fergieweather

No: if anything, tonight's EC Monthly - which, like GloSea, has performed well this season - adds further weight to notion of a chilly start to Spring. It has markedly strong -ve temp anomalies end of Feb & 1st week March, and then retaining (albeit weaker, but that's normal as lead time extends) -ve anomalies right to 3rd week of March. Equally, it shows blocking initially to W then becoming dominant to NW later, with - and this in key contrast to last Thurs output - a drier outcome generally but notably so by 3rd week as signs of NW block establish. This *may* be the model toying with SSW impact. Not sure. Flow throughout is broadly from N (from NW phases at first to more N/NE'ly after mid-month). So nuances aside, it certainly doesn't look mild: at least in *overall* sense. Met Office today retain the broad idea of a prolonged phase of below avg temps into at least early March, possibly further, and EC will undoubtedly strengthen their view. However, GloSea5 *had* been (I haven't checked today) less bullish on colder weather a bit further into the month. Hopefully it's right...and spring can get running after only a temporary chilly start! We shall see. The difficulty here, communication wise, is that some people (not just elements of media) run off with ideas about 'below average' equating to the exceptional conditions of March 2013; or being some coded forecast for Snowmageddon/Big Freeze. Neither are true, of course: we see nil exceptional re severe cold looking ahead. But clearly, snow could be possible. That's not a core part of message until much shorter, reliable lead times.

Just saw latest UKMO 12z Model assessment & MOGREPS update. Essentially as cited above: MOGREPS paints a 10% risk of >5cm snow by end of Fri effectively covering Bristol down to New Forest, across to Dartmoor and up to Exmoor. So, similar risk area to 12z GFS. So, in public forecasts for now, we will just continue to express a low threat of something significant Fri-Sat, for SW/W Country in particular.

It was 15% based on 00z runs. The breadth of outcomes re frontal inroads E'wards (and resultant snow risk) was very marked in MOGREPS. This uncertainty remains now: eg check EC12z (almost dry even in SW by early Sat) versus 12z JMA. It has all the hallmarks of remaining unresolved until short lead time.

Hmm. Not yet convinced. All the forcing has run away SE, perhaps a few PV strips/filaments aside, circulating in the flow N of the departing low centre. Last UKMO profiles I saw for E coast (but not 12z data) looked like StCu of limited depth, under a lot of higher cloud... all unlikely to yield ppn. 12z EC has no snow at all anywhere south of Pennines all the way from T+24 to T+240, but much can and doubtless will change even by Sat.

http://www.metoffice.gov.uk/publicsector/contingency-planners

As mentioned earlier today, for those interested, this is now available on UKMO website. It includes discussion re the anticipated SSW early in March (signposted for some time in GloSea5 and now appearing in other models) and potential subsequent impacts; plus El Niño and other factors.

Met Office assessment remains consistent: - A protracted period of below average temperatures (ie most extended period of such seen for some time) remains key message from longer range output - Timing of any return to mobility after end of this week remains very uncertain - However, persistence of any blocking becomes much less pronounced through early March as mean NW flow becomes more dominant in extended range products - GloSea5 takes us through mid-March on a wetter than average, cyclonic-dominated signal. EC Monthly does likewise, but retaining colder than average conditions - Further transient blocked periods are possible - Self-evidently, 'below average' for late Feb-early March must be interpreted relative to normal seasonal trend: so, that does NOT equate to what it would mean in e.g. mid-Jan!! - Finally, EC Monthly updates again tonight and the UKMO Seasonal Team will shortly update probabilistic expectations for March-April-May.

As usual at this time of year, expect any (less significant falls of) snow by day to either struggle to settle and/or melt at low levels under any insolation. Settling will generally occur readily only at the WBFL elevation or above, but of course still remain a more widespread hazard at night.

Correct: it's expected to be a cold phase not at all unusual for this time of year and nothing like 2013. No talk of severe cold. It'll *seem* more pronounced because of milder winter backdrop, plus longevity (circa 2 weeks; longevity signal then turns v mixed/less bullish, but possible colder/blocked periods at times further into March).

Only points of emphasis I'd add to Gibby's summary are re Fri-Sat: - sufficient signal still within EC-EPS and MOGREPS to maintain a risk of more developmental outcome with gales/severe gales into SW/S, with attendant risk of heavy rain and lower risk significant snow - mismatches this morning between WV imagery and model expectations in critical source area for evolution later this week illustrates the low confidence message re how things will unfold: treat deterministic fields by Fri with a due level of doubt. May remain unresolved until short lead time.

EC-EPS run below average all the way to end of run (see below, Reading shown). Considering too how all 51 ENS members sit in a single cold cluster by 7 March (see below), the support for a lengthy period of NW-N'ly dominated flow, as per GloSea5 and EC Monthly, is starting to get reflected now in other products.

Temperatures are about to start a decline to below average and expectations are for quite a few weeks of that situation remaining, with cold/bocked weather dominating. It could last well into March. Longer range products have only just started reacting to the MJO phase 7 (which tends to have instantaneous response in outcome here) due to lag. Now we can see the direction becoming clearer in GloSea5 and EC Monthly, we await the operational model centres to catch-up. Storm track is expected to become ever more southerly in next 1-3 weeks, with inherent prospect of these passing to S-E of UK and resultant risk of wintry hazards. It's total chalk and cheese change versus winter so far and evidence is now compelling for most protracted period of below-normal temperatures seen for some time. It matches exactly the UKMO Seasonal Team assessment for this winter, but only now is the model evidence becoming well aligned.

In contrast to GFS, the 00z EC manages to keep any meaningful precip away from the south of the UK all through this week (and thus offers no snow at all), but it's clearly a fickle situation with low confidence in many aspects. What joy....

PS Paul, as an aside: with UKV already sold to (at least 3) commercial forecast companies in UK and - I'm told - MOGREPS and GloSea5 soon to follow (this year?), is Netweather likely to become a subscriber for any of that output? I'd imagine any or all three suites would prove popular fare?

...but is it measuring snow melt based just on modelled 2m temperature? Dunno. Appreciate the parallel suite might be 'cleverer'.

Paul, the point I'm making here is that the granularity of GFS snow charts doesn't take account of the critical elements for *settling*. Its showing what would fall into a gauge (as snow... but could, for all we know, melt minutes later). To adequately resolve risk of snow actually lying on the deck, it needs modelling (not just UKV, as you note) that synthesises depth temps etc. PS I think the sole issue where there is clear agreement is how the coming week quite evidently has a risk of snow for parts of the south....!

But it's by definition NOT 'lying snow'. It's the rainfall equivalent accumulation using a PPN total to snow algorithm. That isn't the same as products showing *settled* snow and herein lies one of the major issues with GFS snow charts. Only bespoke HRES products (such as UKV) can adequately synthesise that through detailed modelling of, eg, the forecasted depth temp/land class/topography. Yes, there will inevitably be times when with no marginality to mitigate settling, the GFS amounts will indeed replicate what we have underfoot. But we can't be prescriptive on any solution being the correct one at this range.

For clarity: I'm still posting, just not any further snow prog output from certain models. Cheers.

It's a shame some take to posting discussion here to twitter. Sadly this results in gross misconceptions by mass public and some media of what is being described and discussed. There will thus be no further sharing of this sort of model output from my end, because I'm afraid some evidently run with it direct to social media as representing nailed-on snow threat (it's self-evidently not and was explained as such).

Complex. Showers in N certainly wintry to low levels; but awkward in S with potential for more organised PPN and perhaps wintriness on N'rn and W'rn flank(s). The pronounced, brisk E-NE regime indicated by MOGREPS can only be taken as moderate probability currently, until we see how EC ensembles handle it in coming days.

By 12z Tues, see below (00z run; NOT to be construed as snow necessarily settling). However, and without wanting to sound like stuck record, bear in mind ECMWF develops solid PPN too readily and thus tends to overplay snow coverage (GFS even worse). 00z UKMO-GM not keen on this outcome but given complexities of set-up, it's unsurprising to currently see little consensus. The 'snowier' outcome currently rated 30-50% Tues/Weds. Further ahead, MOGREPS-15 ENS are dominated by a markedly cold, stiff E-NE regime by weekend of 27-28th: far more dominant a signal in that suite versus the more mixed 00z EC-ENS.

Snow signal Tues (as alluded to by Nick and others) is now stronger in 00z EC through M4 corridor. I've yet to see UKMO-GM fields.

...in total contrast to the output from both GloSea5 and EC Monthly, which agree on a cyclonic theme throughout March, generally well-amplified/potentially blocked and with flow between NW-N and occasionally NE'ly.

Who said it had snow by then? Wales aside, a very minor signal by end Tues, eg Cotswolds/Mendips (and bear in mind EC offers solid PPN too readily anyway).

The new parallel (HRes) EC is repeatedly verifying best of all suites lately and albeit running the trend at base of clusters (plume below, 12z, Reading), it nonetheless matches well the colder transition in the sister EC Op. The EC snow signal (by T+210 across much of UK away from the SE, S coast and SW peninsula) is, of course, highly unreliable at this range anyway so not worth posting here. And thus by definition, anything shown by GFS at that range is even *more* unreliable, but that's an old and now wearying story.