Singularity

That disturbance makes a mess of things on this run. It does feel like every little thing that can make it harder to us is very much determined to do so. For whatever reason GEM decided to drop it this evening and that seems to be all we need to get more of a ridge in place next week, but it sadly remains the lesser probability as of this evening. ECM could still get that ridge in on day 9 but it looks tight unless that low out west really slams on the brakes, which it very much hasn't done in the preceding two ECM runs.

Okay so I know I'm threatening to send some people tumbling into madness but this day 6 chart from ECM bares more than a slight resemblance to that ensemble member. Just needs that ridge to trend stronger still. Hmm!

You may wish to consider putting GFS aside for a while as that might improve moods. Or maybe not - let's see what ECM does this evening. As of +120 it's much more similar to the GEM 12z than GFS 12z with the low over by Newfoundland (much deeper with more amplification of jet ahead of it as a result), which suggests a slower progression of the Atlantic troughs may be seen from this new ECM run. This doesn't equate to a finer outcome next week for sure, but it makes it seem more plausible which is a start. Ah here's the +144 now so; A good deal closer to GEM than GFS here. Closer to it than UKMO too, for that matter. Go on ECM - give us a change of heart*. * by which I mean please do so for the love of all things great and good

I must admit I'm struggling to remain calm in the face of these GFS runs that just won't turn the westerly momentum down in the Atlantic-Europe sector. Well not far enough west for us to escape, anyway. Thankfully when we compare the bile-inducing GFS run (left) at day 6 and compare it to UKMO (middle), we find that the latter has the low S of Greenland weaker, further north and with more of a ridge in between it and the trough by the UK, which itself lacks the shallow disturbance on S side that GFS has decided to introduce. All four of those things indicate a setup less conducive to the low S of Greenland storming east and hammering us with considerable force. Of some interest is that GEM has managed to slow the low down a good deal further, but this does seem a bit out-there at this stage. A shame because that run is the first in a while to turn out pretty good next week; If ever we have needed GEM to be leading the way in spotting some very warm/hot weather potential, it's now! Everybody prepare to clutch those straws I do wonder, though, if the model output can be trusted much at all beyond even 4 day's range at the mo, given some pretty odd looking MJO projections and the fact that the observed MJO took a marked step into phase 5 on the H-W diagram yesterday when most models had a move into the COD (central circle that represents weak or indistinct MJO activity): I can't remember seeing all the models depict such a there-and-back evolution of the MJO before. I've seen GEFS do something similar in phases 8-1-2 before but that's as close as I can recall.

ECM and UKMO show a setup conducive to gales, maybe even severe in the case of ECM, in just 3 days time, when GFS and GEM have it little more than very breezy! It comes down to a feature with some tropical characteristics that swings into the upper trough and deepens substantially. The precise track of this makes a big difference to how windy it is Wed-Thu (and how wet for western parts on Thu). Not only that but it makes it harder to get much of a ridge in Fri-Sun although ARPEGE still looks to be having a fair go of it as of +114 despite having the low close to the ECM position at +72. At the moment, regarding next weekend's pattern, GFS is in my opinion the best fit for the GLAAM state, but ECM is the best for the rampant C-Pacific trade winds. As those trades finally look to ease off during said weekend, there is more chance of improvement early-mid next week. ECM and GFS hint at this but with little conviction at this time (and ECM day 10 is just plain rude given the day 9 - give us a break why don't you! ). That being the difference between our current fortunes with near-neutral GLAAM and when GLAAM goes into a low orbit of the GSDM during these summer months, which is when a longer lasting unsettled period tends to occur. I know it all sounds rather tenuous but the fact is we have recently been thrown from the sunny path into the dark jungle and are now struggling to find an escape route while wondering just how much unpleasantness waits in the shadows!

http://www.theweatheroutlook.com/twodata/chart.aspx?chart=/charts/gfs/12_150_mslp850.png?cb=119 The Atlantic though is a nice distance further north and a bit west on this run. Going to need a pretty big shift from ECM to boost confidence in a fine weekend for much of England, though.

The model output is too chaotic beyond Wednesday to make it worth saying much so I might as well go with the wildest proposition; from the day 9-10 ECM 00z and GFS 06z charts, I can easily envision the trailing frontal boundary down west of Iberia triggering formation of a low in that area with a ridge over Europe, this coinciding with a lot of heat having piled up over much of Europe as a result of the 'cyclone cannon' firing large troughs N. of there - with subtropical airmasses flung east across Europe - over the preceding 4-5 days. A sequence seen all-too often in the winter months! Odds are those troughs won't be as deep and the initial build of European heat less dramatic, but this also improves the chances of at least southern parts of the UK seeing some hotter days in early August. As I said, this is a wild proposition and requires a sizable dose of luck (or bad luck if you don't like heat), but I couldn't think of anything else really worth adding to that already covered by others this morning . Of course even this broad idea has been floating about on here for a few days now, to which I say - well spotted!

ECM's eventually like, 'alright fair enough, I'll edge that plume close enough to tease/torment UK residents (delete as appropriate) . It is amazing how readily and far Europe keeps on heating back up at every opportunity this summer. I must express my disapproval of it's decision to produce the flattest jet stream configuration on days 4-5 with the least ridging across the UK. Probably those anomalous easterly trades at work given how this model usually tends toward the other side of the coin. It has the least enthusiastic eastward MJO propagation so there's little working to try and weaken the trades. That the 12z gets anywhere at all (days 8-10) having been so resolutely flat in the earlier stages does I suppose offer a little further encouragement.

GLAAM has stepped up a bit and looks to stay there, and good support for eastward MJO propagation from the Indian Ocean has emerged over the past few days. This should help mitigate the influence of the strong C Pac trades but by just how much is clearly very uncertain given recent model flailing. There is a discernable net trend toward more ridge influence from Europe which is encouraging, as is the fact that more complications to the Atlantic jet are more likely to help rather than hinder us next week (typically more complications emerge nearer the time), but these trends and tendencies are still fairly slight or tenuous so for now I'm remaining cautious with respect to getting my hopes up for later next week. A good shift toward more settled from ECM, which has the strongest C Pac trades and has - likely as a result - been most reluctant to settle things down, would be the best sign so far and start to raise the possibility toward a notable level. Fingers crossed!

Next week Tue-Fri may be one of those occasions where a low breaking away from the westerlies and dropping down somewhere near the Azores could lead to a transient (assuming background signals remain against a sustained eastward Azorew High displacement) very warm or hot spell at short notice. Only faint signs so far that such a feature could turn up, but they are out there. At least that gives us something to watch out for in hope. A straw to clutch .

Cell from near Wimborne did pass not far to west of here, but most of the lightning was obscured by a possible wall cloud. Result was like when someone has a TV showing a fast-paced action movie next door at night and the curtains are drawn. Signing off for the night. A decent event but only half what 2014 managed on each of its two occasions. Clearly though I was within just 20 miles of a biblical deluge and absurd amounts of overhead lightning

Uh... The cloud bases have suddenly started moving rapidly to the west over here. Strong inflow channel...? Possible tornadic signature to cell now just north of Wimborne Minster but given elevated storm bases unlikely to be much at the surface. Probably.

Looking back over the past hour or so there must be some phenomenal rainfall totals between Brockenhurst and Lyndhurst. That's also where most of the lightning I've seen has been from.

So the Southampton activity is now back at mostly flicker-only range, but a cell that erupted over Poole has moved into that range and may well be on a track taking it just west of here. It does look as if the Avon Valley will work its downpour-dodging magic even tonight as I've not had more than intermittent heavy rain so far. Ah there goes the first crash-bang in about 15 mins so it seems the cell from Poole is drawing near

Fingers crossed for you while this storm takes over my world... Man am I going to hit the mattress hard later!

Sounds about right - but then it is nearly 2 am and I've just spent an hour doing slow-mos on my iPhone and deleting the fails along the way The most intense cells have been sneaky and escaped NE from the point of origin, leaving me with lesser action to the south where I have a less obstructed view. Caught 5 bolts in good detail so Im pleased enough

If I recall correctly from earlier forecasting efforts, whatever hangs on should reach as far as East Yorkshire by mid-late morning tomorrow, maybe even a bit further than that before going too far to the east for most. Getting that 'big one' feel here now; lightning of one variety or another every 4-10 seconds including one a few minutes ago that chained for a full 3 seconds. Rain has set in too but only moderate. Could get a whole lot more intense later - seeing some cores reaching over 150 mm/hr on NW Radar which is as much as I've ever witnessed first-hand.

Ah, some more prominent thunder now and more in the way o blinding flashes. South of Bournemouth the activity seems to have consolidated into a vicious arc which shows signs of a more concerted effort to move NNE. Arc formations and storms are usually a potent mix, but typically they have to be surface-based to bring any severe winds down. Could be some localised fairly sizeable hail though.

Finally, a crawler close enough to produce a lengthy grumble of thunder. First of the night, 100 or so minutes after the first flickers became visible

Proving tricky to anticipate the behaviour of this belt of storms. It's not taken on a nice neat line as was seen during the two light shows of summer 2014, instead there are multiple arcs of activity tending to break away landward only to fade out as new and more vigorous cells spring up back out at sea. I believe this is what's known as back-building and is why it seems to be taking illogically long for the overall mass to make its way inland. I've had flickering and occasional crawlers to the south for near 90 minutes now!

Finally seeing a few crawlers to the SE of here - and they're a brilliant shade of orange

UKMO 12z ends up uncannily similar to some of the charts that were cropping up prior to this week's extended fine spell that never was and GFS is not that far away from that - just a bit flatter with the Atlantic jet which is handy in the mid-range but potentially less so in the longer. This output is following an AAM climb that has gained some magnitude in the GEFS projections over the past few days which begs the question - are the models overreacting to such trends for the second time running?

Suddenly it all appears nice and simple with the low drifting slowly but surely east Sat-Tue and then a bit of a ridge in from the west as Nina-like forcing wanes - but with not much sign yet that there will be enough forcing the other way to achieve more than a warm mostly dry south and cooler changeable north until at least next weekend. The 12z GFS last night seemed to find something much more, but it was clearly along way outside the curve (unless it was ahead of it instead... ha - that'd be the day!).

Uh.. what just happened? I mean really GFS, really? A sequence of charts that would rival the most interese heatwave of 2003? It must really be making something of eastward tropical convection propagation to counter the C-Pac trades. Either that or killing those off sooner and more definitively - not had a chance to check these things yet as I'm out and about. Needless to say, UKMO still has none of these adjustments, and from what I can make out GEM is not better.

Quite something watching these ideas bounce between the models. GEM was keen on the merger with S/F trough yesterday leading to a fast exit but a cool and showery NW-N flow behind (this model does tend to over-amplify in a big way though), but handed that over to each of UKMO, ECM and the EC Ens. this morning. Meanwhile GFS/GEFS have been trending toward yesterday's 12z ECM and in doing so has now gone beyond the shift that the ECM Ens. underwent yesterday evening. I am starting to suspect that a factor behind this extreme variability in LP position and movement is a struggle to resolve finer detail regarding a fairly dramatic N. Atlantic SST anomaly pattern and how that affects the strength of thermal winds (i.e. the jet stream) on a relatively small scale; we have a vast area of anomalously cold waters to the west interacting with a swathe of notably above average SSTs extending from the eastern subtropical Atlantic to the North Sea (and increasingly Barents/Kara so far this month). This being superimposed onto the issues handling AAM changes. Orbits in the GSDM plot are not that often as fast - with large dM/dt (difference over unit time) threatening to rapidly amplify any errors in modelling - as we're currently seeing. A word of caution is however due regarding the emerging Euro-Scandi idea from GFS/GEFS, as this model is the most enthusiastic by far with propagating tropical convective signals east - something that's hit the wall a number of times lately. While the convective signal looks weak, such can still tip the balance in such delicate situations. I wish I could see what the EC ens. are doing with the MJO for example but the plots from this model have gone AWOL as they are known to do from time to time.