Annual CET

[Roger J Smith]

If the NOV and DEC values add up to anything less than 11.0, the year will finish under 10.0. I believe that the metoffice use some weighted formula to get their annual average, not just a straight average of twelve monthly values (February is of course underweighted which makes the net effect of their approach a generally higher outcome by about 0.2). So those estimates are only based on the straight average. It would probably require a combination adding up to 10.7 to get the weighted average in below 10.0. With November heading for something like 7.5 that implies we would need a December of only 3.2 to join 2010 in the sub-10 club. 

[Midlands Ice Age]

On 23/11/2021 at 22:58, Roger J Smith said:

If the NOV and DEC values add up to anything less than 11.0, the year will finish under 10.0. I believe that the metoffice use some weighted formula to get their annual average, not just a straight average of twelve monthly values (February is of course underweighted which makes the net effect of their approach a generally higher outcome by about 0.2). So those estimates are only based on the straight average. It would probably require a combination adding up to 10.7 to get the weighted average in below 10.0. With November heading for something like 7.5 that implies we would need a December of only 3.2 to join 2010 in the sub-10 club. 

Interesting to see what happens here then in December...

According to the above we need something below 3.4C for December, in order to go below 10.0C...

Consensus suggests not but we will   see.

MIA

[Stu_London]

Just on the raw data - 3.7C would deliver a sum 10C year, but as Roger states there may be some differing methodology at play, so perhaps a tenth or so lower.

The first couple of weeks of December look cool to average, so some sort of cold spell in the second half would put this in the mix.

Seems like a million years ago when we got sub 9C (2009) - not sure the like of which I will see again in my lifetime.

 

[summer blizzard]

Rolling CET fell to 10.10C for the Dec-Nov period. Reasonable chance of dipping a little below as December to date looks relatively cool to average in modelling. 

[summer blizzard]

CET for 2021 was 10.26C, the coolest since 2013. 

[SqueakheartLW]

Looks like 2022's spell as the warmest start to a year based on the rolling CET has come to an end provisionally based on how yesterday's CET figure pulled 2022 down a bit.

I have displayed 2022 as well as the top and bottom 3 rolling CET values between 1878 and 2022 and also included all years from 2010 to 2021 for comparisons to recent years too.

1st January                                  2nd January                                 3rd January

Pos     Year     Rolling CET          Pos     Year     Rolling CET          Pos     Year     Rolling CET
1         2022    12.700                  1          2022    11.300                  1        2022    10.500
2         1916    10.500                  2          1916    10.000                  2        1917    9.700
3         1917    10.400                  3          1935    9.800                    3        1916    9.600
4         2012    9.800                    13        2012    8.400                    16      2012    7.867
14       2019    8.000                    19        2014    7.450                    26      2014    7.233
23       2014    7.500                    26        2015    7.200                    33      2013    6.867
24       2015    7.500                    47        2013    6.150                    42      2020    6.333
58       2018    5.400                    49        2018    6.100                    43      2018    6.267
61       2013    5.300                    50        2020    6.100                    49      2015    6.033
62       2017    5.200                    59        2016    5.250                    52      2016    5.700
63       2020    5.200                    67        2019    4.900                    78      2019    3.867
66       2011    4.900                    85        2017    3.600                    89      2017    3.300
72       2016    4.600                    92        2011    3.200                    108    2011    2.267
128     2021    -0.100                  126      2021    0.150                    122    2021    0.600
133     2010    -0.500                  129      2010    -0.100                   130    2010    -0.267
143     1979    -3.800                  143      1979    -3.250                   143    1893    -3.300
144     1887    -3.900                  144      1893    -3.350                   144    1979    -3.333
145     1962    -4.900                  145      1962    -4.450                   145    1962    -3.767          

4th January

Pos     Year    Rolling CET
1         1916    9.675
2         1948    9.375
3         1921    9.275
6         2022    8.825
14       2012    7.675
22       2013    7.375
31       2014    6.825
35       2018    6.675
39       2020    6.250
45       2016    5.975
63       2015    4.900
83       2017    3.700
92       2019    3.125
105     2011    2.275
118     2021    1.075
137     2010    -1.050
143     1962    -2.800
144     1979    -3.125
145     1893    -3.600         

After the dizzy heights that 2022 produced in the first 3 days the rather chilly provisional value for 4th January has pulled 2022 down off top spot to 6th position overall. Still a very mild start but with the predicted chilly weather to come for the rest of the week I can see 2022 heading well down and maybe to around mid table by the end of the week as the colder days cancel out the very mild ones that started the year off.

Considering how cold January 1963 was it is strange to see here that 1963 doesn't even feature in the bottom 3 in terms of rolling CET for the 1st to 4th January. It would seem 1962 and 1979 had the iciest starts to any year on the CET record between 1878 and 2022.

[summer blizzard]

Rolling CET rose to 10.34C for the Feb-Jan period. 

Edited February 12, 2022 by summer blizzard

[summer blizzard]

Rolling CET rose to 10.48C for the March-Feb period. 

[summer blizzard]

Rolling CET rose to 10.54C for the April-March period.

[summer blizzard]

Rolling CET rose to 10.76C for the May-April period.

[LetItSnow!]

18 minutes ago, summer blizzard said:

Rolling CET rose to 10.76C for the May-April period.

Once last years cold May get's out of the way I the rolling average will shoot up. We're a long way off the record holders of 1994-1995 and 2006-2007 but if warmer than average conditions persist, it's not unlikely we'd break into 11C. Can't be many years that have achieved that, have their?

[SqueakheartLW]

Haven't done one of these rolling CET assessments for quite some time so thought I would do one just to see how 2022 is shaping up compared with all other years between 1878 and 2022.

These values are for the period Jan 1st to May 8th.

Considering how mild I thought the first part of this year has been I was quite surprised to see we were only in 53rd position with a rolling CET value of 7.494C

Some significant other years we are in rough comparison with are 1976 which we are very close to and similar to 1976 we have had generally drier than average conditions too. Is this an omen for the summer? We also had a La Nina in the 1975/76 winter preceding this hot summer but at present the shift to an El Nino which happened during 1976 doesn't look as though it is going to happen which is the thing going against a 1976 summer.

Also of note is how we are currently trending slightly cooler than 2009 as well despite the colder winter 2008/09 which gave 2009 a colder start than this year.

Now to focus more on the top 3, bottom 3 and the most recent years from 2010 onwards:

Here we can see how 2022 is in fact trending cooler than most recent years in fact and many of the 2010's and early 2020's years were warmer than this year. We are in fact closest to 2015 so far this year below us and 2012 above. Using this metric for recent years doesn't bode well for summer as 2015 was nothing special and 2012 was a washout.

In terms of ENSO 2015 is no comparison to this year with the developing Super El Nino but then again 2012 isn't really a close match either but is much closer than 2015 is.

Rather interesting to me is how despite how cold 2010 started out at it is in fact 2013 that is the coldest recent year, probably thanks to that cold March in 2013.

2007 is running top of all years at this stage but this doesn't take into account what happened from May onwards that year. At present we don't look to be on course to smash the record of 2014 of 10.95C since we are already 1.3C cooler than 2014 at this stage but since there's a large part of the year to go we can't rule it out yet.

[LetItSnow!]

22 hours ago, SqueakheartLW said:

Haven't done one of these rolling CET assessments for quite some time so thought I would do one just to see how 2022 is shaping up compared with all other years between 1878 and 2022.

These values are for the period Jan 1st to May 8th.

Considering how mild I thought the first part of this year has been I was quite surprised to see we were only in 53rd position with a rolling CET value of 7.494C

Some significant other years we are in rough comparison with are 1976 which we are very close to and similar to 1976 we have had generally drier than average conditions too. Is this an omen for the summer? We also had a La Nina in the 1975/76 winter preceding this hot summer but at present the shift to an El Nino which happened during 1976 doesn't look as though it is going to happen which is the thing going against a 1976 summer.

Also of note is how we are currently trending slightly cooler than 2009 as well despite the colder winter 2008/09 which gave 2009 a colder start than this year.

Now to focus more on the top 3, bottom 3 and the most recent years from 2010 onwards:

Here we can see how 2022 is in fact trending cooler than most recent years in fact and many of the 2010's and early 2020's years were warmer than this year. We are in fact closest to 2015 so far this year below us and 2012 above. Using this metric for recent years doesn't bode well for summer as 2015 was nothing special and 2012 was a washout.

In terms of ENSO 2015 is no comparison to this year with the developing Super El Nino but then again 2012 isn't really a close match either but is much closer than 2015 is.

Rather interesting to me is how despite how cold 2010 started out at it is in fact 2013 that is the coldest recent year, probably thanks to that cold March in 2013.

2007 is running top of all years at this stage but this doesn't take into account what happened from May onwards that year. At present we don't look to be on course to smash the record of 2014 of 10.95C since we are already 1.3C cooler than 2014 at this stage but since there's a large part of the year to go we can't rule it out yet.

The final comment reminds me of the first 5 months of 2006 which were unexceptional and quite cold at times even, yet went on to become the hottest year on record at time time. Similarly, 1993 was very mild/warm until June but then the second-half of the year turned very cool - to a lesser extent this happened in 2017 too. It goes to show you never really know.

At first I was very surprised to find out we're only 53rd place in terms of warmth, but then I remember January and April were unexceptional and actually around the 1991-2020 average.

Also out of interest If the same anomalies against 91-20 were imposed onto the 1772-present day average, 2022 would so far look like this instead

3.4  5.8  6.8  8.2

and 2021 would look like this

1.9   4.3   6.1   5.5   9.5  15.1  16.8  15.0  15.1  11.1   6.0   5.7  - 9.34

perhaps what the same synoptics would have brought 100+ years ago perhaps.

[SqueakheartLW]

On 10/05/2022 at 18:34, LetItSnow! said:

The final comment reminds me of the first 5 months of 2006 which were unexceptional and quite cold at times even, yet went on to become the hottest year on record at time time. Similarly, 1993 was very mild/warm until June but then the second-half of the year turned very cool - to a lesser extent this happened in 2017 too. It goes to show you never really know.

At first I was very surprised to find out we're only 53rd place in terms of warmth, but then I remember January and April were unexceptional and actually around the 1991-2020 average.

Also out of interest If the same anomalies against 91-20 were imposed onto the 1772-present day average, 2022 would so far look like this instead

3.4  5.8  6.8  8.2

and 2021 would look like this

1.9   4.3   6.1   5.5   9.5  15.1  16.8  15.0  15.1  11.1   6.0   5.7  - 9.34

perhaps what the same synoptics would have brought 100+ years ago perhaps.

I too was surprised to see how low we were in the list but when deciding to do my latest update I then realised why we were down in 53rd position and this was a stupid mistake on my part.

I compared May 8th's 2022 rolling CET mean with June 8th years instead of May 8th years.

The correct figures for May 8th 2022 compared with other recent years from 2010 to present and the actual rolling position is displayed below

In allowing for comparisons to the correct rolling means we actually were in a high 5th position of 145 years with only 2007, 1990, 2020 and 2014 having warmer rolling figures up to 8th May and we were in fact just 0.577C below 2007, not 1.475C below as originally stated.

Latest rolling 2022 CET mean compared with other years in the data from 1878 to present

Yes, and this time I have compared it to the correct date rather than a month out from where it should have been.

How does the latest rolling CET mean up to 4th June 2022 of 8.507C compare with the other years between 1878 and 2021. I have only shown recent years but the positions are where they stood compared to all the years on the record.

As can be seen here we are still in a high 5th position with as before 2007. 1990, 2020 and 2014 the only years warmer up to 4th June compared with 2022.

The main thing of note here is how we have closed the gap down to the warmest year 2007 from 0.577C to 0.326C and we are running very close to 2014 still.

I maybe incorrectly stated 2014's annual CET record of 10.95C looked safe but that was before I realised I had compared the wrong data and based on the correction 2014 looks under threat again this year and back in 2020 it also looked under threat at this stage but that year fell back slightly and came up just short.

If we go on to have a warmer than average summer and autumn too then we could be on course for our first 11C CET year on the record.

[LetItSnow!]

15 hours ago, SqueakheartLW said:

I too was surprised to see how low we were in the list but when deciding to do my latest update I then realised why we were down in 53rd position and this was a stupid mistake on my part.

I compared May 8th's 2022 rolling CET mean with June 8th years instead of May 8th years.

The correct figures for May 8th 2022 compared with other recent years from 2010 to present and the actual rolling position is displayed below

In allowing for comparisons to the correct rolling means we actually were in a high 5th position of 145 years with only 2007, 1990, 2020 and 2014 having warmer rolling figures up to 8th May and we were in fact just 0.577C below 2007, not 1.475C below as originally stated.

Latest rolling 2022 CET mean compared with other years in the data from 1878 to present

Yes, and this time I have compared it to the correct date rather than a month out from where it should have been.

How does the latest rolling CET mean up to 4th June 2022 of 8.507C compare with the other years between 1878 and 2021. I have only shown recent years but the positions are where they stood compared to all the years on the record.

As can be seen here we are still in a high 5th position with as before 2007. 1990, 2020 and 2014 the only years warmer up to 4th June compared with 2022.

The main thing of note here is how we have closed the gap down to the warmest year 2007 from 0.577C to 0.326C and we are running very close to 2014 still.

I maybe incorrectly stated 2014's annual CET record of 10.95C looked safe but that was before I realised I had compared the wrong data and based on the correction 2014 looks under threat again this year and back in 2020 it also looked under threat at this stage but that year fell back slightly and came up just short.

If we go on to have a warmer than average summer and autumn too then we could be on course for our first 11C CET year on the record.

It all depends on where we go from here. As stated, in many recent years we get a run of mild months in the first half but it collapses in the second-half. I wouldn't be surprised to see it happening again. Interesting how last year was the opposite of that.

[Don]

10 hours ago, LetItSnow! said:

It all depends on where we go from here. As stated, in many recent years we get a run of mild months in the first half but it collapses in the second-half. I wouldn't be surprised to see it happening again. Interesting how last year was the opposite of that.

I think a classic year where the first half was cold and the second half warm was 2013.  January to June was significantly below average, but that ended abruptly early July, heralding a much warmer second half to the year!

[summer blizzard]

Rolling CET increased to 11.0C (V2.0). 

July, September, October and December do present high bars so we should begin falling back in all likelyhood.  

[SqueakheartLW]

On 04/07/2022 at 16:20, summer blizzard said:

Rolling CET increased to 11.0C (V2.0). 

July, September, October and December do present high bars so we should begin falling back in all likelyhood.  

Remember August 2014 is a very low bar to beat so all we need to do to beat 2014 is come fairly close in July, September, October and December.

It wouldn't take very much to beat August 2014's CET value

I still think unless we see things cooling down somewhat that 2022 could be the year that breaks the 11C CET

[Roger J Smith]

I see that in the v2.0 annual means, only one decimal place is shown now instead of two, but they do rankings based on second decimals (this applies to monthly ranks as well). This is why if you look at the tables, tied values are no longer listed chronologically but they are separated by that unseen second decimal (pre-daily 1772 to present means appear to be treated in some random fashion but at .x0 within their grouping, so for example if there is a bunch of Augusts tied at 16.0, the pre-daily ones are sorted at random as 16.00 and the rest are sorted as whatever they are that rounds off to 16.0. 

As to annual means being sorted, that appears to be by second decimal also, otherwise you would not see 2011 ahead of 2018 and 1949 ahead of 2017, 1999, 1990 and 2002 in that order (in the old format, these ties would have appeared as 2011 in 5th then 2018 in 4th, and 1949 in 10th then 1990 9th, 1999 8th, 2002 7th and 2017 6th). The legacy values have been adjusted at least by second decimal so you can't really go over there to check the second decimal situation, it might be different now. I don't know if the second decimal is derived from a straight average of twelve months or a weighted average by number of days or even just an annual average of daily means. 

So in v2.0, the top ten are listed as follows: 

 1. 2014 (11.0)

 2. 2006 (10.9)

 3. 2020 (10.8)

 4. 2011 (10.7)

 5. 2018 (10.7)

 6. 1949 (10.6)

 7. 2017 (10.6)

 8. 1999 (10.6)

 9. 1990 (10.6)

10. 2002 (10.6)

Edited July 5, 2022 by Roger J Smith

[LetItSnow!]

20 hours ago, Roger J Smith said:

I see that in the v2.0 annual means, only one decimal place is shown now instead of two, but they do rankings based on second decimals (this applies to monthly ranks as well). This is why if you look at the tables, tied values are no longer listed chronologically but they are separated by that unseen second decimal (pre-daily 1772 to present means appear to be treated in some random fashion but at .x0 within their grouping, so for example if there is a bunch of Augusts tied at 16.0, the pre-daily ones are sorted at random as 16.00 and the rest are sorted as whatever they are that rounds off to 16.0. 

As to annual means being sorted, that appears to be by second decimal also, otherwise you would not see 2011 ahead of 2018 and 1949 ahead of 2017, 1999, 1990 and 2002 in that order (in the old format, these ties would have appeared as 2011 in 5th then 2018 in 4th, and 1949 in 10th then 1990 9th, 1999 8th, 2002 7th and 2017 6th). The legacy values have been adjusted at least by second decimal so you can't really go over there to check the second decimal situation, it might be different now. I don't know if the second decimal is derived from a straight average of twelve months or a weighted average by number of days or even just an annual average of daily means. 

So in v2.0, the top ten are listed as follows: 

 1. 2014 (11.0)

 2. 2006 (10.9)

 3. 2020 (10.8)

 4. 2011 (10.7)

 5. 2018 (10.7)

 6. 1949 (10.6)

 7. 2017 (10.6)

 8. 1999 (10.6)

 9. 1990 (10.6)

10. 2002 (10.6)

Still so impressive that 1949 still ranks sixth. Imagining the same synoptics of that year in today’s climate and thinking how warm it would be now.

[Don]

2 hours ago, LetItSnow! said:

Still so impressive that 1949 still ranks sixth. Imagining the same synoptics of that year in today’s climate and thinking how warm it would be now.

Given the world has warmed around 1C since 1949, potentially 11.6C?

[Roger J Smith]

I wouldn't call it a debate so much as a discussion, but if it's true to say there is a general consensus that the climate has warmed by almost 1 C deg from the mid-20th century,  then there are two different scenarios in play, or a combination of the two: 

(a) warming occurs because all air masses warm up, or

(b) warming occurs because the storm track shifts north and we spend more time in the warmer air masses

My own research shows that it is probably about 35% (a) and 65% (b). Thus if 1949 was one of those years when storm tracks ran further north than average, perhaps some of the warming should not be added to it, but its already warm air masses should be assessed as slightly warmer now. My own guess is that 1949 would check in around 11.0 now. But it is not within the realm of an experiment that can be duplicated in laboratory conditions. 

Probably the main reason why temperatures have warmed (aside from changes in atmospheric chemistry) is that oceans have warmed. That tends to warm up the surface layers of air masses moving west to east over Britain. Even if these air masses were to enter the Atlantic basin from North American source regions at exactly their "original" temperatures (which is shown to be false, those have warmed also), they would warm at least through the boundary layer to about 1000 metres by contact with the warmer ocean.

The fact that all this is subject to monthly scale variability means that we can still see just about the same synoptics as back in the colder climate, as shown in Nov-Dec 2010 and Feb-Mar 2018 but there is pressure on any such reversion to that form to snap back to milder conditions. Feb-Mar 2018 showed that pressure I would say more obviously than Dec 2010 did. Also April 2021 showed that given the right synoptics you can still sustain a very significant negative anomaly. But the odds against it are slowly growing partly as oceans warm further. There may come a time however when the warming of oceans will reach a limit imposed by mixing in of additional arctic meltwater which may tend to stabilize or even partially reverse trends. I am not personally all that convinced that we face runaway warming, although the human contribution factor is hard to model given conflicting political trends, also the natural variability factors are hard to assess. The IPCC have been saying for several years now that the current warming is against the background of a climate that in natural terms should be cooling. That has always seemed a bit off to me, but the problem is, the only evidence that the background climate should be warming anyway would be in air mass frequency; if you then postulate that climate change is having an effect on air mass frequency, you have no way of knowing what that "should" be in a zero-human modification control experiment. 

Also there is the matter of the Pacific oscillations and clearly the northern hemisphere tends to warm faster when there are strong El Nino events; the cause and effect between human atmospheric modification and the El Nino is an area that I don't even think has attracted any research, if so, the results of it have not yet been made known. I have thought about it casually just at the stage of "what scientific research paradigms should be adopted to study this" and not gotten very far with it. When you think about what causes the El Nino, linking that to climate change is difficult. If the Pacific oscillations are unrelated to human activity, then they should continue to play a role in regulating temperatures. But then there might be modifications of their signatures from complicated feedback processes. My own guess is that human political trends will operate faster than research and the atmosphere will gradually be back to how it is supposed to be within two centuries, then natural cooling trends will set in and people will clamor for a return to high greenhouse gas emissions. You'll have people marching through the (icy) streets of European cities demanding that people drive gasoline guzzling cars (and personal gyrocopters because those will come sooner or later), and fire up the old coal fired electricity plants before the ice age returns. (irony alert)

 

[SqueakheartLW]

Looks at present like we are well on course to be warmest year on the CET mean record between 1878 and the present day if current trends continue.

Here's the eventual top 10 years based on legacy CET data

31st December

Position     Year          Annual CET Mean
1                 2014         10.948
2                 2006         10.863
3                 2020         10.752
4                 2011         10.715
5                 2018         10.679
6                 1999         10.660
7                 1990         10.655
8                 1949         10.642
9                 2002         10.621
10               2017         10.584

Here's how these same years are shaping up at the current point in the year (5th August is latest CET provisional value) compared with this year's provisional rolling annual CET mean value too. Also included are what the August CET's were for these years and the current provisional August 2022 CET to 5th August.

Note the position of these years here is relative to all years between 1878 and 2022, not just relative to the 10 years featured.

5th August

Position     Year          5th August CET Mean          August CET
1?                2022         10.920 (Prov)                        18.520 (Prov to 5th)

1                  2014         10.877                                   (10th) 14.955
2                  1990         10.757                                   (1st)    17.990
3                  2017         10.686                                   (8th)   15.613
5                  2020         10.628                                   (2nd)  17.565
6                  1999         10.595                                   (6th)   16.148
8                  2018         10.487                                   (5th)   16.639
11                2002         10.407                                   (3rd)   16.952
21                1949         10.200                                   (4th)   16.800
22                2011         10.192                                   (9th)   15.397
23                2006         10.138                                   (7th)   16.065

As you can see 2022 is currently provisionally at 10.920 which is the warmest on the CET record between 1878 and 2022 and we currently have a provisional 18.520 CET August which if this was maintained would easily be clear of all 10 years featured and would easily keep 2022 on track to beat all other years on the CET record.

31st August projections

Now what happened with the eventual top 10 years versus what could happen with August 2022 and where would that leave us this year. Here's the actual data and the 2022 projection for 31st August too.

Position     Year          End Of August Rolling Annual CET Means
1?                2022         11.733 (Projected if Aug 2022 CET is 18.520)
1?                2022         11.459 (If Aug 2022 CET is 16.371)

1                 1990          11.458 (Up 1 position relative to 5th August)
2                 2020          11.396 (Up 2 positions relative to 5th August)
4                 2014          11.275 (Down 3 positions relative to 5th August)
6                 2017          11.205 (Down 3 positions relative to 5th August)
10               2002          11.118 (Up 1 position relative to 5th August)
11               1999          11.109 (Down 5 positions relative to 5th August)
12               2018          11.088 (Down 4 positions relative to 5th August)
17               1949          10.923 (Up 4 positions relative to 5th August)
26               2006          10.755 (Down 3 positions relative to 5th August)
32               2011          10.679 (Down 10 positions relative to 5th August)

If 2022 maintains an August CET of 18.520 then we end up with a provisional rolling annual CET of 11.733 which would put us 0.274 ahead of 1990 at the end of August and well on track for warmest year on record. We are 0.458 ahead of eventual warmest year 2014 on this projection which would see us get an annual CET mean of 11.406 if we match 2014 for the rest of the year for September to December CET means.

To maintain warmest spot by the end of August 2022 we only need to get an August 2022 CET of 16.371 which would make the rolling mean 11.459 which would be just 0.001 ahead of 1990.

[LetItSnow!]

I think we're almost guaranteed a top 7 warmest year on record, the only question is whether these warm anomalies will continue or fade away. 2020 stayed very warm right until August (bar July) then faded as we got into the autumn - still warm enough to rank third warmest, but just stopped it from being a record breaker. We need to really continue to see these warm anomalies persist to record anything higher than about 10.8, which I still think is not a done deal. This persistent warmth has now been in place since September last year, and a part of me is inclined to say it may break soon - but it's also more likely than ever this year to beat 2014. It's sure felt more notable than 2014, that year very much was persistently above average instead of ever overly exceptional - indeed the highest temperature of 2014 was a very tame 32degC unlike this years 40degC.

[Earthshine]

15 minutes ago, LetItSnow! said:

I think we're almost guaranteed a top 7 warmest year on record, the only question is whether these warm anomalies will continue or fade away. 2020 stayed very warm right until August (bar July) then faded as we got into the autumn - still warm enough to rank third warmest, but just stopped it from being a record breaker. We need to really continue to see these warm anomalies persist to record anything higher than about 10.8, which I still think is not a done deal. This persistent warmth has now been in place since September last year, and a part of me is inclined to say it may break soon - but it's also more likely than ever this year to beat 2014. It's sure felt more notable than 2014, that year very much was persistently above average instead of ever overly exceptional - indeed the highest temperature of 2014 was a very tame 32degC unlike this years 40degC.

If so much energy stored in the Mediterranean (SST in some places 30°C+) would help keep continental Europe warm well into autumn.  How much influence we get is anyone's guess, however autumn 2006 stayed warm after that exceptional summer.