Stu_London

My biggest concern is that the 2018 BFTE cold spell was cut short in the south with a similar feature. When first modelled nearly everyone on here discounted it and some even said that the evolution was impossible. Now we have a scenario where the cold spell is being stopped before it has begun with something very similar. It is not without support in the ensembles. Might be a blip this time, but if it does gain any traction its curtains I would guess, for the south at least.

We might have cold air to the east but its not a lot of good if the airmass isn't originating from there. Lots of other considerations also. Should be getting something from the east by next Saturday, so keep the faith

Game on with the 18z for a snow event for our region next Sunday - just 144 hours for it to all go wrong. Kent looks like bearing the brunt of it with 30 hours continuous snow. However as it's a boundary event there is a goldilocks zone - any shift south and we lose the ppn - any shift north and all too familiar rent and sleet scenario becomes likely. Overall however proper cold is trending from Friday/Saturday onwards.

18z excellent run for Kent - which is one of the places that has done worst so far this season with a 30 hour snow event dumping several inches. Being a frontal boundary event it is fraught with danger as any deviation south and the ppn stays in the channel. If it pushes north, then marginality creeps in along with slightly higher upper and surface temps. However, barring one run where it placed the ppn south, the GFS has been pretty keen on this over the half dozen outputs.

Taking the OP and the ensembles together, its a very positive 12z run, tempered only be the large spread relatively early, which suggests the model struggling overall to get a handle on things

Amazing how quick the weather flipped after beast 2 - I recall playing golf in shorts and t shirt a week later and coming across drifts that had accumulated in the bunkers

2018 was a lot more consistent, although it did end up hitting us 2-3 days later than first modelled. 2010 was just counting down the hours with some subtle variations along the way. Other cold spells were much more volatile - Jan 2013 was a bit of a model fiasco which dated back to December 2012 - lots of chopping and changing and massive stand offs between the models right down to +48hrs sometimes. March 2013 - the cold was modelled really well - but a couple of big fails with ppn distribution. Going further back 2006 had a big failed beast at about 72hrs - came back with a watered down version a week or so later that more or less verified.

Indeed - but I don't think the weekend will be resolved until Tuesday and after the weekend all the models diverge massively. Still good potential for further cold shots and it can't be ruled out that it goes cold and stays cold for a reasonable period from the weekend.

Realistic view of model output from the 12zs without the bipolar tendencies of the MOD thread. Immediate - rain and sleet - bit of snow north of region - no appreciable accumulation Midweek - mostly on the mild side of a couple of battleground events. Any surprises that pop up will be shortlived as the mild push expected to boss matters until Thursday Next Weekend - Cold looks likely to push back down from the north from Thursday. Extent and timing completely up in the air at this stage, although most models have us back in a cold set up by Saturday. There will some PPN around, where it comes from depends on which model you believe, but if the cold gets here and soon enough, there is some potential for snow next weekend, although marginality could be a factor, so sleet again in Tunbridge Wells. Beyond - I don't know and anyone who suggests they do, is lying.

If the Navgem is on board then I calling the cold spell now. In all seriousness, although the ECM has reverted to a colder solution, still no easterly and still no intra or cross model agreement. I'm thinking possibly by Tuesday or Wednesday as to whether we get the cold spell or not?

I think most will take a degree of marginality of what the 0z was showing - zero marginality on the run - just rain. Still time for these to change a bit as with the overall pattern - best ECM run for at least 2-3 days

We might as well see what happens by looking out of the window at this rate The discord between the various models is making them rather ineffective as forecasting tools.

Better ICON this afternoon. One of the problems with expectations is that some runs have been outlandishly cold and long lasting (1 in 50 year events) and the more you look at perturbations like this the higher the expectation. Anyone who saw the 18Z GFS para last night will be aware of what I am saying, which was probably the gold standard of output cold ramping. A one in fifty year event will come along one day, but the chances are it will not be today. We have been fortunately enough to have two such events (Dec 10 and Mar 13) in the relative recent past, so we know that they still do happen, which is why we still keep looking.

yes quite a big setback for any cold spell with this mornings models with the GFS taking a step towards the ECM, which to be fair has never bought into any meaningful easterly regime There is still lots model uncertainty, which means nothing is nailed on for certain and we still have a cold outlook on the UKMO model and a decent cold block to our east. I think the GFS para was still going cold as of the 00z, but that model is very much on its own. If their was shift back towards a cold spell with the afternoon output, it certainly would not be the first time that had happened in recent weeks, but the reality is whilst you have one of the UKMO or ECM not on board, you cannot bank anything

it only gets cold in the south after day 8!!

Hey its all still up for grabs - where it ends up landing, no one can tell - probably no model has the correct solution. you are right its been all over the place most of the winter - apart a short period around storm Bella where all the modelling was very good. for the avoidance of doubt i was referring to the last few runs where it has steadfastly refused to buy into the easterly that others were showing in the same timeframe.

The GFS only delivers quite a short snap and then even blows away the surface cold it was keen on yesterday

last three runs it been fairly obdurate with its theme

I've seen before the UKMO take a positive turn only to find itself the odd one out. Usually if falls back in line pretty swift if that happens. Oh well, after the euphoria of 18z GFS para, this is definitely a step in the wrong direction. On the plus side, there might be some snow events still - the GFS has a decent one for the south next Sunday / Monday, and the building blocks for a significant cold spell from the east remain in place, even if its just proves to be just out or reach for some of the current modelling

to be fair its not that different to the GFS - timings are a bit different but deep low in atlantic bring mild - holding the block at bay - very much the same theme

Typical model jump around UKMO improves GFS is underwhelming after a reasonable positive 18z ECM - sticks to minimising any cold set up. ECM is being consistent, which is a huge concern for the any cold spell and its longevity - because ECM is not buying in at all.

a five minute covering last Sunday and some wintryness in the cold rain between Christmas and now is the full extent of my severe winter weather - could do with something to brighten the landscape up - the sun has hasn't exactly been in plentiful supply. I might have to stay up for the ECM - despite not going to bed yet

Thats exactly my sort of party!!

GFS is a bit meh - very short snowy cold spell in the south - then gets a bit too excited about a depression in the atlantic

Both good but some marked differences - sharper easterly on 18z Hemispheric heights joined up on 18z but maybe better arctic high on the 0z Certainly would kick the 0z out of bed for passing wind.