Predictive windshear warning
 

The PWS system issues a warning for a windshear detected 1.5nm ahead. Firstly, that's hardly predictive, you'll agree.
Some of the pictorial examples show the culprit CB even 7 miles or so away. Could the PWS warning even be issued for weather further than 1.5nm?
Is the Go Around Windshear ahead warning a rule that one goes around or a recommendation?
What would you do if, You were in very calm conditions at 300 feet on short finals( no precipitation, calm winds, whole runway in sight) and got the Go Around Windshear ahead wx with a very mature and threatening CB on the Go Around path 6nm ahead of you. Would you land or go around?

My QRH is very straight forward. A "go around, windshear ahead" should be responded with a windshear escape manuever, or at pilot's discretion, a normal go-around. You really want to gain as much altitude as you can and start turning away from the warning area, which will be displayed on your ND, before you get there. You might be descending quite a lot in a severe windshear even with engines firewalled.

The only time I would consider ignoring it, would be in nice weather with light winds and no convective activity whatsoever. PWS has been known to produce some false warnings on airports next to large areas of water. But having a CB on upwind... I'm hitting that TOGA and getting out of there.

The answer is very obvious. You would be touching down in one nm. If you went around from 300ft or 000ft due to I don't know what(because you will accept a bad landing), you would go towards the same CB since you are not in a helicopter. The difference would be 1nm closer. You can always turn to avoid it.

Hi,

regarding the tech background of the topic, I'd leave that to a or to your FCOM/FCTM references.

Then

During approach, visual and aural warnings are downgraded to caution between 370ft and 50ft AGL, anyhow as long as You can satisfy yourself that there are no other signs of possible windshear and the reactive WS system is operational, you can disregard those alerts.

In Your example we are missing a few elements to be able to answer your question, for example the lateral and vertical extension of the CB. In your example if 6NM is the distance from the edge of the cell to your aircraft it means the downburst is spreading out wet particles for a huge distance hence there could be some very threatening weather you should be able to observe visually and by wx radar analysis, SIGMET and so on. Point is, the PWS is a valuable tool that must be used in addition to several other pieces of information and to confirm a preliminary weather analysis done by the flight crew, it is there to confirm what You already suspect. Another important point: never start an approach without knowing how you will fly the missed approach, i.e. when and where you will turn with respect to weather and hence coordinate early with ATC.
Hope that helps.

Experience cannot be taught; it can only be gained
 

The question which should be asked is ‘what would you (self) do’, how did I get into this scenario, what were the briefed (pre-considered) alternatives.
Why wait for a warning.
Forum views would be opinion, some based on hard learnt experience, but framed by a (your) hypothetical scenario.
A more meaningful question is why was the approach commenced, continued, to the point of the situation described; what options were considered to ‘undo’ the decision to start the approach.
After that you can seek alternative views from the forum, with reasoned discussion, opportunities to learn - flexible response, not a rigid ‘land or go around’, all of the situations in between.
Also the real world doesn't stop with the decision to go around, as might simulator training.

Experience cannot be taught; it can only be gained. The crew in the incident below did not have the technology of a PWS, but they did have a ‘system’ for the conditions, and survived to pass on the experience.

https://www.dropbox.com/s/51zkz0lznd...unter.pdf?dl=0

There are few if any ‘answers’ in aviation, but there are many situations with opportunity to learn,
“Wind shear, like many other severe aviation weather hazards, is best avoided. Note the warning signs - clouds, temperature, and rain - if in any doubt during an approach, commence a go around immediately; if on the ground do not take off.”

Were any wind shear precautions taken?

Lot of tools in the kit, but one has to open the kit to use them.

https://reports.aviation-safety.net/...321_N564UW.pdf

I’ll do what I’ve done when it has happened to me twice in different airports, land.

At least in the Boeing there is a bulletin that explains under what conditions you may have a false alert, and it warns you to take into consideration other factors such as pilot reports, surrounding weather, and trends in wind and aircraft speed.

There is a list of airports in that bulletin where is proof that may happen but is not limited to those only so you can apply that to your scenario.

Consideration should be given to the surroundings before pressing the toga as you may put yourself in a worst situation, my suggestion is to be go around minded, but allow a few seconds to make the most sensible decision.

At least remember whether AP is ON or OFF and somebody fly the plane. In A350 they could get away but in a nonprotected B737 800 at Douala or Tatarstan airline at Kazan they didn't.

Not in my Boeing… could you elaborate? Or copy/paste?

Airspeed and upwardness
 

Stab, et al, revise the mind set.

At the instance of alerting there are no ‘false’ warnings - you don't know.

A amber alert, prepare, adjust, act; a red warning - react TOGA, ‘airspeed and upwardness’.

‘False’ is a label after the event. Describing a system or location as ‘false’ before a situation is encountered is an indication of a system weakness. The system component of responsibility for a safe outcome is moved from ‘the system’ (manufacture) to the pilot. Pilots are expected to mitigate a deficient system, a system which was originally intended to aid the pilot. (review the aircraft manufacture or vendors philosophy, cf MAX)

Beware what is believed, how situations are framed, and by whom.

'it's the hope that kills you' ...

Not only Boeing but also Airbus.

From Airbus FCTM:

Note:
​​​​​​When a predictive windshear@ aural alert (“WINDSHEAR AHEAD" or "GO AROUND WINDSHEAR AHEAD") is triggered, the flight crew must carefully check that there is no hazard. If this is the case, the flight crew can disregard the alert, as long as both the following apply:

• There are no other signs of possible windshear conditions
  
• The reactive windshear system is operational.
  

Known cases of spurious predictive windshear@ alerts were reported at some airports either during takeoff or landing, due to the specific obstacle environment.

However, the flight crew must always rely on all reactive windshear (i.e. WINDSHEAR) alerts.

I cannot find it in google so I copy/paste from the manuals in my computer.

Note this bulletin is version 3, and is applicable to a specific version of the weather radar equipped by different Boeing aircrafts. It’s not the Multiscan.

It happened to me in different operators with different Boeing models in airports not listed in this bulletin. Previous experience is something to consider always.



Number: CEX-19 R3
Issue Date: May 17, 2021
Subject: Predictive Windshear System Anomaly
Reason: This bulletin informs flight crews of the susceptibility of certain airports to false Predictive Windshear System (PWS) alerts.
This bulletin is being revised to update the list of affected airport/runway combinations. Information in this bulletin is recommended by The Boeing Company, but may not be FAA approved at the time of writing. In the event of conflict with the FAA approved Airplane Flight Manual (AFM), the AFM shall supersede. The Boeing Company regards the information or procedures described herein as having a direct or indirect bearing on the safe operation of this model airplane. THE FOLLOWING PROCEDURE AND/OR INFORMATION IS EFFECTIVE UPON RECEIPT
Background InformationAirlines have reported false Predictive Windshear System (PWS) alerts at a number of airports. The anomaly is only applicable to PWS alerts; all reactive windshear alerts which occur are valid. These false alerts are limited to airplanes equipped with the Honeywell weather radar with the following PWS weather radar processor part numbers:

• 066-50008-0406 (All listed airports)

• 066-50008-0408 (All listed airports)

• 930-1000-001 (SBRJ airport only)

• 930-1000-002 (SBRJ airport only)

• 930-1000-003 (SBRJ airport only)

Honeywell has reviewed data provided by the affected airlines and has attempted to determine if particular airports and runways may be susceptible to “false alerts”. In addition, data have been analyzed to determine if the alerts are more likely during takeoff or on approach.

Honeywell has accumulated sufficient data to suggest that the following airport/runway combinations are susceptible to false PWS alerts:

• EHAM (Amsterdam), Runway 9, Takeoff

• GCRR (Lanzerote), Runway 3, Approach

• KBOS (Boston), Runway 27, Approach

• KOAK (Oakland), Runway 29, Approach

• KPHL (Philadelphia), Runway 35, Approach

• KSNA (John Wayne Orange County), Runway 19R, Approach

• KSTL (St. Louis), Runway 12 (L and R), Approach

• LEBL (Barcelona), Runway 25, Approach

• LFMN (Nice), Runway 4L, Approach

• LGSR (Santorini), Runway 34R, Approach

• RJTT (Tokyo, Haneda), Runways 16 and 22, Approach

• RKPK (Busan, Gimhae), Runways 36L and 36R, Takeoff

• RKSI (Seoul, Incheon), Runway 33R, Approach

• SBRJ (Rio de Janeiro), Runway 20L, Approach

Although these particular airports appear to be more susceptible to false alerts, the data indicates the majority of operations at these airports do not experience false PWS alerts.

Flight crews should use the following criteria to help determine if windshear exists:

• reports of windshear from other aircraft

• visual indications

• tower windshear alerts

• differences between computed winds in the airplane and reported winds from the tower.

Honeywell has developed software modifications that should significantly reduce the occurrences of false PWS alerts. These modifications also include numerous other changes and enhancements. Service Letters 737-SL-34-188 and 737-SL-34-189 provide additional information about these changes.

As Honeywell continues to develop a software solution and to process data, operators are encouraged to continue reporting incidents to Honeywell and Boeing in order to provide the most effective solution possible to this anomaly.

Operating InstructionsIf windshear is encountered, perform the Windshear Escape Maneuver.

It is recommended operators establish policies for flight crews operating into one of the reported airport and runway combinations in the event a PWS alert occurs. The following windshear criteria may be beneficial in establishing policies:

• reports of windshear from other aircraft

• visual indications

• tower windshear alerts

• differences between computed winds in the airplane and reported winds from the tower.

Administrative Information
This bulletin replaces bulletin CEX-19 R2, dated October 22, 2018. Revise the Bulletin Record Page to show bulletin CEX-19 R2 as “CANCELLED” (CANC).Insert this bulletin behind the Bulletin Record page in Volume 1 of your Flight Crew Operations Manual (FCOM). Amend the FCOM Bulletin Record page to show bulletin CEX-19 R3 "In Effect" (IE).

Please send all correspondence regarding Flight Crew Operations Manual Bulletin status, to the 737 Manager, Flight Technical Data, through the Service Requests Application (SR App) on the MyBoeingFleet home page.

I’m sorry but I expect the pilots to have divergent thinking and not to behave like a computer.

As per the bulletin yes, there are false warnings. I don’t say that, Boeing and Honeywell (the aircraft and radar manufacturers) say so.

In the situation described by the colleague, if the PWS is telling you that you have to go around and is a system error that will actually put you into a real wind shear, I would like to think you will take at least a few seconds to assess and decide before pressing a button that has unpredictable outcome.

I think that you should apply what you said to your self, as you seem to have too much confidence in the system “technical and non-technical” as a whole.

Thank you Stab, our aircraft have multiscan so I guess the bulletin is not applicable.

Stab, thank you for the document extract, which on reading it challenges core aspects of safety thinking and human factors.
The expectation of pilots’ abilities in situation assessment, knowledge recall, judgement and decision making is not supported by evidence from incidents and accidents.
Arguing the point to absurdity, then remove PWS because its not required (but reactive is) - all pilots will manage, all situations, all of the time - not so.

Human performance is limited, also influenced by inappropriate training or publications.
Not only have the vendor / manufacturer introduced doubt about the effectiveness of a safety system, so too regulators who apparently condone the text.

‘PWS doesn't tell you’ anything, that’s your interpretation, your mental model, part of situation assessment, part of the judgement process in decision making. Citing the specific hypothetical situation #1, is irrelevant because as previously argued that situation should (ideally) not be encountered - situation awareness again, thinking ahead, considering possible outcomes.

‘A few seconds’; in the Air Zim incident #5, the time between indications being detectable and understood, followed by very swift and correct action, was 20 secs. What is ‘a few’ relative to 20; how fast might we think, or think we can think with the surprise of an unexpected alert (if the alert was not unexpected why are we there).

After Air Zim the industry debated if PWS could have detected this form of downburst any quicker - possibly a simultaneous alert and reactive warning after ‘a few seconds’. The limitation considered a rapidly forming cell and downburst, above the aircraft - descending flight path, and at close range (high scan angle).
Encountering the same situation at one of the susceptible airports doesn't provide time for evaluation; the document provides ‘get out’ statements, requiring pilots’ awareness, etc. Poor safety management requiring pilots to manage a weak system, yet where pilots are subsequently judged wrong, they are blamed because ‘we told you so’.

A useful safety technique is to have a ‘pre-mortem’ assessment; like a post-mortem but before the event - ‘what if’, identify the double-bind, catch 22.

PWS Warning
 

Why is a warning issued within 1.5 NM called a predictive warning. That's 20 seconds lead time in a airplane at approach speed. Hardly predictive imho.

It's still predictive, as you aren't in windshear conditions just yet.

Why are you flying approaches at 270 kts? :E

It may commence as an ADVISORY, become a CAUTION and then become a WARNING. It unlikely to be an instant WARNING. But could be depending on the aircraft altitude and whether climbing or descending. The alerts are only available for a narrow altitude range.

PWS; some history.
 

Within the span of an aviation career, there was:-

1. Windshear. No alert, no warning, no guidance. If you encountered it - over to you - fly the aircraft.
but would you detect and assess the conditon in time, and act swiftly, was the aircraft capable of escape from it.

Thence there was education, think ahead assess conditions, avoidance, recovery.

https://www.faa.gov/documentLibrary/...ar/AC00-54.pdf

2. Technology for board detection and audio alerting; a reactive system. Thence advice for escape manoeuvre, and then guided recovery - use of fast-slow indicator, and then FD guidance developed. Simulation.

Ground based technology for early warning of likely conditions; a semblance of prediction.

3. On board predictive technologies (PWS); look ahead, advisory alerting, crew options, warning - requiring immediate action. Thence advanced severity classification and recovery guidance; some auto flight, auto thrust capability.

There are those who have experienced windshear, and those who will, but who should avoid it at all cost.

https://www.dropbox.com/s/51zkz0lznd...unter.pdf?dl=0

The only PWS warnings i got so far were without any CB close by, but warnings about possible windshear on final on the ATIS, in both cases in extremely windy weather and close to some terrain. However, if i do get a „Go Around, Windshear Ahead“ i will continue to follow it. Not doing so will probably get me on No Fly status if not fired.

[QUOTE=FlyingStone;11235651]It's still predictive, as you aren't in windshear conditions just yet.

Why are you flying approaches at 270 kts?

What if within 1.5 miles or 5 miles is just clear air and there is vicious weather in the go around path.
My point is that the system design has a flaw and 1.5 miles does not qualify to be called a prediction imho.

However, if i do get a „Go Around, Windshear Ahead“ i will continue to follow it. Not doing so will probably get me on No Fly status if not fired.[/QUOTE]

That's my point, job pressure can cause an accident. In my experience the PWS system is flawed. i would follow a reactive warning if I were in or close a CB but nothing else. There are flaws in the PWS system.

That's how it should be, i agree but it's not.
It's a warning and is mostly false. Pilots though react to it just to avoid losing their jobs.

If there’s vicious weather on your standard published go-around path then either you advise ATC that, in the event of a go-around, you require an alternative ATC approved track (of your preference), or runway (if available) or you don’t carry out the approach in the first place and divert. For example, Milan 36 go-around paths are frequently entirely coloured in red on the Wx radar during the Cb season.

Its predicting in a short space of time you will encounter windshear. Therefore it is “predictive”.

The amount of time given is plenty to avoid it.

Would you ignore a TCAS RA if you couldn't see the intruder? (who might be above you, below you or behind you. Or in IMC).

Can you see a microburst ? Can you see wind-shear?

The doppler RADAR can detect fast moving wet air, i.e. wind-shear or a microburst, so the PWS issues a warning or an action. Why would you ignore that?

What have you got to lose by following a PWS - even if it turns out later to have been false? (as said above, if there is a CB in your go-around path, you should have already told ATC that you will be delaying your approach, or at the very least, turning to avoid it.

Better to follow PWS than assume it is a false warning.

From Airbus FCTM:


And

Totally agree with everything else. But, just to clarify, we are trained to always follow RAs, even if you have the other traffic in sight, and I think most companies do.

That was my point :ok:

Yes. When there’s traffic the TCAS can’t see. It’s happened before.

When TCAS can't see the traffic it can't trigger RA. You see it and avoid. The question is when TCAS RA is telling you to descend will you ignore it climb instead? Then how do you know what will the guy do? Not following RA is dangerous.

Delta191 Lockheed 1011 fatally crashed at Dallas FW while going through a single CB on ILs path. That triggered the research to identify the vertical movement inside a CB. It triggers the warning when it crosses a certain threshold.

Oh dear, I truly hope you’re not serious.

If it can’t see the other traffic, it will not issue an RA.

It RELIES on all aircraft following the coordinated advice it’s giving.

Follow the RA and then sort yourself out.

Quite so, added Dr Tetsuya "Ted" Fujita's name to our lexicon. (he passed away in Nov 98... )

The geometry of DL191's encounter is a major badness to the day at Dallas.
The alert inhibiting is quite smart in ir's own way, it should avoid the sort of question that has arisen in this thread... further out from the threshold increases the potential for getting a downflow to increasing tailwind event in a microburst. strategically, taking a hold is not a bad idea, but stuff happens to best-laid plans. Making a decision to continue while having a PWS blaring at the warning level would look pretty interesting at the subsequent court of inquiry. Assuming that you have a QAR type program you are guaranteed to be getting tea and biscuits from management who may not ascribe to your point of view. Reminds me of a discussion around the bar once when a newly minted driver was telling all 'n sundry around the table which of the companies policies he declined to follow by choice, which led to the laconic and polite comment by the Director of Standards, "then, pray tell, what other company policies do you wilfully disregard?". The standard that would apply to a deviation from a trained response would be unless a greater emergency exists. To that end, describing the morning after that the landing with the raucous noise until inhibiting occurred was the lesser emergency would probably get the grey-haired guys at the table busy with their eyebrows... muppet style!

The conversation would be even more interesting if they are digging the gear out of the mud off the side of the runway...

Old Guys

https://cimg2.ibsrv.net/gimg/pprune....2afca797c6.png

runway offs come in various flavors and varieties... the one below had the crew asking for a tow truck... needed a bunch of bulldozers and angle grinders.

https://cimg3.ibsrv.net/gimg/pprune....446112287.jpeg

From Airbus FCTM:
I tried to upload this explanation of a microburst earlier, but the drawing didn't upload. Here is a screen shot:

I have seen aircraft flying under huge CBs as they approach a runway, and hoped for their sake that a microburst did not suddenly descend on them.........

https://cimg5.ibsrv.net/gimg/pprune....58349c63b4.png

The main point of having a detection and alerting system is to improve the level of safety above that based on unaided human performance (thus EGPWS, ACAS, Windshear).

Pilots who believe that they know better - ‘its false warning’, are deluding themselves and depriving the industry of its improved safety record for these type of accidents.
‘False’ is like ‘error’ it can only be determined after the event, which is based on outcome. All alerts and warnings should be considered real and acted on.
The safety statistics indicate that modern technology has far superior performance - error rate / false warning, than the human for those specific tasks.

The analysis of an encounter - Captains in Command, #19, compares that event with the FAAs training model #4 (fig 2, and 3), which was derived from the DFW accident. Although event severity is normally associated with big storms, this is not always correct.

Our Company says that a Windshear Ahead can be ignored if ‘no signs of windshear are present’ which basically means it is all your fault if it goes wrong.

But can you see wind-shear? Can you see a microburst?

No, but your weather radar and PWS can measure it and warn you.

Not sure about the latest FCOM but before when PWS warning was still a memory item it was written that if the crews make a positive verification that no hazards exist then the alert may be disregarded. Alerts can be spurious. It's all about common sense. If it's clear weather outside, and you have a PWS alert then obviously you can assess the situation and ignore it. Now if there are CBs activities on final then it's a no brainer IMHO, follow the PWS warning. We are pilots after all we should not follow blindly what a computer is telling us to do. I have seen a guy asking me to request weather deviation cause the weather radar was showing a predictive overflight Icon, new feature on the latest weather radar indicating the risk of a fast growing cells ahead of the aircraft. It was during day time and there was absolutely nothing!! I was like: Are you trolling me right now? Pilots don't even look outside anymore...

In the situation I’m familiar with, there were 2 intruders, and TCAS issued the RA without knowing about the 2nd aircraft. The crew could not comply with the RA.

An outlier, yes, but technology doesn’t always have the full picture.

See my reply to Vilas above.