Model Output Discussion - Spring Has Sprung

[Mike Poole]

ECM showing that there is potential in the upcoming setup.  Close but no cigar at T216, but the potential to land something more significant is there before that.  

While I’m on, here’s the translated blurb for the AIFS model on Meteociel, for those interested:

‘These maps present data from the experimental European CEP/ECMWF AIFS model available in Open-Data. The maps are updated randomly according to the runs but generally one to two hours after the classic deterministic model is released. The resolution of the Open-Data model is approximately 0.25/0.28° (or approximately 25km/30km). There are fewer fields available than for the classic model. All fields are available up to 360h for runs of 00Z, 06Z, 12Z and 18Z. These data are provided on an experimental basis for testing forecasts made by artificial intelligence/machine learning, the calculation of the model by the AI is done in one minute (!). The model was trained with ERA5 reanalyses and classic ECMWF HRES runs.’

The model takes 1 minute to run - wow!

[Mark wheeler]

Hmm, ECM like a dog with a bone . So close .

Edited March 3 by Mark wheeler

[daz_4]

Quite the winter wonderland for Central Europe. Thanks, but no thanks. Pretty sure it will be a huge outlier.

[feb1991blizzard]

 nick sussex

Yes, just to add insult it blows into a full blown low at 240, only one result from here, the same one we've had all winter.

[Lukesluckybunch]

1 minute ago, Mark wheeler said:

Hmm, ECM like a dog with a bone .

Imagine that 192 chart I. January..you would be looking at highs of -5!

[minus10]

 Mike Poole Thats just crazy...1 minute.....not expert in this but dont current model runs take a few hours?..this AI has the potential to revolutionalise the way modelled forecasting is produced...i should imagine far less computing power is needed although the AI still needs the input of traditional nwp runs..historical to inform its continual learning...but will there come a time when that is not required ?...

[CoventryWeather]

Looking at the short-term this evening, much drier than of late. Some showers overnight and warm dry days. Feeling a lot more seasonal. 

Seems to be a fair few upgrades early on in runs which seems to be making a difference towards day 10. Scandi High is being underestimated again. 

[Mike Poole]

 minus10 Of course, we don’t know how long the AI model took to train itself to get to the point where it can knock out runs in 1 minute!  And let’s not forget it requires as input the self-consistent T0 dataset calculated from the input data - that is a non-trivial calculation in itself which would take a lot more than 1 minute, it kind of gets that ‘for free’ as the main ECM models need it too.

Edited March 3 by Mike Poole

[stodge]

Evening all

Without worrying too much about the medium term or FI this evening, the initia 96-120 hours seems to be falling into place with the Atlantic trough disrupting SE towards Iberia thus confirming some form of Easterly flow for the British Isles. Next weekend could be another washout for southern areas in particular.

The heights to the NE don't look robust enough to hold and after a brief col, the Atlantic trundles in agsin but it's a very slow process commensurate with a very weak jet.

Plenty to be resolved yet.

[damianslaw]

Taking cursory glances at the models presently, when heights to our east are set to develop, you never can be confident just how things will pan out, and in such scenarios I give them far more leverage and room for error, expecting them to change at short notice.

In this context, I can't call which is closer to the likely end destination. We may see a southerly, se, east or ne flow by the time we enter mid month period. 

.

[Methuselah]

 minus10  Well, I'll welcome AI in weather forecasting; it might do away with all those blizzards on Day 10! 

[Lukesluckybunch]

The best ecm mean yet at 192.

[Cloud 10]

So that's 4 ECM Ai runs per day.

 

As if the rollercoaster wasn't bad enough.

 

Also to note the SSW is pretty much underway now.

 

Edited March 3 by Cloud 10

[Mike Poole]

ECM clusters T120-T168:

Support for the easterly on this timeframe growing, with clusters 2 (contains the op) and 3 - both take the low well south under the block.  Cluster 1 more inclined to hold it out west, but this is the solution on the decline, I think.

T192-T240:

I think this timeframe is still work in progress!  The UK rather in a no man’s land that could go one of a number of ways.  More runs needed…

[Cambrian]

12z GEM op is very interesting indeed. No mucking around with that shortwave around Iceland, so has a cleaner route to cold, the low going under the high into France at day 8. 

 

Then the bonus of moving it through to Poland by day 10, reinforcing the feed of cold air from the northeast into the UK and Ireland, with further signs of Atlantic system disruption.

 

Nice clean route, which is what we need this time of the year. 

[January Snowstorm]

 Cambrian not so sure, the Atlantic looks ready and primed at 222hrs

[ANYWEATHER]

Looks like the Atlantic is going to having a huge fight with Mr Scandi .......

[Lukesluckybunch]

Icon 114..looking good cold to the northeast creeping closer!look at them heights north of scandi..we would dream of seeing that in winter!

Edited March 3 by Lukesluckybunch
Adding

[Man With Beard]

 Lukesluckybunch This is a great position to be in. I do feel good about a potential snow situation the week after the coming one - in recent years, models have underdone the speed of everything coming from the east at D8 when accompanied by northern blocking, so I would place my money on the easterly gaining momentum in the coming days, not losing it.

[Rain All Night]

9 hours ago, Rain All Night said:

0z/6z deterministic runs, out to Sun 10th (day 7)

Oh pleeease let that low go under and not blow up, I so want to stay dry...

0z/6z ensemble means, out to Mon 18th (day 15)

These ensemble means suggest that the low won't go under, and that the pattern will flatten out from 13th March (or earlier on the GEFS), with less sign of residual mid-latitude blocking than yesterday. Rats.

12z deterministic runs, out to Sun 10th (day 7)

We see significant divergence after day 5 this afternoon. Not sure what on earth's going on with that ICON run!

Only UKMO is showing the absolute nightmare washout scenario for next weekend, so that's something, though it doesn't look like the south coast will be able to escape some rain in any scenario.

12z ensemble means, out to Mon 18th (day 15)

The ECM and GEM means are now suggesting that the low actually will go under - but also that the Atlantic will still come back into the picture a few days later regardless.

The GEFS again prefers to bring the Atlantic back more quickly and with less equivocation.

Bonus

Meanwhile, far above, it's looking increasingly plausible that the stratospheric polar vortex may not be returning before the autumn. But what, if anything, will that mean for us?

Edited March 3 by Rain All Night

[LetItSnow!]

Caution should be taken with this pattern as I do fear that a situation like the UKMO could occur if things don't land 100%. At least for the south there probably could just be a whole lot of cold rain, as apposed to February's mild rain! All hangs on whether high pressure will get a strong foothold around the 11th or whether the Atlantic will win out. Can't say personally I'm chasing cold charts now but it would be funny if March ended on the chilly side after February - but I don't see this likely just yet.

[Lukesluckybunch]

Early stages..but the core of hp..further North on the 18z!

[WYorksWeather]

On a temperature front - worth mentioning that the 12z ensembles still don't look overly cold.

I am concerned that a failed attempt to bring cold in from the east could as others have suggested lead to further heavy rainfall, which is the absolute last thing we need.

In terms of the ECM, the meteogram shows a generally average temperature trend for northern England, then rising to somewhat above average, before falling back below later next week. Beyond the 12th the wind direction plumes become uncertain to the point of uselessness, so I would discard anything beyond that point at this stage. High confidence though in a south-easterly turning increasingly easterly, before then turning north-easterly and allowing a colder interlude. Unfortunately, the temperature plots along with the precipitation spikes indicate a lot of cloud, wind and rain with this pattern beyond day 5.

In terms of the broader outlook, worth repeating my usual look at SSTs, which I've not done for a while now.

First, the absolute SST values, for days 0, 5, 10 and 15. Not much change here yet - SSTs are clearly near their annual minimum, so not much change from day 0 to day 15.

Below are the anomaly values, for comparison, for days 0, 5, 10 and 15. Not a huge amount of trend to be seen here. Probably the most we can say is that the Azores warm patch decreases in magnitude a bit around days 5 to 10, but then is forecast to recover. The seas immediately to our west also cool down, so we should probably expect that any westerly patterns later in March will bring temperatures that are somewhat less turbocharged than we've seen of late, though anything south of west will be much milder than we'd usually expect if the air is sourced from the Azores.

 

[Lukesluckybunch]

Nice 18z run..we get some kind of easterly around day 8 onwards..I still think we can improve around this timeframe..but finally some consistency with the gfs and ecm!it's still looking pretty decent approaching day 10

[feb1991blizzard]

Respectable GEFS, nothing -10 and blizzards yet but steady as she goes for the opening gambits of spring, setting up something more cold and snowy as we get further into spring.