is our best defence.
The reader of this report will finish it unaware that many New Zealand line pilots were concerned about runway performance in NZCH during the runway works at this time and it appears that Air NZ had elected not to use 29 prior to this incident.
Good on the crew for stopping it 5m before and not 5m after.

[tonytales]

Way back when, before introduction of anti-skid protection, aquaplaning (I seem to remember it was more commonly called hydroplaning at least here in US) was not uncommon. In severe cases the tire rubber would melt and it was explained that the braked wheel raon atop superheated steam which literally melted the tire rubber. Antiskid systems and Maxerets mostly stopped that extreme result except when there were maintenance faults such as crossed brake lines (or wiring) on a bogie type gear.
Runway braking reports are problematical. Back in late 1970's when Maintenance Manager for EAL at KEWR, I would get weekend duty. On snowy days the Port Authority would come and pick me up at my office and take me out for a trip down the active runway or runways. I was supposed to rate the braking action from the non-driver seat of a speeding Police car. Being a conservative fellow my fair/poor braking condition ratings sometimes made the Airport Operator unhappy. I hope they have come up with better ways of rating than in those now pre-historic times.

[stilton]

While this may have been a close
call how can it be classified as an
’incident’ ?


After all they did not overrun the runway

[73qanda]

You’re right Stilton, it didn’t overrun therefore it is not an accident.
It can be classified as an incident because it meets the ICAO definition of a ‘serious incident’.
An incident is defined as:

An occurrence, other than an accident (https://www.skybrary.aero/index.php/Accident), associated with the operation of an aircraft which affects or could affect the safety of operation.

Note.— The types of incidents which are of main interest to the International Civil Aviation Organization for accident prevention studies are listed in Attachment C to Annex 13.

A serious incident is defined as:

An incident involving circumstances indicating that there was a high probability of an accident (https://www.skybrary.aero/index.php/Accident) and associated with the operation of an aircraft which, in the case of a manned aircraft, takes place between the time any person boards the aircraft with the intention of flight until such time as all such persons have disembarked, or in the case of an unmanned aircraft, takes place between the time the aircraft is ready to move with the purpose of flight until such time as it comes to rest at the end of the flight and the primary propulsion system is shut down.

Note 1.— The difference between an accident and a serious incident lies only in the result.


Note 1 is a pretty succinct way of answering your question.

[Tee Emm]

The captain assisted the FO with manual braking. Both crew reported that they could feel the
rudder/brake pedals ‘pulsing’, which indicated the antiskid system was operating. The crew
elected to keep reverse thrust deployed to assist with braking.
The FO reported that he was focusing on the red runway end lights and noticed the aircraft drift
slightly right, which he then corrected to bring the aircraft back onto the centreline.

To some this may sound like nit-picking but there may be an interesting hidden lesson here. When I read "the captain assisted the FO with manual braking" and later "the FO noticed the aircraft drift slightly right which he then corrected to bring the aircraft back on the centreline," it made me wonder exactly who was steering the aircraft while at the same time both were applying the brakes? It is rather tricky having two people applying full brakes with rudder pedals not centralised because one of them is simultaneously also using rudder to rectify for a lateral deviation. Were they both on the brakes correcting for lateral deviation? Was there any operational need for the captain to "assist the FO" with manual braking?

We all know it can be quite distracting to the PF to feel the other pilot riding the rudder pedals. A case perhaps of the FO thinking who is flying this machine?

If the captain was that concerned the FO was not applying enough braking (otherwise he would not have felt the need "assist" him in manual braking), it would have been better for the captain to call "I have control" and finish the job himself. You can't have two Bob's each way in a situation like this? The ATSB reporter who wrote the final draft seems to have missed that fact that both pilots applying brakes simultaneously but not evenly, could have confused the anti-skid system and perhaps inadvertently extended the landing run?

[mustangsally]

Way back in the late 70's operated a Sabre Liner, no anti skid brake and very narrow wheels, right off an F-86. The check out included a simulated wet runway. The idea was to basically plant aircraft on as close to the end of the runway with a rather firm plant, then cut of one motor (no reverse on these old ladies) and apply brakes. If a skid developed, release the brakes and reapply. As I learned after more than several hundred hours, a plant on a rain covered runway was actually much closer to soft and it worked. If you spent enough time on this aircraft you would skin a tire. I moved on before that happened to me. So, when landing on a wet runway, a plant it good, the payers in the back will think it was a greaser. End result of the plant is less moisture between the rubber and the concrete. The same idea apply for landing on icy or snow packed covered runways. Easier to say sorry for the firm touch down than start a discussion with the Chief Pilot or Safety Board.

[AerocatS2A]

PEI, my comment about the 15% being just enough was with tongue in cheek.

Mustangsally, the 737 didn't encounter wet conditions until partway down the runway. Touchdown technique would not have made any difference.

Tee Emm, I had the same thought. Easy to say with hindsight of course, but I'm sure the crew have had a good think about a few of the aspects of the incident.

[George Glass]

Something is missing in all this. DONT land on Rwy 29 in marginal conditions when Rwy 02/20 is available. ATC in NZ an OZ have a very bad habit of allocating runways that are inappropriate for traffic/noise abatement/ATC convenience.Hardly a day goes by in YSSY that heavy longhaul carriers or allocated Rwy 25/07 due noise abatement when the wind is straight down Rwy 16 and wont get a runway change unless the say "require" instead of "request". Same in YMML regarding Rwy 27 vs Rwy16. When a Chinese low cost operator goes off the end of the runway with a 15 knot crosswind and damp runway the subsequent inquiry will be a ripper.

[AerocatS2A]

02/20 wasn't available*. Previous A320 diverted rather than landing on 29.

* Or rather, it was only available in one direction and the tailwind and shortened length made it worse than 29.

[exfocx]

Why would any operator let alone the regulator accept actual ldg dis + 15 % for inflight reassessment unless a non-normal presented itself, and there wasn't an alternate available.

Edit: Dumbfounded by the dry / damp / wet bit as well. There is no such thing as damp other than here, NZ and maybe PNG, plus from my experience ATC is somewhat lax in down or up grading rwy conditions.

[AerocatS2A]

It’s not actual + 15%, it is actual x 1.67 for dry and actual x 1.92 for wet. 1.92 is 1.67 x 1.15 which is where the 15% figure is coming from.

[Capn Bloggs]

It’s not actual + 15%
Yes it is, Scat. Have a read of 20.7.1b section 11.2. Subtly different to the "takeoff weight" determination of 11.1.

[AerocatS2A]

11.2 is for contaminated runways and doesn’t mention a percentage. 11.1 is as I said above, 1.67 for dry and 1.92 for wet. The 1.92 is 15% more than the factored dry figure, not the actual dry figure.

Edit: could be my copy is out of date of course...

[AerocatS2A]

Sorry Bloggs and exfocx, I was working of an old version of the CAO. You are correct.

[Goldenrivett]

After TriStar G-BBAI overran the wet runway at Leeds in 1985 - we all decided the Runway must be short when wet.
https://assets.publishing.service.gov.uk/media/5422fd9ae5274a13140008e5/2-1987_G-BBAI.pdf

[safetypee]

From the very comprehensive and well considered report I could not identify the basis of the published landing data.

Was the AAM based an ‘in house’ computation or did it use Boeing data - performance manual?

Was the QRH based on Boeing ‘actual’ distances, or the more recent OLD method for calculating landing distance?

Would the performance differ between documents, in addition to the likely difference between ‘actual’ and OLD?

Do current documents still show numerical landing weight, where it is difficult to add 15% distance margin, or do they now show actual landing distance, even with the min 15% already added?

General question: re “…15 per cent may be inadequate under certain wet runway conditions and subsequently encouraged operators to apply additional conservatism.”
The FAA SAFO appears to precede the incorporation of TALPA recommendations and the publication of OLD/FOLD by manufacturers. Would the ‘inadequate’ comment still apply to FOLD calculations providing that the correct surface description is used?

Notwithstanding the above, it does not degrade the value of the SAFO remark - need for ‘at least’ or ‘minimum’ of 15% safety factor, particularly as inaccuracies can arise from ground assessment, reporting, or crew interpretation of surface condition or reported braking action.

[john_tullamarine]

Have a read of 20.7.1b section 11.2

20.7.1b is an operational document and has had a long history of the occasional discontinuity with the certification side of things. The AFMs know naught of the 20.7 series. 20.7 might be fine .. but it doesn't give you the data, necessarily, to comply with its requirements.

[Capn Bloggs]

JT, I as a lowly driver see it like this: 20.7.1b details the current regulatory requirements, including the "15% when airborne" rule. One can then use the AFM data (1.67) to work out if one can comply with the pretakeoff requirement (or one can use the AAM charts). Then, one could use the QRH data (1.15), when one is airborne, to work out if one can comply! :confused:

Was the AAM based an ‘in house’ computation or did it use Boeing data - performance manual?
I suspect Boeing. Our charts (different model Boeing) are laid out exactly the same.

Do current documents still show numerical landing weight, where it is difficult to add 15% distance margin, or do they now show actual landing distance, even with the min 15% already added?
I assume you are referring to the AAM? Each chart is runway-specific; it is better for the crew to be shown a weight (which, I believe, complies with 20.7.1b). Interestingly, the AAM charts don't show the effects of various braking actions (the QRH does), only Dry and Wet.

[parabellum]

Does the old formula of 9 X the Square Root of the tyre pressure still get used for a rule of thumb? Given a tyre pressure of 180psi the aquaplaning speed would be approx. 120kts, 210psi would be approx. 130kts i.e. speeds to avoid where possible and unfortunately close to Vref at times.

[Goldenrivett]

Hi parabellum,
Your formula is correct for dynamic aquaplaning when water can't be dispersed fast enough by the tyre tread - but the incident at Christ Church was due to Viscous Hydroplaning (mentioned in the final report).
There is some good info at https://www.pprune.org/tech-log/170865-dynamic-vs-viscous-hydroplaning.html
"Viscous Hydroplaning
This one can fool you, because it can happen even at very low speed. Essentially, it is sliding on some liquid other than water or in a situation where water has mixed with something. For example, an area of a ramp or runway could become contaminated with a number of substances.

The run-up area would be a good candidate here, because a lot of aircraft sit over the same spot for a few minutes. Although not leaking much oil individually, the collective effect of a drop or two from dozens of aircraft can produce a noticeable stain. Then along comes a little rain to lift this out of the pavement surface, and it gets slick in a real hurry.

Touchdown zones are even worse, because they get oil shaken loose by the landing impact and a lot of rubber dust, too. At some really busy airports, they have to periodically go out and grind this stuff back down to the pavement.

Once again, just add water for a nice, slippery mess! In most cases, a good hard rain will wash away most of this, so most occurrences are as the rain starts or after a light shower or heavy dew."

You've probably notice this effect when you turn onto a damp runway in the touch down zone and get that nose wheel skid sensation / noise.

[PEI_3721]

The old rule may not be adequate. ‘Hydroplaning of aircraft tires‘. https://core.ac.uk/download/pdf/53034583.pdf See Figs 1 & 2.
“It is concluded from this analysis that modern tires have lower hydroplaning speeds than previously assumed.”

Also re the accident, whereas a wet runway / hydroplaning might have led to the poor deceleration during the latter part of the landing, it is also possible that a change in runway surface would give similar indications. Note the the report’s discussion of minimum runway surface friction and ‘slippery when wet’ (page 15) and the aerial photo of the runway (fig 9, page 16).

Also; https://www.skybrary.aero/index.php/Aquaplaning

[73qanda]

Didn’t Hobart have a Virgin 737 put its nose in the grass after treating the runway with the same chemical as Christchurch did?
Maybe six kor seven years ago?

[parabellum]

Thanks Goldenrivett and PEI3721 - Appreciated.:)

[safetypee]

Capn, #26 Thanks.
The QRH is of greater interest. Presumably the table ‘ref distance’ is used once all of the relevant corrections have been applied; then calculating and adding the safety margin >15% to give the expected landing distance.

The chart illustrated on page 25 looks like the basic Boeing format for ‘actual’ distances, which is open to debate whether this is realistically achievable, even before considering misjudged runway braking conditions. As I recall this was part of the background to the FAA’s recommendation for >15% additive (SAFO) because the assumptions in calculating the baseline distance were ‘best piloting performance’ in every circumstance.

The remaining question is what was the basis of the QRH distance used in the accident; has it changed (to OLD), and if so what additional distance is given by the assumptions in the OLD calculation; and are crews aware of any change?

[73qanda]

From the ATSB website;

ATSB Executive Director, Transport Safety, Mr Nat Nagy said the incident demonstrates how the subjective nature of runway surface reports and braking actions reports can contribute to the risk of a runway excursion.

“There is little standardisation between how pilots, industry and regulators describe runway surface conditions,” Mr Nagy said. “Information provided to pilots is often from multiple sources, with undefined terminology that may only be relevant to a certain time or type of aircraft.

“To overcome this, the International Civil Aviation Organization will be adopting the United States Federal Aviation Administration’s (FAA) runway condition assessment matrix in November 2020.”

The Civil Aviation Safety Authority (CASA) have advised the ATSB they anticipate aligning their runway condition definitions with ICAO post 2020.

“Until then, pilots are advised to apply a conservative approach when relying other pilot reports for runway surface conditions, in particular, when the conditions are considered damp,” Mr Nagy said.

Additionally the ATSB found, along with the FAA, that the 15 per cent in-flight safety margin applied to actual landing distances may be inadequate under certain runway conditions. CASA has advised they intend to evaluate this finding and update the corresponding advisory material accordingly.

[Capn Bloggs]

The chart illustrated on page 25 looks like the basic Boeing format for ‘actual’ distances
That's a portion of the QRH page: see page 62 of the report. No factors, with 305m air-run distance. You would have to add your own factors; 67% dry before departure (using our CASA rules) or 15% (min) after you get airborne.

[PEI_3721]

parabellum, et al.
More recent and in-depth (!) information, tyres, anti skid, runway texture, here:- (19MB)
https://www.futuresky-safety.eu/wp-content/uploads/2016/01/FSS_P3_NLR_D3.3_v2.0.pdf

With this greater understanding and issues of different tyre construction, variable runway surfaces, etc, I wonder if the current (updated) wet landing distance calculations still provide the same landing distance safety margins previously assumed