On Wednesday, August 30, 2017 at 10:51:01 AM UTC+1, Freddie wrote:
On Wednesday, 30 August 2017 09:29:47 UTC+1, wrote:
Only last night they were predicting heavy rain for kent this morning and
moderate rain for London and the rest of the SE. Actual weather? Nothing of
the sort. A few scattered patches of drizzle with the main rain band way off in
Belgium. How can they get it so wrong only 12 hours in advance? Did the wind
suddenly turn 90 degrees overnight?
Nothing to do with the wind direction (as you present it) and everything to do with the way small variations in shape and speed within the jet interact with the moisture in the lower and middle troposphere. Tiny errors in analysis have a massive impact on the end result - and the impacts change dramatically in a very short period of time. The forecasters know their stuff, it's just they sometimes can't keep up with the changes.

Never forget that the atmosphere is a complex three-dimensional beast that would be almost totally unpredictable without computer models.

Having said all that, I agree that it was a poor forecast. The only way handling of these situations can improve is through better modelling (more horizontal and vertical resolution, and more responsive assimilation of current data) which will enable a better response to when the forecast looks like going adrift.
--------------------------------------------------------------------------
Forecasts go adrift for a variety of reasons.
Better resolution in the horizontal and vertical might help but will not solve the problem.
There needs to be a better understanding of the physics. The interactions are not well understood, especially the feedbacks. Parameterisation has to come into it somewhere. There is a chaotic aspect to it as well.

Len
Wembury
----------------------------------------------------------------------------