skiesful, et al. Sorry, didn’t intend to lecture (# 19); a sharing of thoughts, an observation of how PPRruNe’s think, and a reminder not to rely on a single data source etc, etc.
“Do you know of any airline applying stronger crosswind limits than the manufacturers' suggested limits?” No.
Although I am familiar with several aircraft types, I don’t have access to operator’s manuals. I suspect that the lack of response to the question in this and other threads, suggest that where operators publish a limit, they use the demonstrated value without additional margin. Hence my concern about understanding what demonstrated means and the errors in wind measurement / reporting.
I agree with the ball-park approach. Where to discontinue the approach probably depends on how a limit is determined and stated by the operator.
A hard rule like JAR-OPS RVR minima in low visibility enables a specific position; a more flexible rule like a stabilised approach (more variables) uses several decision points, i.e. 1000ft, 500ft, threshold, and even touchdown. The latter rule might result in more errors as several parameters have to be assessed - more opportunities for judgement. Even so, flexible rules should still have a final hard limit.
The concept of operator crosswind ‘limits’ appear to be even more flexible still (perhaps due to certification terminology, weak operational knowledge, etc), whereas in reality a hard limit is required (accident / incident statistics).
Comparing manual landings with autoland, the latter has a hard limit applied before an approach. Autoland limits include a safety margin due to difficulties in determining the system’s performance in the conditions. This is not so much a technical issue, (the reliability and consistence can be assessed), but more the variability of the conditions and their effect on the aircraft / autoland system.
To me this is similar to the problems experienced by crews during manual landings; they might miscalculate the crosswind component, or misjudge the effect of gusts;- human fallibility.
So just as the certification process adds a margin for autoland, so too should the operational process for manual landings.
I would recommend a hard limit less than the demonstrated value, but greater than autoland (credit for the positive aspects of pilot judgement). The limit would depend on good knowledge and training to provide the essential skills of judging and flying in such conditions (airmanship, currency).
In my limit, if the manufacturer does not specify any demonstration of gusts, then either gusts are limiting or factored, e.g. wind + half the gust. An absolute gust value might not be as important as reminding crews to think about the consequences of the risks from gusts.
More generally a safety margin of 60% of any wind speed error, or a combination with direction error (and calculation) could be used. This probably results in a 5 kt reduction from the demonstrated value (ball-park figure).
When to apply it? Earlier rather than later. Although wind measurement is historic (up to two mins), which might encourage a later assessment, this would result in less time for a well considered judgment and greater pressure to continue the approach (plan continuation bias; ‘press-on-it is’ – the closer you get the more you continue). In difficult conditions, turbulence, gusts, etc, crews should avoid increasing workload at a late stage of the approach where greater attention must be allocated to the flying tasks, thus a late consideration of a limit has greater risk. I suggest that the decision is made at 1000ft; this a consistent point with any RVR limit (if applicable) and a stabilised approach (IMC). An earlier decision also provides opportunity to refocus on the missed approach procedure, thus avoiding the rush of manoeuvring due to last minute decisions.
A lengthy view originating from a short question, but that’s like most safety issues.
Refs:
Crosswind Certification - How does it affect you?
Safety aspects of aircraft operations in crosswind.