North-Easterly Blast

I have looked back in the records and, in recent decades anyway, the warmest September that I have found that has been followed by a decent winter for cold was 1985, whose September had a CET of 14.6. My conclusion is that when September is cool or even close to average there does not appear to be a pattern on the type of winter that follows. There certainly appears to be a strong link that if September is significantly warm (say more than 1*C above average) it appears that it is very rare to get a cold winter to follow it, and the theory has certainly been a strong one in recent times at least. That said, it appears that it is still possible to have a September that is "slightly above average," rather than significantly warm, like in 2009 (CET 14.2), (1978 was also 14.2) and get a cold winter to follow, and we just scraped in with 1985's 14.6. (It would be difficult to consider a September really warm to test the theory, if it is only around half a degree above average). On the other hand I think that really warm Septembers (CETs 15*C and above), or possibly high 14s or more, are really bad news if you want a decent winter to follow, and recent times have clearly shown this, and even recent winters with other good background signals have still not brought much in the way of proper cold patterns after really warm Septembers. The September averages are as follows (1961-90 13.6; 1971-2000 13.7; 1981-2010 14.0; 1991-2020 14.2). Based on this I would say that a September CET from the high 14s upwards could be considered warm. On top of this, September 2021s CET of 15.9 was the seventh warmest on record back to 1659. So I would say that when you have a September that is in the top ten warmest on record then this winter was certainly written off last September. I think the warm September + mild winter theory, or at least not favouring a colder winter to follow, is as strong a link in the UKs weather patterns, as mild Februarys not favouring a cold March to follow. I was also doing some research on that recently, and found that notably mild Februarys (above 6*C CET) in recent times at least, have nearly always followed by mild Marches (over 7.5 CET); in recent times only 1995 was followed by a colder March, and in 2011 and 2020 the March CET was not in the 7.5+ category. That said the mild February = mild March theory clearly does not always work the other way round (1991, 1994, 1981 being examples). On the other side of the September theory, it appears that it is not clear that there is any trend in the type of winter that follows. While nothing in the British weather patterns appears to guarantee cold weather for the UK, my conclusion is that really warm Septembers do not favour colder conditions during the following winter; in the same way really mild Februarys do not favour colder weather in the March to follow.

I think that this winter was already dead in the water last September, as it is a well known fact of the UK's weather patterns that it is very rare to get a cold winter after a warm September. Anomalous warmth in September in the UK is a killer pattern for the winter that is to follow. In any year when September is really warm I think that you can write the following winter off from being anything special in the September.

I think that January 2006 saw another anticyclonic pattern that never really came together and set up favourably to bring a significant cold spell to the UK; in fact winter 2005-06 was a very average winter for temperatures with a fair amount of blocking but never favourably set up for a really cold spell to manifest itself in the UK; I believe that there was a bit of inversion cold late / Jan early Feb that year, and we finally late on that winter, did get a much colder pattern setting up in late February into March; winter 2005-06 was very much a failed opportunity for a cold winter; there was a fair amount of cold air near the UK that year and a fair amount of blocking, but disappointingly the blocking never really set up favourably to allow the really cold air to make it to the UK, so synoptic-wise (when you would look at the charts and think there would have been a decent chance of a cold winter) it was a disappointing winter from a cold perspective, in a blocked pattern never coming together, even though it was not a mild winter overall. Winter 2005-06 could be another example of the warm September theory ruining the cold prospects for winter as September 2005 was pretty warm. Unless a better pattern develops in February I think winter 2021-22 could well be looked upon as another missed opportunity where the pattern just failed to come together.

I have found this winter so far not particularly of the mild zonal type, certainly nothing like 2019-20 and 2013-14 etc, and the only notable mild spell this winter was at the turn of the year; the rest of the time is has just basically been "nothing special", and disappointing in the sense that there has been a fair amount of high pressure around, but never in a favourable position to get proper cold to the UK, and charts have at times looked half way or getting towards being favourable for a proper cold spell for the UK but it has never come together, and we have often been left with temperatures been stuck at about average, with hardly any real cold. I am just wondering if, judging by the pattern we have had so far, has been more like a "failed cold winter" for the UK.

You are right. This winter was nailed against being anything special back in last September with its anomalous warmth, as there appears to be a strong link in anomalously warm patterns in September significantly reducing the chance of cold patterns developing in the following winter, and in any future year, my view is that a cold or at least colder winter will always be very unlikely after a warm September. I am now wondering if winters like 2017-18 and 2020-21 may now be regarded as "cold" or even "severe" in the UK, and that winters like 2009-10 are dead in the water along with the sort of cold spells of Dec 2010 and what we had in the 2012-13 winter, and if spells like these will ever be seen again.

Do you take into account how foggy a winter is when deciding its overall score, as well as snow and frost? As fog is the white stuff that fills the air and makes the air white away from surfaces, it would be great if all the three "white stuffs"; frost, snow and fog, are factored into the winter index, along with average daytime maximum temperatures. As fog, at least away from the coast and high ground inland, has traditionally mostly only happened in the colder winter months, it would be great if how foggy a winter is counts for the winter index figure as well as snow and frost. I am surprised that winter 2019-20 has a score above some of the other poor ones, as colder weather even in the form of short snaps was in very short supply that winter. What is the score for winter 2021-22 so far by the way?

Winter 2020-21 was roughly overall average in terms of temperatures - although you are right that late December to mid February overall was cold, in actual fact predominantly so, but without any real major severe spells. 2017-18 was another recent winter that was roughly overall average - it did have a cold spell in the first half of December, a bit of cold zonality in January, and a fairly cold February, with a week's very cold spell late on lasting into March. That said we have not had a winter that could reasonably be described as a cold one, with a number of significant cold spells since 2012-13. Winter 2021-22 so far just stands out as a winter where the pattern never came together to bring serious cold to the UK, but apart from the exceptionally mild spell over New Year, it has not been notable for mild zonal weather either.

To get significant cold into the UK from that, the low to the north of Norway would need to move south-east, the high to the east of us would need to link up with the Russian high, and the PV in the NW Atlantic would need a breakaway low to dive south-east into Europe; it is half way there but still a long way to go.

In the end we will maybe not be far from average in terms of mean temperature for this month apart from the first three days, but I would not call a pattern of anticyclonic nothingness for possibly two thirds of the month an average weather pattern for January - little in the way of any really cold days, and little fog compared to what you would expect with a calm pattern, and any "proper cold synoptics" reserved for about three or four days in the first week. This month is very likely to be a month where its overall CET does not tell you a great deal about the weather that the country experienced during the month.

Metcheck have their CET tracker at around 4.25*C up to today. I acknowledge that this is often lower than Hadley's final confirmed figure, but sometimes it is similar. Net weather's tracker figure is often a bit higher than Hadley's final confirmed figure.

If that GEFS mean verified exactly as shown a week today, I would say that there is a good chance that higher ground (the Pennines) would see snow falling, but for lower levels (under 1,000 feet a.s.l) I would say falling snow would be unlikely.

Yes you are right. I think in the last two decades, Aprils 2008, 2012, 2016, 2018 and 2021 saw snow fall over some part of the country or another at some point in the month.

It has also been a virtually fogless winter for most places. This high pressure has failed to deliver any widespread fog, like many anticyclones would have done years ago. It has just been the whole pattern that has failed to come together to deliver any significant cold, and there has not, and looks unlikely out to the end of January, that there has not been a cold spell to speak of, despite high pressure around much of the time, and many areas have seen no fog, and snowfall has been limited to the few areas that caught some in the short lived northerly early this month, and some places that had some in late November before winter started. Last year's warm September has destroyed cold prospects for this winter, and this winter was already dead in the water last September after its anomalous warmth, and the warm September = poor winter theory has just followed suit again.

Yes we have had a fair amount of blocking for most of the winter so far, but it has just completely failed to set up favourably to bring a significant cold spell to the UK. This January HP spell has just been complete nothingness (just as though the UK's weather has gone on holiday), never building to a place where any truly cold air can reach the UK, no inversion cold days, little if any fog for most places either, just nothing at all. The only worse pattern than the HP pattern of this month that in my view I can think of is something like 2013-14, and to a lesser extent 2019-20, nothing but raining all the time. At least this month's HP pattern has not been as bad as the pattern of those two recent winters, but that is about all it is an improvement upon some recent winters.

We have not had a February that cold in 36 years. To see something that cold we would need to see a significant cold outbreak lasting for most of the month. We have actually not had a really cold February since 1991 (1.5 CET). The coldest February of recent times was in 2010 at 2.8*C, although more recently Feb 2018 was close to this at 2.9*C.

That is what I thought. In fact the whole winter so far (apart from the exceptional mild spell at the end of December / start of January) has not felt all that mild; it has just been so disappointing that the pattern has simply never come together from a significant cold spell perspective, and even this HP spell has never lived up to potential in years gone by in bringing widespread fog and proper cold days, and neither did the HP spell in the week or so before Christmas, and the northerly early this month turned out to only be a toppler, and the patterns have at times become "half way" to delivering a proper cold spell for the UK but just not come together to get the UK properly cold.

That is just what I thought. In fact I think that the whole of this winter so far has not felt all that mild apart from the five day exceptionally mild spell at the end of December into the start of January, other than that it has just been stuck at average or just above the rest of the time. I think that the main disappointing part of this winter is that the pattern has just never come together to bring a significant cold spell to the UK, both just before Christmas and during this month with the blocking setting up totally unfavourably. I cannot understand why it happens; the HLB just before Christmas failed to materialize into a cold outbreak from the north or east, and the anticyclonic spell this month has just set up totally unfavourably to allow a cold spell to materialize from the north or even to set up favourably to allow significant fog or proper cold days. We have not had a large amount of tropical maritime zonality so far this winter, and a fair amount of high pressure around, but it simply has not come together to bring anything special for the UK. Unless things change in February then this winter will basically just go down as almost a winter of nothingness when the pattern just failed to come together.

It is just so sickening that when we have an easterly QBO in place and an eastern based not so strong La Nina that we had a really warm September that tested the theory again and really reduced the chance of cold weather this winter, and we are just left with a pattern that just simply does not come together to set up a significant cold spell for the UK - and that has been the case almost right from when winter started. This winter was already dead in the water back in September following the anomalous warmth that month. If you want significant winter cold in the UK then you do not want a really warm September or otherwise you can write the coming winter off then.

Has anyone any idea why we do not get cold and fog in the winter anymore or very rarely? This high pressure spell has been so disappointing in terms of its delivery, with hardly any fog. To get cold and fog under winter anticyclones nowadays what do you need?

I think that the La Nina in 1998-99 was stronger than this year's and was more central based. As you say, there was a northerly spell in the second week of that February that brought a much colder week in a largely mild winter.

That is just what I was thinking - it has just felt average all the time with a few frosts - and very disappointing as regards fog and low maxima on a large scale at least. The only worse weather than this that I can think of is 2013-14 when it was just 6-9*C and raining all the time. Why does the British winter today, especially in more calm HP periods, fail to deliver fog? I know that Atlantic driven / windy weather is not the right weather conditions for fog, but many years ago calm anticyclonic periods often brought widespread fog. This HP has just been about the worst I can think of for its delivery not being up to its potential for wintriness, in the sense of significant widespread fog and inverted cold based low maxima.

I do not know how relevant that would be 360 years later. Certainly in recent times if not recent decades it certainly has existed that there is a link between warm Septembers and milder winters to follow, and it is very rare to get a cold winter after a really warm September. I think that the background signals of an easterly QBO, a moderate eastern based La Nina, and shortly after a solar minimum were good background signals for a significant cold spell to manifest itself this winter, but I really think that coming after a very warm September that chance was significantly reduced, making the point all the more marked that warm Septembers are winter killer patterns for the winter to follow.

September 2009 wasn't extraordinarily warm, it was not that far above average (14.2), compared to 13.6 (61-90) and 13.7 (71-00). Another example of something similar to this was 1978, also 14.2 CET. I actually think that although there is not much of a correlation, it appears that it is still possible to get a cold winter after Septembers that are just above average (low 14s), but significantly above average Septembers appear to be a winter killer. For comparison, the warmest September that I can think of that has been followed by a good cold winter was 1985, which was 14.6 CET, at or approaching 1*C above average. I cannot think of a September warmer than 1985s 14.6 that has led to a good cold winter. September 1981 also had 14.5 and led to the December / January freeze that winter. I think that when September is about the mid 14s CET, then I can think of a few good winters that have occurred after them, but Septembers warmer than that, I cannot. I think if we had not had last year's really warm September, I think that there would have been a good chance that this winter would have seen the pattern come together for a significant cold spell to reach the UK. Warm Septembers really are winter killer patterns. It appears to be a well known fact of the British weather patterns that it is very rare to get a cold winter to follow a really warm September, and they are always more likely to be followed by milder conditions during the winter. A cool or average September is certainly better news with respect to colder winter weather to follow, but there is by no means any correlation that way round as there is with the warm September being a winter killer pattern. I find it weird that the September trend link appears to only be evident at one end - after a warm September.

Out of that list I can only think of the Jan 2013, Jan 2009, Dec 1981 and Jan 1985 SSWs that brought a significant cold spell to the UK shortly after, and possibly Feb 2001 did have some HLB and a northerly spell late in the month. I can also think of mid Feb 2018 that is not on the list. There was one in Feb 1979 but the big freeze that year had been going on since the end of Dec 1978, so it was not a factor in bringing the freeze up that year. There is one listed in Feb 2010 but there had been significant cold before then in that year.

The warm September theory being controversial, well I will say that although it is clear that previous records suggest, in recent times at least, that there is a link that a warm September certainly does not bode well from a perspective of colder winter weather in the following winter, and a warm September is more likely to be followed by milder conditions during the winter, but on the other hand, it is not clear that a cool September or even a close to average one has any bearing on the sort of winter that follows. I can only see evidence that the September link works one way with the warm September ruining the chance of cold weather in the following winter, but the other way round nothing is clear. In reality a cool or average September is only a weak factor in favour of a cold winter to follow, but the other way, a warm September is a strong factor of a winter killer for cold weather in the following winter. Certainly if you want a reasonable chance of cold outbreaks in the following winter you do not want to see anomalous warmth during September, and I would say that you can almost write the following winter off from being decent in September if it is very warm in that month.