staplehurst

I suspect the warning for East Anglia is more catered for the ice aspect rather than snow accumulations. (That said of course we have accumulating snow right now in SW Essex). The odd cm or two might be possible very locally, mainly above 100m, but I suspect it'll struggle to settle for most places (if not falling as rain/sleet in any case)

Probably because it's such a small-scale feature (in the grand scheme of things compared to most larger-scale upper-level features). Looks like the upper low slowly weakens as it continues to clear eastwards through tonight, so expected a slow weakening of the precipitation over SE England over the coming hours, but it could still snow for some time yet given it's slow movement and decay.

Here's a loop of the upper low / PVA that's driving this persistent area of snow through today:

Looks like the 06z EURO4 with a lot of snow edited in - you can see where the (relatively narrow) band of precipitation would be in the original model where the line of rain over the North Sea is. Snow has been drawn in north and south of this line to 'blur the edges' and spread the risk I would imagine

I take it the 20-25cm last March is a distant memory?

The Sunday night / Monday North Sea wintry showers the EURO4 always overplays - similarly 2 weeks ago it suggested 10cm of lying snow in Norfolk, when most other models only had 1cm. Eventually it backed off the idea - this is a known issue in the EURO4 with it's convective rain rates being excessively high, and hence convective accumulation too high. This plot from 21z may go some ways to help explain the marginality of tonight's rain/snow. Red shading highlights areas where the dewpoint is warmer in reality than EURO4 expectations, while blue depicts colder.

It has been mentioned multiple times that this event will largely start as rain, and then gradually transition to sleet/snow from west to east as the cold undercut takes place. So it shouldn't really be much of a surprise if you've got rain at this early stage in the evening? It is all going to the original plan...

Precipitation type radars will never be perfect because they are based on model algorithms - so if the model hasn't quite captured the right vertical profile, then the 'precip type radar' will also be incorrect. Also, a lot of these precip type radar products won't adjust for evaporative cooling in the heavier bursts of precipitation. The Met Office precipitation radar (the grey maps they occasionally tweet) uses model algorithms but also compensates for evaporative cooling and adjusts accordingly.

For those asking why they had rain this morning when others had snow... This upper air sounding (a plot of height through the atmosphere on the Y axis (units: mb) and temperatures on the X axis (units Celsius) highlights the tricky precipitation type associated with this morning's trough. A small 'warm' (>0°C) layer melted the falling snow to rain, which then became supercooled in the shallow cold layer below = freezing rain, forming instant ice on impact with ground. This warmer layer mainly affected the western half of the precipitation, hence Northants / Beds / Peterborough etc experiencing freezing rain. Warm nose was less-pronounced farther east, and so Norfolk / Suffolk and down into Kent etc had snow falling in places and a dusting.

The ICON is also notoriously poor at advecting sea-generated showers inland. Usually the only time you'll see showers inland during the winter months in the ICON-EU is if they're courtesy of an organised trough. Countless times you can run the precip accumulation charts out to the end of the run and see how all the precip from the showers stops along the coastline. As an aside note, last Wednesday the EURO4 was consistently predicting 10-11cm of snow in Norfolk from the brief showery northwesterly flow on Thursday. In reality very little to no accumulations occurred, and the model eventually backed off the idea and only had 1-2cm by the morning runs on the Thursday. So I suspect the snow depth charts are a little exaggerated in the EURO4 world for some areas at this lead time.

How do you mean? No storms are / were expected in Cambridgeshire today

It's more an issue with the GFS physics than the fact it's a global model per se - this issue tends to feed into high-res models that use the GFS for its boundary conditions, such as WRF and NMM, so even though they are higher-res models they also have a habit of being too keen on developing showers in the same situation. Several other global models that have the same resolution as the GFS are pretty much dry today.

I've looked at the potential but decided not to issue anything - it really is just the GFS producing anything, all other models are practically dry. Some decent CAPE, but virtually no shear, dry looking profiles too. As you say, if anything did develop it would be pulse-type rather than anything organised, and hence relatively short-lived. So some convection may bubble up in SE England, but will likely struggle to gain sufficient height to produce very much - GFS has a habit of overdoing these marginal risks due to its issues with excess moisture.

Lewis took some photos and videos, and uploaded them here: https://www.facebook.com/lewis.blythe.1/posts/10155386124505913

Indeed, here's the evolution of the forecast for Saturday. In general, everything developed about 50 miles further west than earlier expectations, and hence no point keeping a 'wrong' forecast online when nowcasting would suggest the risk had shifted slightly west.

Worth noting that up until now, no MDT risks have been issued - and hence all forecasts so far have suggested storms are more likely to not happen than to happen (SLGT = 30-45% chance). There are very few forecasts issued in this country that carry high confidence, and the greatest uncertainties are often mentioned in each forecast text. We make mistakes, we're human - the point is to always verify your forecast to see if it went to plan or not, and if not try and find out what went wrong to ultimately avoid that mistake in the future. Two examples from the past month or so (15th March vs 3rd April) where the forecast didn't quite go to plan on one, but succeeded quite well on the other. (Black dots are the detected lightning strikes during each forecast period). In the case of the former, a post-event analysis was posted explaining what went wrong >> .

For those concerned, compared the 06z EURO4's expected snow accumulation for 17:00 today (this includes sleet too, and is mm water equivalent rather than pure cm) vs how much in total may have accumulated by tomorrow afternoon. You have to take into account any melting that has already, so its possible it is over-estimating totals in some places right now

maybe in this event, but remember 10-11 Dec?

Adjusting the 00z Norderney tephi for a 6C with a dewpoint of -4C (representative of North Sea conditions today) gives you convective cloud top of about 2km (6,700ft ish). Inversion then present at this level, with very dry air above. Wiating for the upper cold pool to approach this afternoon/evening to help moisten the profiles a bit more and increase the convective depth. But yes, inversion could be playing a role with the radar, as can blowing snow and the snow itself make the radar over-read the intensities.

Still +2C on the north Norfolk coast too

Easter bunny about to pounce down the North Sea!

It won't matter what radar display you use, it's all the same data from the Met Office (precipitation type will be done independently using model data to determine whether it's rain, sleet or snow etc)

I'm not sure Norfolk was ever predicted to have large amounts of snow? East Norfolk could get quite a bit when the flow swings back E then SE on Wednesday morning with the deep instability around

ARPEGE did well, many days in advance, to consistently pick up the swathe of 80-85mph gusts in Norfolk. Admittedly it stepped back a little during yesterday in its peak wind strengths.