fergieweather

The 00z EC ENS to d15 and clusters look anything but mundane to me. Markedly below average TMax against climatology (taking Reading as an example).

Yes, +ve MSLP centred primarily to W or SW of UK, maintaining temps below average; but some members shift it N, hence risk of much colder flow.

Polar continental

Worth noting that UKMO prognosis for January has shifted now. They suspect PNA may be influential; GloSea5 has moved over weekend to suggest anticyclonic/blocked regime will dominate throughout Jan, with milder interludes only occasionally and temperatures generally lower than average. Although severe cold is still considered unlikely, they note that currently around 25% of yesterday's EC EPS members and several GloSea5 members allow reorientation of the block (principally expected to the W) to shift such as to bring a Pc flow. GloSea5 keeps the broadly +ve MSLP anomalies all the way to the end of January, maintaining a colder theme compared to climatology. Deja vu, or on the money this month? We watch with interest... but gloves & scarves ready from next weekend (at least that initial bit of the story comes with high confidence, albeit cold front timing has wide error margin of 12-24hrs).

Yes they absolutely do. However, global teleconnections influential for the UK remain extremely weak for the forseeable and thus of little forecast use. The only consistent element in latest longer-term forecasts is the unusually strong SPV signalled to remain throughout January, which - taken in isolation - would imply a leaning to +ve NAO. However, longer-range predictions aren't so clear-cut on NAO and the sudden lack of almost any useful signal in ECMWF beyond 1st week of January is, in itself, considered very unusual. If basing prediction on GloSea5 alone, the anticyclonic theme remains dominant - but weakening - further into extended range (ie by d15), with this sinking south to allow a mean westerly flow across us and a gradual upward turn in temperature. However, given that a conspicuously colder phase looks set for first few days of 2017 (scope/longevity/severity uncertain) and signals then suddenly vanish thereafter, forecast confidence beyond week 1 of Jan remains much lower than normal.

Well, we will see. There's no real support in 12z ECMWF EPS for anything so pronounced, albeit both it and the preceding 00z 15-d ENS do drop median TMax below avg (v model climatology) into early Jan.

Yes, consistent broad ideas into early Jan but no useful signal into extended range. MJO remains at very low orbit next 20 days, lending credence to ongoing UKMO view that it remains of no real predictive value into the forseeable for the UK.

Hi Nick. Latest available (Mon) ECMWF ensemble mean forecast into extended medium range keeps it at persistently low amplitude phases from 6 nearer-term through to 8 into 15-20d cycle, albeit with a fair number of members leaning eventually to either moderate amplitude phase 8, or low amplitude 1 / 2. Will see how tomorrow's run compares.

Yes but it's something of a magical mystery tour. Monday’s EC Monthly offered rather uninspiring anomalies on face value, but masked a whole raft of spread from individual ENS stamps. It's understandable why UKMO continue to mention some risk of colder (N/NW'ly) synoptics manifesting into first half of Jan, but the most likely direction of travel is an imponderable at present.

Latest DECIDER products lean to increased risk of NW'ly flow early into New Year; equal chances of high cell being centred over France or W of Ireland by that stage.

You'll note the to-fro in GFS deterministic runs between mobility to blocking/anticyclonic bias in extended medium range. This reflects the bimodal clusters at same range in ECMWF (majority favour blocking, but only just). Recall the same dichotomy from a week or so back... and how protracted it became until models settled into favoured outcome? Watch for plenty of this 'A versus B' conundrum in near future...

Agreed: available evidence including this evening's ECMWF Monthly leans more towards +ve MSLP anomalies initially to our SE, then migrating west to sit just west of and across the UK through early to mid-Jan, with accompanying below avg ppn anomalies and broadly +ve temp anomalies. Looks promisingly more settled versus the next few days; rather similar to last week's run but without any 'colder blocking' evident in the pattern, at least on face value. I've only had a cursory look at the output however: will examine all the 4 week stamps & clusters later, when more time available.

Well, very low confidence...!! As expected. We will see whether ECMWF Monthly this evening keeps banging the blocked drum, or switches and offers the much more mixed signals of GloSea5 (slowly progressive, some blocking, but ensembles unconvincing re ultimate outcome). Dominant signal for remainder of winter (albeit as previously discussed, a divergence in ECMWF & GloSea5 output) is for broadly +ve avg temps (some colder phases possible); +ve NAO (mostly, with possible exception part of Jan) and SPV remaining stronger now. Broad GloSea5 pattern ultimately favours high pressure to S of UK, helping perhaps to keep things drier overall but there's equal probabilistic weighting on +ve/-ve PPN anomalies. December Contingency Planners doc offers further context.

There's no dispute re poor broadscale performance of seasonal modelling for December and consequently all output into J-F-M is viewed with great suspicion. Matters aren't helped by the divergence between GloSea5 and ECMWF products: the former offering weakly blocked Jan; indeterminate Feb; mobile March. EC meanwhile remains steadfast for pronounced blocking Jan & Feb. Both models then lean to +ve temp anomalies during spring. Thus, albeit the ongoing output opts rather more for a blocked/likely colder period developing 1st half Jan (and 00z EC clusters show over half of members heading to a mid-Atlantic ridge early Jan), this has very low confidence, given conflicting signals; weak global teleconnections and the upsurge in SPV apparently at odds with longer-lead output.

Both possibilities are strongly represented. By 31 Dec, only 6 members offer +ve NAO. That *shouldn't* be read as all the blocked solutions being colder (at least not at that timeframe: *may* be different further into 1st half Jan).

Plenty of interest starting to emerge in the spread of extended ECMWF clusters. Let's see if these nascent signs of pattern change into Jan (as per EC Monthly & GloSea5) continue to gather pace now in the ensembles from operational centres.

EC MONTHLY: again like last run. Blocked & dry anomalies developing 1st half Jan as heights move W to north of UK. Colder than avg anomalies southern UK 2nd week Jan (not seen that signal for a while!). East-NE flow. So continuity between last runs and ditto with GloSea5. Don't shoot messenger. Just describing the output: not making any comment on outcome.

The notion of MSLP rising in Atlantic into Jan, with increased chance of blocking and a shift from SW to E mean flow (with risk of colder periods), remains the UKMO longer range assessment (low confidence of course).

This "front loaded" winter thing isn't a phrase ever used in any UKMO assessment. They talked about a higher *risk* of cold weather in 1st part of winter, due to background signals (weak SPV initially; blocking which could result in increased E/NE flow). Whilst we've seen both blocking and recent E'ly phases, the probabilistic risk of *prolonged* cold weather was never high in previous diagnosis and in very latest seasonal update this week, any such signal is lacking anyway. So in summary: yes, there was a clear potential for cold weather into early winter, which was flagged to contingency planners and public; however the caveat re this not manifesting was also made very clear (eg in Adam Scaife's UKMO blogs).

Hi Steve: not sure which models you mean (missing the stratospheric trend), but GloSea5 certainly didn't. Here's snapshot of output from way back in late October (hence not latest modelling, which I can't show). Red is forecast mean; blue the hindcast mean. Note much weaker period foreseen during Nov; then marked uptick in 10hPa zonal winds (most members heading above average) during December. This signal was repeated and strengthened each run thereafter. UKMO were confident on the turnaround in SPV that has now unfolded, and GloSea5 timing for this was actually good. So it's incorrect to suggest UKMO missed this: quite the opposite; they were ahead of the game.

:-) 'phone Adam Scaife and his team before they meet on Monday, Nick... as that's the very enigma they're trying to resolve even before discussing the next 3 months! The fact remains that trying to fathom where these longer range outputs lead us *is* creditable: after all, they underpin everything we later see modelled at shorter range. And just as we see UKV misrepresent (or even miss altogether) the behaviour of a shortwave trough even at T+6hrs, we'll sure as dammit see times when a seasonal coupled model messes-up at T+600 hrs. It'll happen. We accept it. We run hindcasts; we learn; the model learns. Science advances. Snippet by snippet... Whilst it's very easy to trash the seasonal or 30d models (primarily by people seemingly obsessed with some sole desire of snow, or who wrongly assume blocking=frigid), it's intriguing that many of those who devalue the long range bespoke models then willingly accept deterministic output so faithfully at those extended ranges as 'proof' of the extended lead models being 'wrong". That's a judgement we can reasonably make at around 5-7 days. But not at 10-15.

....albeit ECMWF has it focused in exact same area (bear in mind these are 500hPa GPH anomalies in both charts: not MSLP)

It's hard to fathom any semblance of favoured direction, isn't it!? I'm awaiting what UKMO Seasonal meeting on Monday makes of it all (they're discussing J-F-M in entirety). Of course, any possible SSW in Jan would throw everything akimbo anyway.

Current thinking for festive period via ENS is 60% likelihood HP remaining proximal to SE/E (more settled); more unsettled to NW/W. Alternative is 40% probability more widely unsettled with cyclonicity spreading across entire UK for Christmas, with resulting temps ranging normal to a bit below (showery; some wintry).

Yes, but it's the evolution on into January that is a focal point of ongoing interest (certainly at Exeter), with GloSea5 continuing each run to support Monday's ECMWF theme of migrating height anomalies westwards during week1-2 of Jan, with lows tracking much further north over this. The net result is a generally blocked output 1st half Jan, with switch from mean SW or at times perhaps NW flow (Xmas period) to E'ly into Jan and resultant higher likelihood of colder outbreaks. Confidence is of course very low at this lead time and not least - as UKMO point out - hindered by a lack of clear global teleconnections into that period.