North-Easterly Blast

No, I do not think that September 2022 was exactly cool - it had a CET of 14.4, which was still slightly above the most recent 1991-2020 average, and it was still quite a warm month by the older 1971-2000 and 1961-90 standards. More correctly put, I agree that September 2022 was less above the average, compared to how July and August were, and the October and November that followed. January 2022 although it started extremely mild, and was mostly dominated by high pressure over the UK, with little in the way of cold outbreaks from cold synoptics, did turn out to be a pretty average month by the most recent 1991-2020 average, but even then it was a fairly mild month by 1961-90 standards, and was still above the 1971-2000 and 1981-2010 averages, so it definitely was not a cold month.

Given the output at the moment, myself and likely many others are fearing that we have already had the best of what winter 2022-23 has to offer, which was the cold spell from December 7th to 17th. Last month the cold spell did produce the first colder than average month since May 2021, but given the output at the moment, it is difficult to see January being anything other than above average, and possibly a very mild first half.

What about 2010? That December had a CET of around 0.1 by mid-month I believe.

Well to make this month the coldest since 2010 it would only need a CET of 4.7 or less. That is almost a certainty now. In order for 2022 to record the first ever 11*C+ calendar year the December CET needs to be 2.5 or more. It looks set to turn much milder by the 18th / 19th of this month, but some models are suggesting that the cold could return after that, but to be fair, a mix of milder and colder outcomes are being shown as possibilities in the ensembles, over the Christmas period and beyond. Although the first half of December will have been notably cold, with a significant cold spell in the second week to mid-month, it is still too early in my opinion to say where this month's final CET is likely to end up.

What some may call an exceptionally mild November / December 1953, was actually cool in comparison to the same pair of months in 2015! Nov 1953 still had a CET of 1*C less than 2015, and in Dec 1953 the CET was still 2.8*C less than 2015, so 1953 puts into perspective how exceptional the same pair of months put together were in 2015.

The CET for December 1962 of 1.8 was still 2.5*C less cold than Dec 2010, and 1.5*C less cold than Dec 1981, so yes Dec 1962 was still a significantly cold month, though by no means as exceptional as 2010 or 1981.

I do not think that December 1962 was severely cold for all or most of the month. I think that month really became severely cold from about the 22nd onwards, which set the scene for the whole of the rest of the winter. I think that the first three weeks of December 1962 were more like rather cold though not significantly so. Whereas December 2010 saw severe cold for most of the month right from the start, as did December 1981 from about the 8th, and December 1950 was in some ways a bit similar to December 2010; very cold for most of the month albeit not as severe as in 2010. Whereas December 1962 was slightly different; in being generally below average but not significantly so for the first three weeks, then severely cold for the final ten days. I would not say that December 1962 really fits the same pattern as the Decembers of 1950, 1981 and 2010, as the first three weeks of Dec 1962 were certainly nowhere near as cold as the Decembers of 2010 or 1981 or even 1950.

Some great looking charts appearing in the model output, but to me it looks as though there is the chance of some excellent cold looking charts for the UK on paper, but there looks to be a chance that the lack of cold air over the continent may mean that the weather we experience in the UK is not desperately cold. Looking further ahead, there is some excellent Greenland blocking being shown and whilst some model runs do show the really cold air making it into the UK, there are other runs which keep troughs of low pressure very close to the UK and the really cold air not reaching us. I think it is likely to be this weekend before we know fully if the really cold air from the Greenland High will make it to the UK.

Looking at records and summaries, December 1950 appears to have been in some ways a bit similar to December 2010 albeit with the cold not as severe, but still persistent cold for most of the month and significant snowfalls for many parts of the country. The cold of December 1950 may not have been as severe as December 2010, but the general pattern in the weather charts looks fairly similar for both of them. December 1981 was another very cold December that was in many ways very similar to 2010, albeit not as cold right from the start as 2010 was, but when the severe cold spell kicked off at the end of the first week of Dec 1981, the whole pattern was as impressive and the cold just as extreme as Dec 2010. Looking at all three of those Decembers that were severely cold, none of them went on to see a winter in which sustained cold held sway, with winter 2010-11 in particular finishing up very mild in February. In actual fact if December is severely cold, it rarely bodes well for the rest of the winter. The best example that I can think of for a severely cold December that went on to see the rest of the winter being cold is 1878-1879; now in that winter December was severely cold, and in many ways much like 2010, and the severe cold lasted through January 1879, and although winter did moderate somewhat in the February, it was still quite cold by today's standards.

Well the incoming 2001-2030 averages for almost the first 22 years of them are already known, but the final ones will not be known until 2030.

The most notable feature that I can see about the warmth of 2022 is that it has not contained one single colder than average month (that is even below the most modern 1991-2020 average let alone below the older averages), although we still have no idea yet on where December will end up. The previous warmest calendar year on record, 2014, still contained one month (August) that was below the older averages (1961-90). The second warmest year on record, 2006, still contained one month (March) that was below the older averages (1961-90), and also a February and an August that was at or in between the older 1961-90 and more modern (at the time) 1971-2000 averages. The third warmest year on record, 2020, still contained one month (July) that was below the older 1961-90 average, and also a September, October and December that were in between the older 1961-90, 1971-2000 and the more modern 1981-2010 average. The fourth warmest year on record, 2011, still contained a June, July and August that were all below the older 1961-90 average, and a January that was fractionally below the 1961-90 mean. It is now looking increasingly likely that this November will see a CET finish over 9*C, which will be another significantly above average month. The first ever 11*C+ yearly CET now looks highly likely unless December is very cold. That said although not coinciding with a calendar year, the rolling 12 month mean CET has gone above 11*C on a few occasions in the past, most notably during 2006/2007, and it also went over 11*C at some point in 2018/2019 and I think that it also possibly went over 11*C for some period over late 2013 to late 2014, and also for some period in late 1994 to late 1995. Something that I believe has never happened before, it is a truly remarkable statistic that 2022 has not had a single calendar month that was below even the most modern 1991-2020 average (which for most months are warmer than any older set of averages) let alone the older 1981-2010, 1971-2000 and 1961-90 averages, although as yet no-one can say how December will turn out. At this stage the models are hinting that the first week of December is likely to be anticyclonic and dry although probably not overly cold, but beyond that no-one can say yet. On top of no single below average month yet in 2022, we have actually not had a below average month (older 1961-90 average) since May 2021, and we have also not had a single month that was even below the most modern 1991-2020 average since August 2021. That said I think that November 2021 was possibly just fractionally under the 1991-2020 mean.

Even more grimmer in late January and in February, especially if there has been limited cold during the first half of the winter.

It just goes to show how difficult it is to get the UK cold - now and in the next few days we have a nice cold pool of sub -5 850s and even a little pool of -10 850s only a few hundred miles to the east of the UK - yet the UK still manages to avoid the cold getting to its shores - it just shows what rotten luck it is; how now with no Bartlett High and some higher pressure further north, the pattern still does not come together for the cold pool to our east to pay the UK a visit.

The main thing for sure is that there is not even a cold snap let alone cold spell on offer, and most of the country has not even had any frost yet, and there is little in the output to suggest that will change before the end of the month.

What I actually mean to say is that the link between September's weather and the following winter is clear that it appears to work one way with only warm Septembers, in that it is very rare if not unheard of to get a cold winter after a warm September, whereas after a cool or average September it is not clear that it has any bearing on the following winter either way. I think that you can write the coming winter off from a cold perspective after a warm September, but when Septembers are cool or average I do not think that there is a guarantee either way.

The persistent warmth of 2022 really began in September 2021, which was a very warm month with a notable late heatwave in the first half, and we have not had a below average month since May 2021, and we have in fact not even had a month that was even below the most recent averages but not the older averages, since August 2021. Although to be more precise, November 2021 at 7.3 CET was fractionally below the most recent 1991-2020 average of 7.4, but still above all the older averages.

I think that the warming trend in the October CET became evident from the 1960s onwards, and CET records show evidence that there was a warming trend in the November CET from the 1930s onwards. One remarkable statistic is that the month from mid October to mid November this year was the warmest since daily CET records began in 1772. Although we do not have daily records prior to 1772, this year has very likely seen the warmest mid-October to mid-November period since the CET record began in 1659. I calculate that mid-October to mid-November this year has a CET of 12.2. So although it is not a record warm CET for a calendar month, it has still been a record warm month spanning a mid-month to mid-month period, nevertheless.

The pattern is getting there towards a cold easterly setup - but not quite yet on that chart as the upper air is still not particularly cold at that forecasted point - but afterwards the pattern goes downhill.

Although we have not broken a record for an overall CET for a calendar month so far this year, the period from mid October to mid November this year almost certainly looks to be the warmest such monthly period since daily CET records began in 1772. Certainly the warmest CET monthly period from mid October to mid November on record must be one of the most notable month's warm spells of this year.

Although Jan and Feb in both 1988 and 1989 were not cold. It looks more impressive if you look at the period from 1977 to 1987; the number of cold Jans and Febs look a real treat, and both those months averaged at around 2.8*C for the CET for the 11 year period between 1977 and 1987! The Decembers had a prolonged spell of cold ones in the 1960s, as most of them were below average that decade, but then just over a decade and a half later, January and February were both generally very cold from 1977 to 1987.

I do not think that there is a significant correlation between November's weather and the following winter; sometimes we have had cold winters after mild Novembers (2009, 1984, 1978 though that Nov became cold late on), but sometimes no after mild Novembers (2015, 2011, 1997, 1994, as all winters were mostly mild following these mild Novembers).

You are correct that the 2010s were not renowned for cold Christmases or Decembers other than 2010, but on top of what you say above, late Dec 2014 was quite cold with snow for some places more so further north. Looking at the Novembers, 2016 was below average, as were 2013 and 2019 although with no seriously cold outbreaks.

The solar minimum after solar cycle 22 did coincide with a pretty cold winter in 1995-96, which had a number of Scandinavian Highs and easterlies coming and going through the winter and the occasional Greenland High. The end of solar cycle 20 around 1977 was followed by the severe winter of 1978-79, and winters 1976-77 and 1977-78 also had cold outbreaks at times. I think that the end of solar cycle 18 was in the first half of the 1950s and the winters in the mid 1950s brought a number of notable cold spells. Regarding what you say about solar cycle 21 (which was an odd numbered solar cycle) which ended in about 1985; I will also add in that prior to winter 1986-87, 84-85 and 85-86 also brought notable cold spells, and in actual fact mostly longer in duration than the one in Jan 1987. So odd numbered solar cycles are not the be all and end all looking at the above. One of my major worries is that in recent years with the most recent solar minimum, winter 2020-21 did have a fair amount of northern blocking for the most part, and it did try to be a cold winter, but low pressure was often too far into the south of the UK or to the west, and most of the cold outbreaks were focused on the north, with the winter being fairly average in the south, and in the grand scheme of things it wasn't a particularly cold winter; certainly nowhere near as cold as 2009-10. Looking at this point it looks likely to me that just after the most recent solar minimum, winter 2020-21 was the UK's chance of a cold winter, but the pattern did not quite favourably come together, and we could well have missed the boat for this time around just after the most recent solar minimum.

A remarkable statistic about last month, is that it saw the warmest second half to October on record since at least 1772. The second half of October 2022 had a CET of 13.6; the previous warmest second half of October on record was in 2014 with a CET of 13.2. 2005 now ranks third at 13.1.

I think that the polar stratospheric temperatures on average reach their highest levels in late June, up to around -30*C at the 10hpa level and around -40*C at the 30hpa level, and then on average cools rapidly through the autumn to averaging their coldest levels in late December at around -70*C at 10hpa level and -76*C at 30hpa level. What has happened over many recent years is that stratospheric temperatures have stayed at these cold levels through January and February and few SSWs or even gradual warmings have taken place at this time in recent years. For the first time during early November this year stratospheric temperatures have fallen just below average for this time of year. We on average would be about -70*C by the time we are into November with only gradual cooling on average until December. It is rare to see a SSW before late December in any year. What we would want to see now moving into the start of winter from a UK cold and northern blocking perspective, is the stratospheric temperatures stabilising at where they are now or a very slight warming would be the best.