How I rank high res models:

WRF NMM

ARPEGE/AROME

DWD/ICON

HIRLAM

EURO4

However keep in mind the NMM uses GFS data, so it could well be that the ARPEGE is my favourite now.

Here are the Arpege 81 to 84 precip type charts:

After a week or more of being the model that holds everything back, it's now quite a lot more progressive than ARPEGE (the new favoured model it appears!) and GEM.  Huge uncertainty still....

They just signify heavier snow. The charts are still in development and I'll be integrating with TWO once the next version of the app is live.

So basically all snow.... 

Edit... except for Planet Thanet. Sorry JB!

Model chaos!  Just when things couldnt get more wayward they do.  Here is a chart which is a possibilty:

Arpege 114:

With A Wetter for the GEM around the same time:'

Grrr! ARPEGE is merd! Down with the ARPEGE! 

PS - If the now flavour of the month ARPEGE handles lengthy high pressure systems well, does that make it the new kid on the block?...

Actually, if you look at the high-res Arpege charts...

There's a load of rain first

but then even Thanet gets a decent wedge of snow on the back edge

I always giggle when Meto showing progressive charts illicit ''uncertain'' or ''knifeedge'' or indeed no comment whereas GFS showing progressive charts are normally termed a ''pile of shite'' or ''bin it''

The chart you want to see is always the best one!  

Just 24 hours ago the Chilterns was the place to enjoy a clear, frosty and sparkling long weekend.

Now we could have snow for 3 days running - I'll believe that when I see it as I for one will not be sucked in by any of the charts for Sunday onwards until 12z tomorrow.

A nice winter's day today. Sunny and cold in the wind with p.m. temp hovering between 4 and 5 degrees.

I prefer the GFS 384 chart - nice easterly - but I don't know how to post charts!!

There may be rain on the ARPEGE then, but just look at Sunday!

Hilarious. GFS now looks better than UKMO in the short term!

GFS now everyone's friend again

GFS deep fi brings the cold back again from the east.

Key issues into (the nearer-term) medium range are compounded by mismatches between 00z/06z model solutions and imagery. The latter is already looking more amplified in the Atlantic versus output, favouring the slower solutions. This slower progression was always suspected and remains the main caveat set against ongoing deterministic modelling, which continues to show marked swings including run-to-run within the same suites (and even lately between UKMO-GM Operational and Parallel suites). No detailed analysis of current output is really trustworthy into the start of next week... 

From IF on NW

Another variation on a theme from JMA at 84hours:

But I thought GFS was a pile of sh.....

Is it just me or is the only way to get ARPEGE precip involve splitting up the files into individual time steps. Its so annoying, and I don't envy you trying to integrate them.

In a time of such model chaos, the sense of anticipation with the ECM from me is minimal, as we know that at the moment beyond about 60 hours all output is not to be trusted.  I'm loving this FI starting at 60 hours malarchy.

While we wait here's the NAVGEM 114 which then HP drifts over to allow Atlantic in by 180:

Personally I favor a combination of the ARPEGE and the JMA will be the solution. 

The wintry mix overnight Saturday

I am a little surprised that today's supercomputer's nor human intervention cannot model in the impact of a late Northerly running hurricane