Annual CET

[Weather-history]

Should January be on the list? I though that was just 0.3C below.

You are right, it was only 0.3C below the average.

[SomeLikeItHot]

With April being the 4th month in a row with CET below the 61-90 mean I thought it might be interesting to see what each months correlation was with the following 12 month period, so I constructed the correlations between the Mean cet of each month and the Mean CET of the following 12 months. Below its the table. 

 

 

 JAN   FEB     MAR  APR    MAY    JUN     JUL    AUG    SEP  OCT   NOV    DEC 

0.30   0.27   0.31    0.28   0.26   0.14   0.19   0.24    0.25   0.22    0.21   0.23

 

 

And the thing to note is that while there is not a huge range in the values, the 4 highest values on the table are Jan - April. Months that this year are all below average. May is the next highest correlation, lets see how that turns out.

Edited April 25, 2013 by SomeLikeItHot

[Weather-history]

It looks likely that we are going to record 4 consecutive below average seasons. Summer 2012, autumn 2012 and winter 2012-13 were all below average and against 1961-90 average at that. It's going to take a truly exceptional May for spring 2013 not to be below the 1961-90 average. It needs to be record breaking against 1971-2000 and 1981-2010 averages.

[Stu_London]

annual rolling CET 01.05.12-30.04.12

 

 

9.04C

[vforvendetta]

Dear weather - history,

 

Always find your posts excellent, bit if a statto myself.  What would you say the chances are now of us going sub 9 for annual CET at the end of May?  I think we need to finish at 11.2 or below for May don't we?  If we finish  the year sub 9, will this be the first time since the mid eighties that we have had to sub 9 years in a decade?

 

cheers

[Weather-history]

Dear weather - history, Always find your posts excellent, bit if a statto myself.  What would you say the chances are now of us going sub 9 for annual CET at the end of May?  I think we need to finish at 11.2 or below for May don't we?  If we finish  the year sub 9, will this be the first time since the mid eighties that we have had to sub 9 years in a decade? cheers

Thanks for your comment.Yes I think it is about 11.1C or less for a sub 9C. I think we are in with a reasonable chance. It will be the first tie since the 80s of two sub 9C years in a decade.

[SomeLikeItHot]

Looks to me like the annual rolling CET will drop below 9 again around the 28th - due to the very warm spell last year that started on the 22nd, being matched by a cold/cooler spell this year.

[SteveB]

Now we know that May has come in -0.7c under the monthly average, that means the first 5 months of the year are all below.

 

How are we looking for a sub 10c year, and what would need from the rest of the year  to stay below 10c, or even below 9 or 8c

Edited June 1, 2013 by SteveB

[Relativistic]

I think, if I have done this correctly, for sub 8C (pretty unlikely), we'd need the rest of the year to average approximately 9.8C. The mean for the rest of the year for 1961-1990 data is around 11.7C.

For sub 9C we'd need to average 11.5C, just a smidgen below the mean, and for sub 10C we'd need to average around 13.2C (also looks very unlikely).

[SteveB]

Cheers 22n10b, another sub 10 likely then

[Weather-history]

If the next 7 months were as warm as those of 2006, this year's CET would be 10.26C

If they were like those of 1993, it would be 8.57C

[Weather-history]

2010: 8.83C

Feb 2011- Jan 2012: 10.84C

June 2012-May 2013: 8.93C

Quite a sandwich contrast.

What is so strange is the coolest summer for 23 years lies in the middle of that warm period.

[SomeLikeItHot]

Now we know that May has come in -0.7c under the monthly average, that means the first 5 months of the year are all below.

 

How are we looking for a sub 10c year, and what would need from the rest of the year  to stay below 10c, or even below 9 or 8c

 

Regressing the CET mean to May vs final CET we would expect with a current CET to May of 5.44 a final CET around 8.9 at the end of year (based on data from 1900 and later).

 

In the entire history only 1 year has acheived a CET above 9.5 from a CET to May of less than 5.8, that being 1947 which had a CET to May of 5.2 and final CET of 9.57. 

 

coldest Start to May to get a CET over 10 was 1857 which has a CET to May of 6.2.

 

So history says

[*]it would take a recordbreaking turn around to break a CET of 10.

[*]we are quite likely to end up less than the 61-90 CET mean (9.5)

[*]and probably a good chance of CET going less than 9.

A lot depends on what happens in december as the winter months have a much larger variance in outcome than the warm months.

[SteveB]

I would be more than happy to have June/July/August post above average CET's & the rest of the year to come in around or below average & still get below 10c.

[Weather-history]

Looks like the coolest first half to a year since 1986

[GRHinPorts]

Might be worthwhile pegging the annual running CET figure as of July 3rd since it might be the lowest we see it for some while to come.  

[SomeLikeItHot]

With June coming in at 13.6, the second end of month in a row with the running annual mean below 9 - first time since the 80s.

 

July will need to be 16.1 or less to keep it below 9.

[VillagePlank]

Playing around with CET series, this morning.

 

I took the entire CET series, and averaged the annual temperatures. I then detrended it by linear regression. I then calculated the running decadel standard deviation, and this is what I got

 

 

Which, I think, is quite fascinating. Visually, there's some sort of quasi-cyclical behaviour going on. If you count the years between the minimum of each quasi-cycle, you get this,

 

 

Which averages out to exactly 30 years between minima. What could this mean? Well, the standard deviation is a measure of how far data points are away from the average, so should we expect more records to be broken near the maxima of the cycle since the values are likely to be further away from the mean? Does it also mean that during this cycle's minima the CET becomes more predictable since the deviation from the previous ten years annual CET temperatures is (much) less. Does it also imply that the CET temperature record is some function of the previous ten years strongly suggesting the presence of hysteresis?

 

Also, of interest, is the rather obvious observation that the trend of the running standard deviation is going down (obvious, visually) This seems to be clear evidence that the CET temperature series is getting more .. ahem .. boring: does this imply less very hot weather and associated thunderstorms, and, perhaps, less cold weather with less snow? Of course, the corollary is that current AGW proposes more extreme weather events as CO₂ increases; this is not supported by empirical observation for the CET zone which shows the very opposite to be happening.

 

What do you guys think?

 

Edited July 5, 2013 by Sparkicle

[SomeLikeItHot]

Playing around with CET series, this morning.

 

I took the entire CET series, and averaged the annual temperatures. I then detrended it by linear regression. I then calculated the running decadel standard deviation, and this is what I got

 

cet_stddev.png

 

Which, I think, is quite fascinating. Visually, there's some sort of quasi-cyclical behaviour going on. If you count the years between the minimum of each quasi-cycle, you get this,

 

min_diff.png

 

Which averages out to exactly 30 years between minima. What could this mean? Well, the standard deviation is a measure of how far data points are away from the average, so should we expect more records to be broken near the maxima of the cycle since the values are likely to be further away from the mean? Does it also mean that during this cycle's minima the CET becomes more predictable since the deviation from the previous ten years annual CET temperatures is (much) less. Does it also imply that the CET temperature record is some function of the previous ten years strongly suggesting the presence of hysteresis?

 

Also, of interest, is the rather obvious observation that the trend of the running standard deviation is going down (obvious, visually) This seems to be clear evidence that the CET temperature series is getting more .. ahem .. boring: does this imply less very hot weather and associated thunderstorms, and, perhaps, less cold weather with less snow? Of course, the corollary is that current AGW proposes more extreme weather events as CO₂ increases; this is not supported by empirical observation for the CET zone which shows the very opposite to be happening.

 

What do you guys think?

 

 

Interesting post.

 

You can replicate that pattern quite simply, by taking rolling 10 year STDevs as the effect of the trend within any 10 year window is small and the STdev is just calculating the variance and explicitly deducts the average in each period.

 

One thing you would expect to see is spikes in StDev if there was regime transitions in the climate as you would get maximum variance when the STdev calculation period spans the transition between the two regimes and lower variance when within a regime. An obvious candidate would be the AMO as I think that we can probably see several of the spikes occur very close to the transition years around (1920, 1960, 1990s, and possibly around 2007). However I am no expert on the AMO or its impact, just throwing out some ideas.

Edited July 5, 2013 by SomeLikeItHot

[BornFromTheVoid]

Not sure that the AMO fits the bill...

 

 

The data is here http://www.esrl.noaa.gov/psd/data/correlation/amon.us.long.data

 

[SomeLikeItHot]

Not sure that the AMO fits the bill...

 

AMO 10yr.JPG

 

The data is here http://www.esrl.noaa.gov/psd/data/correlation/amon.us.long.data

 

You are correct, To get the sort of effect that I was talking aobut you need to offset the AMO by around 5 years. - ie the spikes in variance actually preceed the AMO by 5 years, even so its hardly water tight still below is what I was talking about. (11 year central average)

 

 

 

 

 

Edit: Grr the label should say annual CET variance, not average - also they are 11 rather than 10 year averages in sparkicles post just so taking a central average is easier.

Edited July 5, 2013 by SomeLikeItHot

[Isolated Frost]

2010's first 6 months finished at 7.1c here, 2013's provisionally at 6.4c, which is quite something. Though July was notably warmer this year, and it will be extremely difficult to near November and December, but it's likely to be very close, and 2013 will end up as one of the coolest of the last 20 years.

[Relativistic]

If we could go sub 9C again that would be excellent, although I fear that all this talk of a very mild Autumn is threatening that. I'm thinking that September will finish above average, but hopefully from then on things will cool down again. The last two very cool November's were 1993 (4.6C) and 1985 (4.1C), so it would be interesting to have another one like these two, or better.

The very warm Summer notwithstanding, sub 9C Is doable at a push. September to December would have to have a cumulative anomaly of around -4C. With what is looking like an above average September, if we had a cool October to November period and a notably cold December then it could happen. Seems unlikely though with, as I've already mentioned, all the talk of a mild or even warm Autumn.

Edited August 31, 2013 by 22nov10blast

[Relativistic]

After a quick calculation, if August finishes at 17.0C (will likely finish lower than that), and all of the remaining months of 2013 finish bang on the 1961-90 averages, then we will end up at around 9.3C as an annual CET value.

[March Blizzard]

I think we need 3 of the remaining 4 months of the year to be below average to get 9 in total, which would beat 2010's 8. Probably a tall order. Parity with 2010 is obviously more likely but by no means a given. We would probably be looking for a cool - cold Nov/Dec.

When was the last sub 4C November? Sub 3C must be rarer than dog eggs when you consider the coldest on record is 2.3C.

To answer my own question the last sub 4C November was 1925, last sub 3C was 1915. Dare I say we're "overdue" a notably cold one?

Edited August 31, 2013 by March Blizzard