Thursdays HeadacheAdded : Monday at 11:55 You know, you can count pretty much on one hand how many times over the last few years the global models have struggled with a single feature. What normally happens, is that one of the global models, whether it's the GFS ► or ECMWF ► will latch onto an idea and then eventually the other will catch up.
Take a look at the pressure chart for today and you can see the low pressure system bringing rain across the British Isles at the moment, but the feature in question is to the Southwest of this :-
![]()
Over the last week or so, the GFS ► and ECMWF ► have both toyed with taking the system across England and Wales, then reverting to sending it across France before changing their mind again and bringing it across Southern England!
This morning? Well, here is the GFS ► output for Thursday :-
![]()
Here is the ECMWF ► output :-
![]()
We're not talking T+240 here... We are talking T+78, just a couple of days away?!
The models are struggling with this system for two reasons. The first is that the circulation was part of a developing tropical circulation last week and thus lacked the upper level steering winds which meant the timing of the movement of the system was that much more difficult. The other was the lack of detail in the mid level circulation, in short, the models couldn't find the "fix" of where the centre of the system and therefore the centre of gravity was.
As a result, the models fluctuated between keeping it a seperate system and then allowing it to engage with another low pressure system to the Northwest on Wednesday.
It's not just the main models which are struggling. The mesoscale models also having other ideas and some of the new AI models we use also offering no clear track of the system.
In theory, the models should come together with a clearer consensus today, but after the last two weeks of uncertainty with this single system even that might be asking a bit too much.
Interestingly, the Super Ensemble ARCUS which is a blend of all the models and ensembles sides more with the GFS ► :-
![]()
But, given the spread between all the output even this has a fair sprinkling of salt on the output.
Rest assured, as soon as we have a clearer idea then we will run with that, but if you think the models are flipping and flopping like mad over the next week or two then it's all down to this one weather system and fortunately events like this are very rare in the modelling world.
Just as well, otherwise we would have no hair left!
METEOROLOGIST : MARSH | 
Update : A dry afternoon with a maximum temperature of about 5 degrees . Expecting the te...
