beng

Simply that small variations in how the models are handling the distribution on energy in the jet are likely resulting in big changes to the predicted surface patterns at the moment. Hence the rather large change in the ECM medium range output this morning.

Thanks Nick, explains why we're seeing the models apparently flip-flopping at the medium range.

Certainly in my area in the SE, we're seeing less storms come up from France - we used to get several big MCS systems every summer as I recall, but since the mid 90s, these often steer to the East and become Kent clippers if that. I'm not sure how many other areas in the country would have been impacted by these systems, but presumably a lot because they were big beasts that would away rumble all night as they moved North. From memory they typically arrived in the Surrey area between from 6pm on-wards. If we do get a storm now they seem to be later - closer to midnight for some reason. I think there's potentially an obvious change that might be responsible for less imports and it relates to the AMO - which shifted into its warm phase around the same time that we saw a drop in the number of these imported storms. I'll speculate that during the warm AMO phase, there's more westerly energy in the middle/upper atmosphere and that is providing a greater steer of these storms to the East meaning they tend to miss the UK. If correct, we may see an increase in imported storms over the next decade as the AMO begins to drop off into its cold cycle. https://climatedataguide.ucar.edu/climate-data/atlantic-multi-decadal-oscillation-amo

I'm trying to convince my wife that we need to move to the Kent downs - I'll use your photo as evidence if that's okay!

https://judithcurry.com/2013/04/10/historic-variations-in-arctic-sea-ice-part-ii-1920-1950/ This is quite a good read. Of course the problem we have is that no one really knows how good the data was back then (so hard to compare with now), but there's definitely some evidence of a decline around the 30s and 40s (would fit with the AMO cycle too) - although it seems to have been quite variable too (i.e. would bounce back strongly during the cold winters).

Great read & very interesting, especially regarding the relative warmth in the arctic helping drive the blocking. Leaving AGW climate change aside for a minute, this also ties in quite nicely with what we know about the 1940s - as there are records showing a reduction of ice in the arctic around that time (likely ocean circulation changes) while at the same time there was some extreme winter blocking. Ice then recovered and was fairly extensive in the 60s I believe (and then even more expansive through the 70s) so probably a different process in that decade leading to some of the blocked winters that occurred.

This time last year I remember posting a chart of the reconstructed 1962 November SSTs saying that they were the polar opposite across the Atlantic to what we were seeing; the waters west of Spain and down to the Azures were warm: 2017 This year, the Atlantic SSTA profile looks a lot closer to the 1962 setup and should support splitting of the jet stream to our West. Obviously it's only part of the jigsaw that influences the winter pattern (I am not forecasting a 1962/63 re-run!) and the Pacific anomalies are different to 62 (and then there's the stratosphere etc), but at least it's better than last year (if you're after cold) when the Atlantic profile was not favourable. 2018

What a great thread - great discussion and kept friendly and constructive. I'll be honest here - I pulled that list of years by simply reviewing (eye balling) the solar cycle graphs - so it was somewhat subjective; you have a point. I think Summer Blizzard's idea of using the sunspot counts per year has merit, but it's hard to get this perfect as a 12 month minimum may run between 2 years meaning that each of the years has a relatively high count; some minimums are longer than others too - such as the last minimum. We're all agreed (I think) that we have to take these analogues with a pinch of salt, but they're always fun to play around with and there's definitely a blocking signal given off in these analogues for minimum years. We can't conclude it means we're about to repeat 1962 or 1947 (for any Express Journos reading this) only that there's an implied signal for more blocking than normal this coming winter. Anyway, to try and be a bit more scientific, I have used the following years which all had a count <30 - should I use 30 or 10 or 40 - I have no idea - so I stuck with 30. Here's my dataset: http://www.sidc.be/silso/yearlyssnplot 1953.5 20.1 2008.5 4.2 2009.5 4.8 1954.5 6.6 1996.5 11.6 2007.5 12.6 1986.5 14.8 1964.5 15 1976.5 18.4 1985.5 20.6 2017.5 21.7 1965.5 22 1975.5 22.5 2006.5 24.7 2010.5 24.9 1995.5 25.1 1997.5 28.9 Here's the surface pattern analogue for January - still a blocked signal - although maybe a tendency to cold and dry on average. By the way, according to the data at SIDC, 2018 is currently averaging 8.55 sunspots, so we can get an idea of where it fits with the years above.

Steve Murr's raised an interesting point on one of the other threads regarding how we need to be careful with analogues as historic charts don't necessarily reflect conditions in the Arctic. This seems like a valid concern and there's no doubt that ice expanse has been very low over the last decade or so relative to the 60s-80s cold period. I think there was a drop in ice coverage too in the late 30s, early 40s - but obviously we didn't have satellite coverage then, so whether this is a natural cycle relating to the AMO and PDO cycles - or a reflection of climate warming - that's a discussion for elsewhere as it'll only end in arguments and distract from what I want to show. Attempting to take on board Steve's concerns, and reducing the dataset to only include years since the warm AMO began, we get the following Dec/Jan composites for solar minimum. Now granted it's not great to reduce the amount of data in the composite for obvious reasons, but hey there's not much else we can really do here to try and see whether we should still expect the solar cycle induced pattern to hold in the modern day. The good news for cold weather fans is that the blocking pattern is very much still there if we focus simply on the last 2 solar cycles; the main issues that could override it would seem to be: 1.) Strong or Very Strong Enso event 2.) Strong Atlantic SST signal for mild. It seems virtually impossible that 1 will play out, although a weak El Nino which seems possible could actually make this a more snowy pattern IMHO. Issue 2 is is in play though in that Atlantic SSTs are currently favourable for a mild stormy winter, but these may change over the next couple of months. If they don't, then it will be interesting to watch the battle between the atmosphere and the ocean to see how that pans out. If the ocean overrides the solar cycle pattern this winter, I doubt it will next winter.

If anyone doubts a link between solar minima and blocked winters, here's a couple of reanalysis charts for you featuring low solar Januaries since the 1950s. Also please note that although this set includes 1963, removing it makes little difference to the implied pattern. I felt justified in leaving it in since it was a low solar winter. Apologies if I've missed a year from this set that people think should be included. Decembers also tend to be cold and February if anything is the warmest month of the winter relative to average. Most of these winters were not cold throughout, but they nearly all featured notable spells and there are quite a few severe months that occured. The coming winter is currently expected to have a weakish el nino on top which should give impetus to the southern jet and favour systems running into the blocking. The strength of that el nino may will influence how far North these systems are able to get. In theory we should be looking at a colder than average winter with some snowfall at times. The fly in the ointment right now is that Atlantic sea surface temperatures (cold, warm, cold tripole right now) do not currently support this forecast, but these may shift through the autumn.

Assume it's a nowcast then before we know how wide the shower train might be, think I'm just going to be too far North based on that, unless things get really explosive.

Worth bearing in mind that in N France and the low countries receive more snowfall than the global models expect now, then the actual 850 temps may be lowered slightly - assuming the synoptic pattern verified. At t120 on the ukmo the surface wind has an extremely short sea track too, so chance of surface cold undercut off the continent. Extremely complicated setup IMHO, glad I don't have to give a forecast!

The surface wind is due East here on ukmo 120, very short sea track.

http://www.meteociel.fr/modeles/ecmwf.php?ech=168&mode=0&map=0&type=0&archive=0 ECM brings about a big thaw across most of Europe by 168 including Germany, Ukraine and Poland. Possible - perhaps, likely, no. Plenty more model runs before we know exactly what this low pressure system will do and exactly what's going to happen.

No thanks - really not my cup of tea, a complete slushy mess! Hopefully it will correct slightly further South come the day.

Cold pool over W Ukraine/Poland is v slightly bigger and more pronounced - on this morning's ECM this moves around the base of the Scandi high and helps keep the low to the South in the end; so maybe a step in the right direction for keeping the low further South - whether GFS will manage this on this run - time will tell.

Actually, it's very close to the ECM @t144 for the same time (t132) - also this morning's GFS was close at the same time range - so we're getting some agreement for the position of the low at that range (midnight Thursday).

Be careful with the GFS, especially the 12z run. In my experience, it's has a flaw around the 120-144h mark for really blowing up lows. I expect this evening's 12z to be a complete stinker in respect of the low. It may be correct, but would wait until 0z tomorrow before having and confidence at all in that track.

https://www.independent.co.uk/news/uk/this-britain/call-this-a-real-winter-1855923.html There's a bit about it here.

I read about it, didn't experience it!

Boxing day blizzard of 1927 springs to mind, but longer lasting since that started as rain.

Thanks, much appreciated Steve.

Will there be enough moisture with a SE flow and shorter sea track?

I don't think it just the cry wolf issue, but also that there's a psychological issue that applies here (especially after 4 mild winters previously) - normalcy bias. https://en.wikipedia.org/wiki/Normalcy_bias