Devonshire

Here you go, hope this link helps: http://www.emc.ncep.noaa.gov/gmb/STATS_vsdb/ Thank you - lots of info there - I will put aside some time to study it!

I think you are bang-on - although some shouldn't be surprised at being called out as 'towel-tossers', toy-chuckers' and - dare I say it 'bed-wetters' (no offence meant against any genuinely affected enuresis sufferers. back to models - my flabber was pleasantly gasted by tonight's ECM run - later stages. Has anyone got any stats about 10 day verification gfs vs ecm ops?

Am novice, so be kind if I am fantasising - but could that suggest the possibility of a cut-off Iceland high coming up?

Casper the Ghost points the way at T96

I can identify with this. I make sense of it in this way: some commentators sit back and look at trends in particular models as well as movements between the models to or away from each other and particular solutions while others fixate on the latest run - reading the former can be very illuminating, the latter less so. Last winter I discovered the 'ignore user' function (hover over a poster's info) - I soon moved up to 9 pages of ignored users!! It makes for a much less frustrating read - and much more informative when something is 'on the cards' and the forum comes alight. (I also wonder if mods might consider adding a metric to posters' info-panels beside their posts, indicating how many members ignore them! - maybe a step too far?! - please feel free to 'ignore' me!) In terms of what the models are doing at the moment, as has been pointed out, they have been struggling to 'sniff out' a strongly-likely and consistent solution going forward - something they will try to do even when there is no such thing, i.e., when there is weak 'momentum' in the atmosphere in our neck of the woods (for want of a more technical term) and things really could go one of very many ways. Anyway, that's how I make sense of frustrating patterns of comments.

ahhh - now that I can appreciate: verification stats of mean versus op, but isn't it in the nature of clusters that there are more than one - you therefore have to choose which cluster to assess for verification - I am in danger of disappearing up a black hole at this point ... but having worked with multivariate modelling in a totally different field I really do like cluster analysis, and you can get an idea of (or indeed actually calculate) a mean from clusters, but in no way can infer clusters (or any information about distribution) from a sole mean. There, the bee firmly out of the bonnet!

I have enormous respect for your meteorological input (as with many on here) - but from a statistical point of view, surely the clusters will always provide more information about what is going on than a mean - isn't that a pretty hard and fast rule?

I might have phrased this better: I believe we are at a similar pre-minimum stage in the cycle as winter '62/'63 (cycle 19) - in response to Yarmy's point about geomagnetic energy lagging solar cycle (and implication that most favourable solar-related conditions for harsh UK winter are a little post-minimum). Our harshest winter in last 100 years came just pre-minimum.

Tentatively - isn't current solar cycle position a good match for cycle 19 (i.e., Winter 1962/63)?! Although we could do with QBO going strongly easterly and La Nina being rather weak for a better match. (It is interesting to note also that the cycle 19 minimum came off the back of a VERY high maximum - unlike current cycle). For what all these things are worth.

This accords with my views - based on my reading re: solar/lunar/Earth dynamics and cycles. Sorry for an enigmatic post, but I've come hotfoot here after looking for a winter thread (prob a bit too soon) - via a search for White Christmas odds! A southerly jet with regular incursions from a very cold Arctic/Northern Europe is my punt for what is on the cards. Mods: feel free to move this somewhere more appropriate.

Sorry, I am going to have to 'call' you on this, (especially as it may be of use in your 3rd year project ) - the nature of statistics includes the very high likelihood (i.e., near-certainty) that for any significant period there will be above as well as below average readings! I am sure I might make this model-related, let's see: yes, when considering the model output for a given run, notice that the mean temperature for ... (goes off to make a cup of tea) ...

I would think solar cycle more worth pinning hopes on that the law of averages - and that looks promising between now and 2020

No intention to suggest ignorance! - probably the opposite; those more experienced and with the knowledge often throw out a one-liner without realising the analysis in their head doesn't automatically leap into the minds of those viewing. For instance on the chart you posted there appeared to be 4 different fronts in close proximity (from my schoolboy geography) and one static chart doesn't tell me what is going where, what with the East-West to-ing and fro-ing - there might be an interesting story in there but I wouldn't be able to guess! It often feels as though there is a communal attempt at Haiku formulation going on once the one-liners really start flowing.

Umm - movement of what, from where, to where, with what possible effect?? (I realise yours isn't the only 1-liner in here this morning, but it is rather perplexing - this is the model discussion thread after all!)

... and to continue the logic: never despair (or be overly optimistic) over what one official agency or 'expert' forecasts at any particular time!

I for one have been loving the crisp bright sunny days and heavy frosts here on Dartmoor! Frankly, have read enough posts for last two months bemoaning the imminent return of wet, mild zonality that never turns up. Snow forecast here again this Thus am. What's not to like?! I only believe the models up to about 48 hrs this winter ...

Is it just me - or does that ridge look like a poor little squirrel holding up it's last nut of the winter, only to find that it's all gone bar a little bit of shell??

Actually - no, it is not fact (even if you choose to capitalise the word). I may not be a model-interpretation expert but I do have a basic grasp of logic and rhetoric: 1. Since we are nowhere near the halfway mark of winter, you cannot say that we have not had/will not have a 'front-loaded winter' - it may still turn out that way if we get a sudden development of 'proper winter'; 2. We may yet get all the way through winter with no more significant frosts 'down south' - in that case, we (southerners) will have had a 'front-loaded winter' (if not a spectacular one); 3. It may turn out that we had a front-loaded winter in terms of blocking patterns over/North of UK offering the *potential* for cold - we may not get these for rest of winter. I am as disappointed as many that we have not had serious cold/snow yet (well, maybe not as disappointed as many on here!), but lets not trash the LRFs just yet. Let's wait and see ...

I just saw a report of a paper suggesting that solar flares come in patterns that make them predictable to some extent: https://arxiv.org/pdf/1612.03504v1.pdf Its too technical for me, but in the right hands it might be interesting as input to experimental/cutting edge models having some sort of short-term utility (esp regarding impact on terrestrial/satellite electrical systems - but also storm intensity/tacks? A similar patterning was reported last year - this link explains it with a video for the hard-of-thinking like me!: http://www.suspicious0bservers.org/january-30-2015/

Hi SS - I started a thread on this a week or two back:

sadly, just Muppetry - the 'ignore user' function is great for this thread, but every now and then I get to see what Kermit and Fozzie and the rest are up to when they are quoted by more sensible members (up to three pages of ignored users now in a little over a week!) - hint for casual observers of this forum: register and you get to 'hide' serial offenders! ... I know, model comment only ... apologies to sensible posters ...

My understanding from having read carefully the posts of the more experienced/better informed posters is that the daily models have not been expected to resolve the situation suggested by the background signals YET - though we should be soon approaching that time wrt mid-December. Until then it difficult to remain calm while the daily model outputs roll out. There seems to have been a definite outbreak of Pre-Midwinter Tension on the forum of late!

That sounds interesting GF - can you say any more?!

Thanks for some interesting info jonboy. Re: I think this gets to the issue - if planetary alignments, and more specifically the circuits (magnetic and otherwise), do indeed influence the likelihood of flares, then I wonder how long it will be until this can be incorporated in short and medium-term forecasts in a meaningful/useful way. I would think the precise orientation to Earth and timing would be the tricky part in any prediction though (as opposed to a general increased likelihood of flaring)

After some musings in the Model thread about up-ticks in solar activity and the possible (dire!) effects on an upcoming potential cold spell I wondered if this topic might be worthy of it's own thread (I wasn't sure if I was being ticked off for posting inappropriately in the Model thread, but then couldn't find an appropriate thread!). I will repost a link to a video that discusses the effects of geomagnetic storms on tropical storm development on Earth: The question is posed: what is the point of giving credence to long-range (Earth) weather forecasts if sudden bursts of solar activity can lead to big changes in weather patterns, seemingly 'out of the blue'. (Allied to this are issues of solar cycles and cloud cover - but I guess this is getting into climate - and I am really interested in weather here). Perhaps more of a practical question is: how can forecasting models (Earth weather) - and forecasters - account for and incorporate space weather into short, medium and long-term forecasts? Any thoughts? ps - don't expect me to respond to technical questions on this thread - I am just an interested layman!