staplehurst

They said snow was possible to low-levels in the heaviest precipitation, but unlikely to settle below 100m

I should probably change that - that's where I used to live several years ago when I first signed up, but have been in Norwich for quite a while now!

Just from a brief glance without looking more into the specifics, looks almost sting jet-esque

We're pretty much at the point where TV broadcasts will offer little value in detail now - it's more down to nowcasting, watching the radar and seeing how things evolve. Likewise site-specific forecasts on websites or apps will offer limited usefulness because there is no human input and they will change with each new model run that comes in.

Yes, most national presenters are trained or qualified meteorologists. The vast majority of regional presenters (in the BBC) come from a more journalistic background, though some have a strong interest in/understanding of the weather. In a situation like this though, I guess the issue with some broadcasts is that the model data used for graphics can be up to 12-15 hours old, with several of the broadcasts pre-recorded earlier too. While this is still very useful for the general public, I suspect for many in here as weather enthusiasts you're probably better following posts in here, looking at the various model output that routinely comes out throughout each day etc as that will give you more up-to-date insight into what might happen for your local area.

Yeah, shifts are very infrequent - tend to average 1 per month at the moment. Last one was early November, next one 27th December

12z ARPEGE has snow into north and east Kent now, the southerly shift continues...

It was certainly a busier Saturday shift than usual today, thankfully off tomorrow!

In a basic sense, it's still way too early to take model output that far ahead with more than a pinch of salt, as it will keep changing with each run. The broad theme is there though - much colder air arriving on Thursday behind the cold front, thereafter there could be more organised trough features running south in the flow that models won't be picking up on yet. Incidentally, a lot of global models always struggle with inland penetration of coastal showers. Another thing to keep an eye on is the low that parks itself up over Scandi - depending on its proximity, it might try to push some slightly less-cold air in from the northeast at times. Intrigued to know where you'd heard just sleet showers were forecast for Lowestoft specifically? There was always a good chance places like Lowestoft and Southwold would get a light covering of snow from those showers given the favourable offshore wind direction.

Data from Naples - looks like pressure fell to 939mb as the eye moved through around 2100 GMT. Winds dropped to 6mph having been sustained 50mph just over 30mins beforehand.

Here's the current difference between observed water level and the normal tide level...

Worth pointing out I guess that the sudden increase in water level is still at or slightly below the normal tide level, but if that rate continues then obviously there will be significant issues, especially once the winds swing round...

6ft now in 30mins, and the wind hasn't even swung round to a W or SW'ly yet...

3ft rise in about 15mins at Naples after the gusts of 135mph

Interesting to see water levels going up with onshore flow in eastern Florida, yet the reverse in western Florida - that will change fairly rapidly I suspect once the centre passes northwards...

Tracks from 4 days ago (6th), where only a small minority of ensemble members had Irma going up the west coast of Florida, or even staying out over the eastern Gulf. There were a couple of runs of the GDPS that had Irma going as far west as New Orleans a few days ago. and of course just a week ago Irma was forecast to be either a fish storm or make landfall somewhere in North Carolina according to the majority of NWP guidance back then...

Just been watching the Prime Minister of Antigua and Barbuda describing the extent of the damage on Barbuda on his visit there this evening - this screenshot explains why (amongst many things) the residents of Barbuda couldn't make any contact with anyone... The aerial footage etc can be found here:

I thought it odd several hours ago that the PM of Antigua and Barbuda seemed so upbeat at how the events had unfolded - thankfully no deaths on Antigua and no major destruction - but it came across as if he thought the storm was perhaps not as strong as they were expecting and he seemed completely unaware of the fact the eye had gone over Barbuda. He did acknowledge the fact that they haven't had any contact from Barbuda since 3am, which was presumably before the eye had even properly passed over the small island? Now that he's been to finally visit Barbuda, he has apparently seen '90% of buildings damaged or destroyed'. https://twitter.com/ldobsonhughes/status/905528629010067456

Yep, 'fraid so. Seen 'live' footage of the Wray tornado this evening being claimed as Hurricane Irma. Most pics / videos these days I often find myself questioning their authenticity - sad times.

I've seen that video before - it was filmed in early July in China

Vast majority keep Irma's centre mostly to the east of Florida (may get pretty close to SE Florida) and then potentially make a landfall along the NC/SC coast - but a lot of chopping and changing to go...

It was mainly the lack of confidence in how widespread lightning would be that prevented any upgrades to MDT. For the UK, it was quite a narrow corridor of lightning and makes it hard to be that specific...

Certainly feels like a decent summer for Norwich with many thunder days (lost count) compared with previous years.

I feel it was reasonably well-forecast? Certainly quite a few storms around, hope one can make it up Norwich way...

I think those charts are showing surface-based CAPE (SBCAPE), which would be derived from lifting air parcels from the surface and then calculating the resultant CAPE available. The storms on Tuesday night will be elevated above the surface (and hence not necessarily affected by surface parameters), so MLCAPE (mixed or mean layer) will be a more useful guide as to how much CAPE may be available.