IanT

Rain here in Woking is truly extraordinary. I have never heard anything like it. Will see what damage it has done in the morning...

Still falling. We have about 8cm now. Looks like the end though... Time to take the dog out for a walk!

Big flakes here... IMG_2111.MOV

Plenty here just down the A3. I’m sure you’ll get some soon.

Snowing heavily here, just S of Woking. Temp -1.2c having risen by 1deg or so. Not worried it’ll be an issue for here looking at the radar.

For once it all looks to be being brought forward, and the front itself looks bigger. I think this is going to be a more widespread event than was anticipated. Good! Where do we think the SE sweet spot will be..?

Oooh no...! Haywards Heath very definitely not Gatwick. Not if you live there anyway... ;>

We have entered a period when - perhaps as a result of the much discussed SSW - there is significant uncertainty in the evolution of our weather within NWP horizons. Many posts in the MOD thread say that the models aren’t handling this well. My take is that the models are accurately reflecting the uncertainty, and it’s actually the posters who aren’t handling this well..!

The lack of explicit acknowledgement of probabilities lies at the core of the problems that surface every winter on MOD thread. Charts at T240 are discussed with the same excitement as those at T48. The extremity of the output drives discussion, with inadequate discussion of the likelihood of verification. I don’t know what the answer is! Perhaps a thread for discussion upto (say) 120h and another for anything thereafter? Or a concerted community effort to pay more attention to qualifying the likelihood of the outcomes being discussed? It is pointless saying “If that verifies COBRA will be meeting” without saying something about whether it is a 2% chance or an 85% chance.

My experience of modelling comes from financial markets, where it has long been understood that most people are very poor judges of probability - perhaps especially where outcomes that provoke an emotional reaction are concerned. This is why lotteries exist. Reading the discussions here is eerily familiar in terms of cognitive biases. Many are on display! For forecasters the main CB’s to be aware of are confirmation bias (we are prone to see what we want to see) and anchoring bias (our ability to interpret new data is skewed by recent data). Thus it is human nature to be more interested in the idea that the Thames might freeze over in 10 days time once that possibility has been flagged, than in the 5cm of snow forecast for tomorrow. Understanding our predilections is an edge in itself. As in financial markets though, quantitative modelling allied with human experience is the best toolkit we have.

High. We’ve been discussing essentially the same output - snow in 7 to 10 days time - for weeks. Nothing in the charts to suggest that will change materially in 24h.

So where is the “POMIC” (point of maximum interest in charts) now? 6th Jan..?

I think this idea of “the big day”, the point of maximum interest in the charts, the “Hugo point”(!?) - might be a useful concept. Prior to the now-casting stage it’s interesting to know (or discuss!) whether the Hugo point is getting closer or receding into the distance. If the Hugo point is holding steady - or better still advancing - that’s good news for the many posters here seeking wintry hazards. My sense is that many of the failed cold spells of recent years have been hinted at through a receding Hugo point before discussions here have been swamped by disappointed reactions from posters who feel they’ve been cheated etc. The current Hugo point appears to be January 4th. I will try to keep track of its evolution.

An alternative explanation is that many differentiated evolutions are possible from the present set of initial conditions, and the models are reflecting this accurately.

Although 10 days away...

Those charts are for the 9th January, not “by the New Year”...

ytilanoz...? Very rare.

It's all about the difference between probabilities and possibilities. Most of the posts we read here are explicitly highlighting possibilities. They may become forecasts if and when their probabilities are higher and conviction can be built. Weather enthusiasts tend to focus on possibilities, whilst pro forecasters are professionally constrained by probabilities. Forums like this are fascinating because they provide a venue for the Probables v Possibles match...

Hello sailor...

Cooler here (Woking) than generally modelled. Showing -1.5c vs +2c. Heavy frost. A few flakes mid-morning may just be possible... Well I did mention modelling

The adage doesn’t imply that a cold spell will produce snow. You may have hold of the wrong end of the stick. How many times have you seen snow without cold?

Can you point me to the source of this please?

I think this is utter nonsense. There’s no upside for the government in deliberately misrepresenting the weather forecast. Not happening. We know that the MetO are more cautious about forecasting snow than many here. We also know that they are generally more correct!

Did they post any charts to support their assertion? Did you at least look at them? If not then please accept that unsubstantiated stuff from Twitter isn’t really what people here are looking for...

Well it’s frequently brilliant at T+10 in the models. Not yet on the ground at T0.