The Rise and Fall of Central England Temperatures; Part 3 2000-2019

Guest post by Tony Brown

This is the third examination of Central England Temperatures (CET) in a series that commenced in 2015 and which has charted the recent decline in temperatures from their highest values. The two previous articles in this series are referenced here;

The Rise and Fall of Central England Temperatures; Part 1 covering 2000 to 2015

The Rise and Fall of Central England Temperatures; Part 2 covering 2000 to 2017

When referencing any ‘decline’ we need to put that into context against CET’s overall substantial rise in recent decades. The official CET dataset used in this article, which is compiled by the UK Met office is linked here and shown in Figure A);

It should be noted that the values between 1538 and 1658 are my own reconstruction and are not used at all in this current paper.

Note: Weather comprises the day to day events that we all experience. Climate is officially the trend of the weather (often temperature and rainfall) taken over a continuous thirty year period. The two terms have sometimes been used in an interchangeable manner here, when a period of more than a year is being examined.


Figure A

According to the Met office; ‘Since 1974 the data have been adjusted to allow for urban warming: currently a correction of -0.2 °C is applied to mean temperatures.’

Over the last couple of years an exercise has been carried out within the Met Office to re-evaluate urban warming values affecting England, bearing in mind the 25% increase in population and many additional buildings and infrastructure since 1974. I understand the results are now in the public domain so have summarised and paraphrased a series of email conversations conducted with the Met office over the last few weeks.

The new version of HadCET will be fully documented in a scientific journal paper to appear in due course. Meanwhile, the existing CET software (has been used) to include ‘official’ values for January 2020. (You may need to refresh your browser to see them)

Although the current CET revisions are not yet finalised, I do have enough data put together to be able to examine the likely urbanisation corrections. In principle, the urbanisation corrections are applied from 1974 onwards but, because our daily and monthly CET values are only provided to one decimal place, it must take until 1980 for any of them to exceed 0.05 degC and hence make a difference when just one decimal place is quoted. (Accordingly) 1980 appears to be the earliest year for which a correction will be applied. 

The corrections vary by calendar month, but by 2000 some of them are as large as 0.3 degC for mean temperature; they do not continue to get larger after that.  Note also that the adjustments for minimum temperature are one-and-a-half times as large as those for mean temperature, and for maximum temperature the adjustments are only half those for mean temperature – this is because the urban fabric tends to hold on to daytime heat through the night, especially in summer, so urban minimum temperatures remain further above the rural equivalent than is the case for maximum temperature.”

CET uses three weather stations to record temperatures, contained in a triangle roughly in the centre of England, bounded by London, Bristol and Lancashire. As far as possible stations are placed away from urban areas but the country as a whole is small and crowded and as the Met office has recognised for many years, has been affected by urbanisation. No doubt there will be much discussion over the changes to HadCET once the scientific paper has been published, so no further reference will be made to it in this article.

‘Climate’ over the last 30 years;

The evolution of the CET temperatures over the last three decades can be seen in the series of graphics below. Figure 1 represents a scientifically derived 30 year climate period. In this respect it can be seen that the data during this period shows an upwards trend of 0.7c per century. This is a snapshot, and over a century temperatures don’t generally adhere to the relatively short term trends used to calculate them. This can be seen in the other graphics, where the shorter periods used will result in exaggerated periods of rise or fall.

When does the modern decline in temperatures commence?

The graphs shown in Figures 2-8 were compiled in order to see where the inflexion point came, when the overall annual temperature stopped rising and instead went into reverse. As can be seen it was becoming close to that state in the mid 90’s but the inflexion point seems to have been 1998 (Figure 6). This was the year of the large El Nino, described here, although that year was by no means the warmest in the recent CET record.

The remaining two graphics after 1998, (Figures 7 and 8) were compiled to ascertain whether the trend continues, if calculated from dates later than 1998.

There are several features we can observe in order to provide context, by stepping back and looking at the longer record shown in Figure A). The first is that the UK decadal weather through the centuries is highly variable, with numerous peaks and troughs and this has an impact on the full 30 years that represent a climate period.

When there has been a peak there has eventually been a decline and this can mean a 30 year ‘climate’ period can include a sharp decadal rise, a considerable decadal fall and a relatively static decade, which is then averaged out, and the nuances of the shorter periods of more extreme weather become less apparent.

The second feature is that the current modest decline since 1998 (Figure 6) is from a historically high plateau and covering only 22 years does not represent a 30 year period that can be scientifically termed ‘climate’. Rather it falls into the category of ‘interesting,’ as this observed decline does not correspond with the volume of scientific and media attention that has led to the UK Parliament and numerous local councils declaring a ‘climate emergency.’ Intriguingly it can be seen that no one living in England during the 21st century has experienced an overall warming trend.

The long term upwards trend

There has been a generally upwards trend in temperature since 1690, typified by notable jumps, then a fall back to some extent, with the period since around 1880 being generally more warmly benign and less extreme, as the trend continues upwards. At this stage it is impossible to predict if Figure 6) from 1998 represents merely a hiatus in the longer term warming or is part of a genuine trend to cooler values.

Does CET have a wider significance?

England is a geographically small area, but because three sets of records are used to compile CET and due to the UK’s geographical location, it is said to be representative of a much wider area. See the ‘Long Slow Thaw’ Section 6, a study by the author in 2011. The section was headed “Can CET represent a wider geographic area and establish the existence of a Hemispherically significant cooling period?”

As Mike Hulme remarks

“ (CET) is also quite well correlated with land temperatures over the entire Northern Hemisphere. At an annual level this correlation is about 0.4, but when average values over 10-year periods are compared this correlation rises to about 0.75.”

A variety of other science luminaries including the UK and Dutch Met office also see a reasonable correlation with the Northern Hemisphere. CET does therefore appear to have some broader relevance for ascertaining broad modern and historic trends in areas outside of England.


Looking at Figures 1 to 8, it was intriguing to note the manner in which changes in the character of one or two seasons (warmer to colder or vice versa) impacted on overall annual trends. Nowhere is this better illustrated than Figure 1, whereby Autumn suggests a sharp upwards trend of well over 2 degrees Centigrade per century. By 1995 (Figure 3) this had become a negative figure and at the turn of the 21st century (Figure 7) it had become minus 1.71C per century.

The Autumn figures from 2004 (Figure 8) show an even steeper decline, but whilst again it is ‘interesting’, it can be recognised that it has no scientific basis as in any 15 year period a couple of exceptionally cold or mild autumns will have a considerable impact and the century trend is highly unlikely to continue.

CET topics for future article

In the second part of this article we intend to examine the evolution of CET since 2004 (Figure 8) and also examine the past 350 years of CET records in order to put the current period into historical context. A detailed examination will be made of previous peaks and troughs and also of the apparent ever changing nature of the seasons, from warm to cold and back again,which appears to be a feature of the extended temperature record.

These variations last for varying amounts of time but appear frequent enough to warrant querying the view that the UK has an equitable and rather unchanging climate, with a certain constancy to the seasons.

Indeed at first sight, what can be considered as a ‘normal’ Spring, Summer, Autum or Winter appears to depend on which decades and which century you are looking from, and the more variable they are, the more impact individual seasons may have on overall annual trends.


Figure 1


Figure 2


Figure 3


Figure 4


Figure 5


Figure 6


Figure 7


Figure 8

Consequences of continued temperature decline

England has a large and growing population dependent on ever increasing amounts of food from its farmers and who need to keep warm, consequently a short term decadal decline of even half to one degree would have serious consequences for this country, as could adverse changes in seasonality.

Tony Brown February 2020

Acknowledgements; With grateful thanks to Ed Hoskins for compiling the many graphics used in this article. Ed has a great site which covers numerous aspects of Climate, Co2 emissions, Energy and much more.

All data used in this article is derived from statistics maintained by the Met Office and available here;

Figure A) comes directly from it.

0 0 vote
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
Dodgy Geezer
February 15, 2020 2:34 am

“….England has a large and growing population dependent on ever increasing amounts of food from its farmers and who need to keep warm, consequently a short term decadal decline of even half to one degree would have serious consequences for this country, as could adverse changes in seasonality….”

1 – I am sure that farmers and the general public are able to optimise their activities depending on the temperature and seasonality. One should NOT take current optimum behaviour and extrapolate from that to show a decline in general efficiency. Instead, some indication of the width of the optimisation band should be given.

It may be, for instance, that warmer average temperatures of one degree would ruin current planting, but allow farmers to plant a different type of crop with greater yields…

2 – I thought that globalisation and trade meant that countries could specialise in producing for world markets? So English/Welsh/Scottish/Irish farmers may not be able to compete with Argentine beef, New Zealand lamb, or Canadian wheat in any case…

Ben Vorlich
Reply to  Dodgy Geezer
February 15, 2020 3:47 am

This has been a long term trend in the UK. Where I grew up in the 1950s oats and hay were grown in marginal areas as animal feed. In fields not used for crops Lazy Beds and other evidence of crop growing was obvious. Hill grazing is now mainly forested, with non-native conifers which resumably are now the most profitable crop.

Although 7000 hectare and 14 million trees have given way to even more profitable windfarms.

Jeremy Poynton
February 15, 2020 3:06 am

“According to the Met office; ‘Since 1974 the data have been adjusted to allow for urban warming: currently a correction of -0.2 °C is applied to mean temperatures.’

In a warm day last summer, the difference in temperature between our local supermarket car park and where we live a couple of miles away in a leafy village was 6 (SIX) Degrees.

Just saying.

Reply to  Jeremy Poynton
February 15, 2020 3:32 am

the temp difference between out main street in a very small rural town and my home a mile away is very noticeable on a hot day and yes at least 5c I would say.
everything closer together, cars and aircons etc

Reply to  Jeremy Poynton
February 15, 2020 3:47 am

At least some allowance is made for uhi, unlike most data sets. Such allowances tend to be ‘smeared’ across a very wide area that will include lots of temperature differences thereby reducing their overall impact.

Yes, there can be huge differences between urban and rural and whether this allowance is appropriate will be argued from many different viewpoints.

Lets look forward to the paper the Met office will be getting peer reviewed and see how this is all calculated.


Ben Vorlich
Reply to  tonyb
February 15, 2020 3:54 am

It’s interesting that Met Office weather forecasts acknowledge differences of greater than 5’C on still days/nights yet their data sets don’t.

UHI is very dependant on wind speed in my view. Storm Dennis passing through will even temperatures across Central England giving no more than a degree or two, a blocking high will lead to a difference in excess of 5’C without taking that into account then daily/weekly/monthly averages are pretty meaningless. This February has been wet and windy, February 1963 had a lot of still cold days and nights.

John Finn
Reply to  Ben Vorlich
February 15, 2020 5:59 am

It’s interesting that Met Office weather forecasts acknowledge differences of greater than 5’C on still days/nights yet their data sets don’t.

A UHI average of 0.2 to 0.3 degrees is about right. There might be a wide variation in temperature on some days but not all. The CET can be validated against rural stations and is shown to be pretty consistent. I have access to local temperature records from a location which is slap bang in the middle of the CET region. The local trend is very similar to the CET.

In any case, the absolute UHI is largely irrelevant. It’s the trend in UHI in which we’re really interested.

Ben Vorlich
Reply to  John Finn
February 15, 2020 8:29 am

I’m very skeptical that a blanket adjustment for UHI does anything for anything particularly a trend. If you add 0.3’C to thirty years worth of data why bother?

John Finn
Reply to  John Finn
February 16, 2020 1:27 am

Ben Vorlich February 15, 2020 at 8:29 am

I’m very skeptical that a blanket adjustment for UHI does anything for anything particularly a trend. If you add 0.3’C to thirty years worth of data why bother?

It doesn’t add anything to the trend. However, the 0.3 deg would be the effect of the UHI ‘trend’. In other words 2010 readings could be boosted by 0.3 deg compared to mid 1970s.

Clyde Spencer
Reply to  Ben Vorlich
February 15, 2020 11:42 am

Furthermore, on windy days, the temperature differences in surrounding rural areas will be asymmetrical. That is, one can expect temperatures downwind from urban areas to be higher than the upwind areas, and areas outside the peri-urban periphery parallel to the wind direction. So, if all meteorological stations are used for homogenization to correct for UHI, the so-called ‘rural’ average will be too high if the downwind locations are included.

Reply to  tonyb
February 15, 2020 8:06 am

“At least some allowance is made for uhi, unlike most data sets.”
There is a question as to how much you should adjust for UHI. There is a case for adjusting if the place you are measuring is not typical of the region. That is a measurement issue, because if you measured somewhere else, you would get a different answer. But if the UHI effect is common to the region, it is not a measurement issue, and should not be corrected. The temperature really did go up, and the measurement should reflect that. One it is established what happened, you can then argue about why.

“UK has an equitable and rather unchanging climate”
I think you mean equable.

Reply to  Nick Stokes
February 15, 2020 8:36 am

Hi Nick

My dictionary defines it as ‘reasonable’ which is surely correct, but yours is better. Setting that aside I hope when the Met Office paper comes out that you will read and comment on it. I do think UHI is important but its overall impact needs to be kept in perspective when it is applied to the wider area


Reply to  Nick Stokes
February 15, 2020 8:46 am

But if the UHI effect is common to the region, it is not a measurement issue

Translation: These aren’t the droids you’re looking for.

If only poor old Nick was smarter than the average bear, then his maybe his wiseguy mind tricks would be effective.

Gerald Machnee
Reply to  Nick Stokes
February 15, 2020 11:08 am

UHI is common to large cities not regions.
0.2 is small for a large city.

Reply to  Gerald Machnee
February 16, 2020 1:24 am

We live in a small arable hamlet in the UK which is consistently 1°C lower in temperature than the nearest small country town 5 miles away.

Clyde Spencer
Reply to  Nick Stokes
February 15, 2020 11:58 am

If one is attempting to only describe and document trends in overall regional or global temperatures, then I agree that urban temperatures should not be ‘corrected.’ However, if one is interested in trying to tease out attribution — particularly if CO2 is the driving force — then urban adjustments should be done because we know that it is the differences in reflectivity and specific heat of urban building materials, conversion or rural to urban land-use, and lack of significant transpiration that drives the urban heating and retention of heat. Not correcting for UHI ends up confounding urban forcings with the effects of ‘green house gasses.’ Just as one should analyze the behavior of oceans separately from land, urban land and rural land should be analyzed separately — if one is really interested in learning what is happening and why!

Michael Jankowski
Reply to  Nick Stokes
February 15, 2020 5:30 pm

“…There is a question as to how much you should adjust for UHI…”

That’s one sad commentary. At this point, there certainly should not be any question.

Chad Jessup
Reply to  Nick Stokes
February 17, 2020 6:40 pm

“But if the UHI effect is common to the region, it is not a measurement issue, and should not be corrected. The temperature really did go up, and the measurement should reflect that. On[c]e it is established what happened, you can then argue about why.”

Excellent points.

Reply to  Jeremy Poynton
February 15, 2020 3:53 am

Measured by your cars thermometer? Car in sun? when you had driven for a couple of minutes what was the temperature?
The absolute temperature is not a suitable measure (this is why anomaly is used) – Take the average supermarket temp (over 10 years+) subtract that from the recently measured temp. Take the average temp of your leafy village over the same period from the measured temperature. Now compare the anomalies.
Report back!

Tim Gorman
Reply to  ghalfrunt
February 15, 2020 5:17 am

Anomalies associated with averages are also not suitable measures. The entire climate determination would be better off using degree-says, especially cooling degree-days and warming degree-days as a measure. The impact of the climate occurs at the edges of the temperature envelope, not at the “average”. The average can be impacted by higher minimums just as easily as higher maximums and vice versa. The average masks where the real changes occur. Integration of the entire daily temperature record used to calculate cooling and heating degree-days would be far more useful for determining what is happening with the “climate”.

Joseph Zorzin
Reply to  ghalfrunt
February 15, 2020 5:42 am

This is probably a dumb question- but will the anomaly at the supermarket be the same as the anomaly at a nearby rural area? I should think it would be higher- but just guessing.

Gerry, England
Reply to  ghalfrunt
February 15, 2020 10:01 am

I have seen differences of 5C regularly when driving out of the London suburbs to my rural home at night and as a return of a journey made a couple of hours earlier. I would think that was a quite valid measurement. I have also seen a difference of 4C on a winter night where it was a freezing -1C at my house and then 3C on the edge of Gatwick airport and then dropped back to -1C with no more than 15mins stopped in the middle of the journey. I was surprised by this one and it was a one off but no reason not to think it is representative of the UHI around Gatwick Airport and the warehousing all around.

Reply to  Jeremy Poynton
February 15, 2020 6:21 am

a correction of -0.2 °C is applied

this is a joke…right?

and 1978….exactly when temps jumped up and off the chart

Stephen Richards
Reply to  Jeremy Poynton
February 15, 2020 10:54 am

In winter, on a still night, the difference can be 10°C by the admission of the UKMO.

Reply to  Stephen Richards
February 15, 2020 11:40 am

The term “heat island” describes built up areas that are hotter than nearby rural areas. The annual mean air temperature of a city with 1 million people or more can be 1.8–5.4°F (1–3°C) warmer than its surroundings. In the evening, the difference can be as high as 22°F (12°C).

Patrick Healy
Reply to  Gator
February 17, 2020 1:45 am

So what would be the “real” temp at Heathrow airport measuring station the other year , when we had the hottest day evah?

February 15, 2020 3:26 am

“to see where the inflexion point came, when the overall annual temperature stopped rising and instead went into reverse. ” That’s not an inflexion, that’s a turning point.

Bloke down the pub
February 15, 2020 4:05 am

‘England has a large and growing population dependent on ever increasing amounts of food from its farmers and who need to keep warm, consequently a short term decadal decline of even half to one degree would have serious consequences for this country.’
It would be reassuring to know that the UK government understood this before they introduce stupid legislation to ban gas fired boilers. Sadly, I suspect that they do not.

Reply to  Bloke down the pub
February 15, 2020 4:11 am

I suspect they know nothing of our climate. The idea that we can switch over our gas boilers to highly expensive heat pumps-needing a considerable boost from other energy sources- introduce tens of millions of electric cars and power all this by renewables is naïve in the extreme.


Reply to  tonyb
February 15, 2020 5:02 am

A somewhat generous description of political in(s)anity, I would submit!

Reply to  tonyb
February 15, 2020 7:30 am

In the Netherlands the realization is just setting in that the country has been talking itself up the same politically wishful energy creek and now can’t find the proverbial paddle.
Several recent studies there show that in order to go electric, the country will need to vastly increase the use of natural gas for generation as neither wind nor solar have a hope in hell of covering the required increases – thus in actual increasing its CO2 output.

Julian Flood
Reply to  Bloke down the pub
February 15, 2020 10:38 am

Let me once more tell my true story about the UK’s Minister for Energy and Climate Change. I’d been banging onto him for some time about faulty climate science and the unreliability of wind energy. As a county councillor I’d met him several times and he is very polite. at one meeting he came bouncing up to me.

MfE&CC: Julian! Solar! Solar’s the thing! Prices dropping, really cheap, it’s the future!
JF: Yes, Minister. But you’ll have to find a way of storing the energy.

A telling pause. The MfE&CC didn’t know you use electricity or store it, it doesn’t somehow just hang around. He hadn’t been briefed and as his background was Oxford PPE, Cambridge economics and straight into politics he had never had contact with reality beyond his circumscribed tribal boundaries.

He recovered his poise and stopped looking gormless.

MfE&CC: Yes! Store it! We’ll have to store it! He bounced away.

Either his civil servants were playing silly bs or they didn’t know either. Let’s hope his current staff are better — he’s now the Secretary of State for Health and n.coronavirus is on the horizon.


February 15, 2020 4:11 am

I recommend a visit to the Met office web site and a look at the mean temperatures and sunshine hours from the start of the 20th century. What is amazing is that the temperature rise of about 0.4 dog is precisely mirrored by the trend in the increase in sunshine hours. Can anyone explain what seems to be an ongoing decrease in cloudiness. How does this relate to gig effects! What is of concern is that a continuing reduction in cloud cover could be more dramatic than the dramatic forecasts of the current-models.

Reply to  Spen
February 15, 2020 4:25 am

Spen: to save people hunting, here’s the link to graphs of UK temperatures, rainfall and sunshine going back to 1910 from the Met Office:

Reply to  Carbon500
February 15, 2020 4:47 am

Send to Willis Sun has no effect on Climate no pun intended

Tim Gorman
Reply to  Eliza
February 15, 2020 5:08 am

The sun has no effect on climate?

Huh? With no sun we would have no climate at all. Just a frozen ball sailing through the heavens.

Reply to  Tim Gorman
February 15, 2020 5:38 am

Being sarcastic cheers

Reply to  Tim Gorman
February 15, 2020 8:01 am

“Eliza February 15, 2020 at 5:38 am
Being sarcastic”

Sarcastic? Not even close.

Intentionally mean spirited, yes!

Julian Flood
Reply to  Spen
February 15, 2020 10:42 am

“Can anyone explain what seems to be an ongoing decrease in cloudiness.”

Yes. Fewer aerosols decreasing stratocumulus cover. Why the Blip? (See Wigley) refers.

I know that’s a bit Mosherish but I’m rather busy. Please look at “Seawifs oil spills” and ponder the implications.

Bloke down the pub
Reply to  Spen
February 16, 2020 1:52 am
February 15, 2020 4:47 am

According to a certain professor Phil Jones, there is nothing scientific about “climate is 30 years”. From: Phil Jones To: “Parker, David (Met Office)” [and others]:
Subject: RE: Fwd: Monthly CLIMATbulletins, Date: Thu Jan 6 2005

“Just to reiterate David’s points, I’m hoping that IPCC will stick with 1961-90. The issue of confusing users/media with new anomalies from a different base period is the key one in my mind. Arguments about the 1990s being better observed than the 1960s don’t hold too much water with me.

There is some discussion of going to 1981-2000 to help the modelling chapters. If we do this it will be a bit of a bodge as it will be hard to do things properly for the surface temp and precip as we’d lose loads of stations with long records that would then have incomplete normals. If we do we will likely achieve it by rezeroing series and maps in an ad hoc way. There won’t be any move by IPCC to go for 1971-2000, as it won’t help with satellite series or the models. 1981-2000 helps with MSU series and the much better Reanalyses and also globally-complete SST.

20 years (1981-2000) isn’t 30 years, but the rationale for 30 years isn’t that compelling. The original argument was for 35 years around 1900 because Bruckner found 35 cycles in some west Russian lakes (hence periods like 1881-1915). This went to 30 as it easier to compute. Personally I don’t want to change the base period till after I retire!

David Parker knew the value of sticking with 61-90, as it had some of the coldest years of the 20th century and was guaranteed to show warming when compared with a period where temperatures had recovered to more amenable levels: At 09:22 05/01/2005, Parker, David (Met Office) wrote:

“There is a preference in the atmospheric observations chapter of IPCC AR4 to stay with the 1961-1990 normals. This is partly because a change of normals confuses users, e.g. anomalies will seem less positive than before if we change to newer normals, so the impression of global warming will be muted.”

The CET period 1931-60 was warmer than 61-90, at 9.6 C against 9.47 C. The periods 1721-50, 1751-80, 1781-1810 as well as 1931-60, all had hotter average summer temperatures for the period than 1961-90.

There were always reservations about what constituted a climate period: “Representing Uncertainty in Climate Change Scenarios and Impact Studies” ECLAT-2 Workshop Report No . 1 Helsinki, Finland, 14-16 April 1999

Edited by : Timothy R. Carter , Mike Hulme and David Viner, Published by the Climatic Research Unit, UEA, Norwich, UK. September 1999

“Climatologists commonly describe the present-day climate using observations from a recent thirty-year period (e.g. 1951-80 or 1961-90). The performance of GCMs at simulating present climate can be tested with reference to such information, although measurement errors, interpolation errors and sampling errors lead to considerable uncertainty regarding the true baseline climate (e.g. New et al.., 1999).

Climate is also known to vary naturally on multi-decadal (e.g. 30-year) time scales and for reasons that have nothing to do with anthropogenic forcing.

Determining what is the true level of natural climate variability on 30-year timescales is not therefore straightforward. Observational data are limited to at most usually 100 years or so (and in any case may already contain an anthropogenic signal).

Another way of obtaining estimates of multi-decadal variability is to use multi-century unforced GCM simulations. However, with only a limited observational record and uncertain multi-century paleaoclimatic reconstructions of climate (Jones et al., 1998) available for comparing with model outputs, it is difficult to judge how effectively such model simulations represent the true natural variability of climate.”

It was however, declared at Kyoto in 1997, that “the science was settled.”

The CET annual average declined from 1949 to 1986, by 1.88°C, in spite of an atmospheric CO2 increase of 36.5ppm. There was a sharp rise of 0.72°C in 1988 and a further 0.73°C in 1989, to reach current temperatures with a high of 10.63°C in 1990. That has been exceeded in 2006 and in 2014, temperatures have been falling since. 2019 was the 23rd warmest in the CET record from the Met Office’s own rankings.

Global atmospheric CO2 went up by 56.95 ppm since 1990, global CO2 emissions rose by some 14.6 billion tonnes, or 65%. Net result in the UK, nothing, temperatures have not tracked CO2, there has been no year on year increase. Annual changes in CET temperature do not correlate with annual changes at Mauna Loa. Annual changes in emissions do not correlate well with annual changes at Mauna Loa.

Is the climate actually changing? If so, from what, to what? When is “pre-industrial”? Before 1750, as first put forward by Hansen, or 1850 and now 1880 that it has since morphed into?
“Countdown to Global Catastrophe” 2005

“The countdown to climate-change catastrophe is spelt out by a task force of senior politicians, business leaders and academics from around the world – and it is remarkably brief. In as little as 10 years, or even less, their report indicates, the point of no return with global warming may have been reached.

The report says this point will be two degrees centigrade above the average world temperature prevailing in 1750 before the industrial revolution, when human activities – mainly the production of waste gases such as carbon dioxide (CO2), which retain the sun’s heat in the atmosphere – first started to affect the climate.”

Clyde Spencer
Reply to  dennisambler
February 15, 2020 12:10 pm

You said, “Arguments about the 1990s being better observed than the 1960s don’t hold too much water with me.” But, we are assured by the alarmists that the Law of Large Numbers makes the early data adequate for our purposes. /sarc

Steve Keppel-Jones
Reply to  dennisambler
February 20, 2020 6:27 am

Looking at the CET record, I wonder whether anyone was panicking in 1740 when temperatures had warmed up by 2 degrees since 1690. Clearly they were only a decade away from boiling to death at that rate.

February 15, 2020 5:00 am

Corrections and adjustments to short-period guesses as to how many fairies can fit on the head of a pin.

February 15, 2020 5:00 am

When does the modern decline in temperatures commence?

The first question should be “Is there a modern decline in temperatures?”

Starting in 1998 CET has a linear trend of -0.05°C / decade, so could be said to have fallen by around 0.1°C, but that trend is not remotely significant.

I think there is far to much year to year variability in CET to try to derive trends over a decade or two.

Reply to  Bellman
February 15, 2020 5:44 am


I do make those points by including all the graphs and pointing out that it is a small decline from a high plateau that is not scientifically valid. However we could also turn your interesting question round and ask “Is there a modern uptick in temperatures?”

Listening to the media hysteria the answer would be in the affirmative and many believe the increase to be very substantial, including the numerous councils appointing climate emergency officers.

I have no idea if this apparently downwards trend will turn into a scientifically valid one. However if all the UK governments plans are based on a substantial increase in temperatures and instead there is a fall, that does have repercussions and consequences


Reply to  tonyb
February 15, 2020 7:59 am

I’d say there was more evidence for a modern rise than for a recent decline., depending on how you define “modern”. For example, since 1970 the linear trend is +0.22°C / decade with a p value < 0.001. (This is using annual averages, but isn't corrected for auto-correlation).

I don't know what will happen to England in the future. It's possible that climate change will result in a cooler UK, but so far I don't see any substantial evidence for a cool down. I would hope that the Government aren't basing future plans on the last few years of CET, though these days I wouldn't be surprised at anything they do.

February 15, 2020 5:30 am

Few years ago I compared CET to Arctic temperatures but since the data before 1880 has wide margin of error, any likeness in the earlier trends may be a coincidence.
However, looking at the spectral power distribution there is considerable similarity.
I have not done any further research, but above doesn’t mean that the CET temperature rise and fall follows Arctic, there could be a large delay between two or even (less likely) that the two are in anti-phase.
Arctic warming in great part is due to the Atlantic currents moving large volumes of the equatorial warm waters north and the CET area would get earlier benefit from it. If so and the CET recent trend was cooling (which at the moment is not statistically significant) then it could be speculated that the Arctic might follow in not too distant future (?).

Reply to  Vuk
February 15, 2020 5:51 am


A couple of years ago I was in contact with Phil Jones in order to try and find out some information on Hubert Lambs work on wind directions.

He sent me a copy of ‘The Kings Mirror.’ This is a Viking piece of work where basically their observations were that the weather/ice in the lands to the North were not synchronous with the lands to the south. This is a 1917 version.

I am not claiming the Vikings were right but Its reasonable to assume though that sooner or later what was warm becomes cold and vice versa as water currents shift heat around.


February 15, 2020 5:51 am

Why doesn’t the Met office simply discard temperature measuring stations that have been potentially corrupted by urban heating and just use those that have not been urbanized? Surely this would be better then trying to guess what the urban heating amount is? But then I guess they might not be able to announce a “warmest day eva” record from a temperature station at Heathrow airport!

February 15, 2020 6:12 am

The CET is composed of four stations. One of those stations was based at Ringway until 2008, and another at Blackpool.

Trouble was ‘Ringway’ is another name for Manchester International Airport, and the measuring station was right in the middle of the taxiways, opposite the jet-engine run-up bays. Every aircraft that taxied past, and every engine test, blasted the thermometer with hot air. All this monitoring station did, was to record the huge change at Manchester from small regional turbo-props to vast intercontinental jets.

Oh, and the Blackpool monitoring station was also on an airport, which had jet traffic up until 2015.

There is no way that the CET data was representative of ambient rural conditions, after 1985.


Reply to  ralfellis
February 15, 2020 6:28 am

From WUWT on Jan 29, 2017 Guest essay by Tom Barr
“The Met Office relies upon just three weather stations to record the Central England Temperature: Stonyhurst (Lancashire), Pershore (Worcestershire) and Rothamsted (Hertfordshire).

Reply to  Vuk
February 15, 2020 7:12 am

Ringway and Squiers Gate were used from 1974 – 2004, but treated as a single station. Since 2004 Stonyhurst has been used instead.

Clyde Spencer
Reply to  Vuk
February 15, 2020 12:16 pm

It appears that your link is broken.

Reply to  ralfellis
February 15, 2020 6:32 am


I had this conversation about Ringway with David Parker himself at the met office a couple of years ago. From a paper;

“But in November 2004, the weather station Stonyhurst replaced Ringway and revised urban warming and bias adjustments have now been applied to the Stonyhurst data after a period of reduced reliability from the station in the summer months.”

My conversation went along the lines that the characteristic ‘hump’ we can observe in Figure A) above may have been as a result of the choice of stations. Not that he would admit that of course but Ringway in particular was not a good place to site a station


February 15, 2020 6:34 am

Thanks for a very interesting post: informative and balanced.
I was struck by how the average temp figures jump around, especially when looking at the seasonal ones. Winter 2007 was roughly 4 degrees warmer than 2011, for example. It reminded me of a nature study I read a few years ago which looked at how wildlife coped in two successive Aprils which were markedly different in weather (not surprisingly, the bad weather year had a serious impact but was not devastating).
Am I being simplistic in looking at these short-term swings and thinking that an overall gentle change is nothing to worry about?

Reply to  Mikehig
February 15, 2020 7:25 am


It is why I want my second part to look at seasonality as it is fascinating to note often large differences from year to year. It is when those changes are longer lasting that the problems could ensue, especially IF those changes are for cooler temperatures, as everything is predicated on it continually getting warmer.

The changes in Autumn in particular are quite surprising.


John Finn
February 15, 2020 7:11 am

I’ve just looked at temperature data from 3 rural stations to the west of the CET, i.e. Valley (Isle of Anglesey, Wales), Armagh (NI) & Valentia (off SW coast of Rep of Ireland). The decadal trends since 1976 are

Armagh 0.23 deg per decade (1974 to 2017)
Valley 0.23 deg per decade
Valentia 0.24 deg per decade
CET 0.23 deg per decade

UHI is not an issue.

Michael Jankowski
Reply to  John Finn
February 15, 2020 6:30 pm

Isle of Anglesey, Wales, has a population density of 250 people per square mile according to Wikipedia. US census bureau considers rural to be less than 100.

Valley looks pretty small, but I don’t know exactly where the weather station is located. Some googling suggests it is at the airport, which added a terminal and started having regular commercial flights in 2007 (RAF training prior to that). But I’m sure there’s no contamination of the record.

The City of Armagh had only about 14,777 people in the 2011 census. But it is also only 3.97 square miles, which calculates to 3,722 people per square mile. That’s on par with Houston, Dallas, Columbus…you get the picture. Sure, it only grew about 16% between 1984 and 2011, but it was projected to grow another 9% between 2016 and 2026. So there is and was a difference between Armagh today and 1974. But I’m sure there’s no contamination of the record.

Try harder.

John Finn
Reply to  Michael Jankowski
February 16, 2020 1:23 am

which added a terminal and started having regular commercial flights in 2007 (RAF training prior to that).

Ok – let’s look at the Valley trend up until 2007. Ooops – it comes out at 0.33 deg C per decade

RAF Valley has been in operation since 1940.
The Armagh Observatory was founded in 1789 and is located in 14 acres of landscaped grounds. The local buildings date back many years.
You don’t mention the Valentia Observatory.

You are desperately grabbing at straws. Our local weather station is located in a remote position on school grounds which date back several hundred years. The population of the nearby city actually fell between 1971 & 2000 and has only recently recovered to its 1970s peak. The temperature trend is remarkably similar to the CET trend.

Rhys Jaggar
February 15, 2020 7:22 am

The discussion of temperature vs farming is really only critical for early sowings and late harvests: quite frankly, most crops could not give a stuff if daily maxima were 21 vs 20C and daily minima were 11 vs 10C.

Going from 10C to 8C at night is no good for various summer crops, whereas excessively mild autumns are wonderful in terms of increasing growth on cabbages, swedes, leeks, winter radishes, fennels etc etc.

Late spring frosts, frosts in October are what will actually affect farmers a lot. Ditto serious droughts, serious deluges and flooding.

Most crops grown in the UK have a relatively wide margin of safety in terms of the seasons: potatoes, carrots, parsnips, leeks, onions, beetroots, swedes are all gong to still be grown even if temperatures dropped by 1C.

Scotland might be more problematical, but the South of England is not going to starve unless the farmers are totalitarian idiots (there is no evidence that they are or ever have been).

February 15, 2020 7:31 am

The CET monthly maximum and minimum temperatures are moving in step up to about 1990, but in the last three decades the minimum has gradually fallen behind, by up to 0.5C in the last 3-4 years.
Maximum temperature warming trend is 1.08C/centaury
Minimum temperature warming trend is 0.7C/centaury
Note: Sharp downward spike coincided with the Iceland’s April 2010 Eyjafjallajökull volcano eruption.

February 15, 2020 8:00 am

Please, please, spend the bits and bandwidth to make the numbers and lettering on the charts readable! (Or, perhaps, have the charts be links to the same chart at larger scale.)

Reply to  Ellen
February 15, 2020 8:13 am

Ellen hi
I’m sure you will find that the numbers and lettering on my charts above are easily readable.

February 15, 2020 9:26 am

Tony, a couple of comments.

First, let me suggest that the CET is not really suitable for some analyses. The problem is that it is a multiply spliced record. The stations that make it up have been changed no less than seventeen times! Here are the dates of the splices:

The splicing makes the CET unsuitable for any analysis of long-term trends that cross any of the splices. This, of course, doesn’t include this analysis of yours.

Next, I took the liberty of enlarging the graphics, as I couldn’t read them.

Finally, you might look into a nearby long single-station record, the dataset of the Armagh Observatory. I wrote about it in a post called “Is Armagh Burning“. Data is available here, and a description of the data is here.

Best regards,


Reply to  Willis Eschenbach
February 15, 2020 11:33 pm


Thanks for enlarging the graphics. That makes it much clearer. I have personally discussed CET with David Parker at the Met Office and also with other scientists there. It is probably the most scrutinised data base in the world and as such, changes are made only after careful reflection.

We are fortunate in having numerous weather reports that enables us to believe that the work Manley and others did over the years has provided a useful record that Parker has built on, and that at each ‘splice’ the records were reanalysed.

I tend to belong to the Hubert Lamb school of thought on temperature records and reconstruction in as much ‘we can understand the tendency but not the precision’ .In other words, records such as CET provide an invaluable insight into the climate stretching back to the Little Ice age, but we should not believe they are accurate to fractions of a degree.

Mind you, that can be seen in many records, even modern ones.

The fact that CET is continually reanalysed and such important elements as UHI revisited I see as a strength not a weakness.

Yes, I am aware of Armagh and others and have used it as a reference point many times.

Once again many thanks for improving the clarity of the graphics.

best regards


February 15, 2020 10:00 am

“The Autumn figures from 2004 (Figure 8) show an even steeper decline”.

The CET station composition changed in 2004, leading to a sudden anomalous cooling from that date:

comment image?w=840

The figure above was derived from an independent assessment of the regional average temperature variations, using around 30 stations in Central England . Berkeley Earth data can also be used to show the 2004 inhomogeneity caused by the change in station composition, simply by subtracting it from HadCET data.

February 15, 2020 12:26 pm

All I can see in the charts is variable weather. It would be useful to add rainfall data as the other major weather feature.
But I don’t see any climate emergency. So why is the British government behaving as if there is?

Steve Keppel-Jones
Reply to  Robber
February 20, 2020 10:31 am

Because it’s not about a climate emergency at all. It’s about a lack-of-socialism emergency… everything else is just window dressing.

Burl Henry
February 15, 2020 12:35 pm


The Hadley Centre listing of CET’s shows the average temperatures of the following years to be:

1875 9.48 deg. C.
1876 9.53
1877 9.26
1878 9.26
1879 7.44
1880 9.10

What do your records show?

February 15, 2020 1:27 pm

Thank you Carbon500 for provided the Met Office link. See 04.25am

But can someone answer my question about increasing sunshine hours and possible decreasing cloudiness. It’s driving me crackers.

Julian Flood
Reply to  Spen
February 15, 2020 2:26 pm

Our civilisation is spilling vast quantities of light oil which makes its way to the oceans. Oil smooths the surface, decoupling the water from the wind up to Force 4 (personal observation). This reduces the number of breaking waves. Waves produce salt aerosols which are lofted up to the boundary layer where they act as condensation nuclei and form a stratocumulus cloud layer. Fewer waves, fewer aerosol particles, less or thinner cloud, lower albedo, warming, panic and hysteria.

Smoothed ocean surface, slower currents, less sub-surface mixing, less nutrient for DMS producing phytoplankton. Less DMS, less cloud, panic, hysteria

Massive increase in farming, tree and scrub clearance. More erosion, more dust clouds, more dissolved silica in the oceans. Diatoms maintain their dominance until later in the spring bloom, suppressing DMS phytos. Less stratocu. Warming.

For smooths, Google Ben Franklin, Clapham pond. For dependence of stratocumulus cloud on aerosol amounts Google shiptrack images.

Julian Flood
Reply to  Julian Flood
February 15, 2020 2:28 pm

Well, you did ask.


Patrick Healy
Reply to  Julian Flood
February 17, 2020 3:13 am

Thanks for that. As an old sea dog who was one of the “rubber bucket” Merchant Navy (much derided) Weather Obs Ships, I can vouch for the amount of sand/silica which blew off the Sahara into the south Atlantic annually.
Presumably it still happens.
On the subject of “oiling troubled waters” it was/is a well known and practised method whereby bunker oil was sprayed overboard in order to give sailors some chance of survival in wartime, by lessening the waves. Maybe it worked – who knows.
Anyway I am disappointed that our present Prime minister can have the future direction of our economy dictated by his deep green concubine – allegedly!

Reply to  Spen
February 15, 2020 2:51 pm

Hi Spen
Let me have a go.
a) I think it is a bit pointless averaging annual sunshine hours across mostly cloudy country as UK/England where west and east coast are having mostly different conditions.
b) Averaging monthly hours at single place somewhere in the middle to me makes much more sense. As it happens Oxford University weather station has such data. I have looked into it 4-5 years ago
As they say: I don’t see much in there to write home about and ‘you pays your money you takes your choice’.

February 15, 2020 2:37 pm

Tony ==> Have you seen John Belville’s little book — I wrote about it in “Historical Note: Greenwich, England Mean Temperature, 35-yr Daily Averages 1815-1849“.

Belville’s eBook is found here. If the link doesn’t work, search Google Play eBooks for John Henry Belville’s “A Manual of the Thermometer”.

Reply to  Kip Hansen
February 16, 2020 2:03 am

Yes I noted it down when you wrote about it before but as my large pay cheques from the sceptic deniers fund have mysteriously dried up and my three archivists have had to be let go, it was useful for you to post it again so I can note it again and file it accordingly


Reply to  tonyb
February 16, 2020 6:09 pm

Tony ==> You’re welcome. I found it when looking for data on the construction and accuracy of 19th century thermometers.

nw sage
February 15, 2020 4:54 pm

Interesting to see that climate is defined as a trend in temp, rain, humidity, wind etc,etc. This means that a climate of zero is desired. ie no change in any or all of these characteristics, alone or combined in some way! Do any of these folks realize what they are saying??

February 15, 2020 4:58 pm

Thanks for the article.
From a farmer talking in the pub perspective, there doesn’t seem to be a “cold year” or a “ hot year”.
The seasons are all over the shop.
Historically, do English farmers have any idea what the next year will bring based on the current seasons weather?

Reply to  Waza
February 15, 2020 11:45 pm


Looking back to books that relate to farming as far back as the 14th century, farmers are often accused of ‘whining’ about the weather. Back then farmers would plant what previously grew and what they had seeds for. As can be seen, sometimes they would be lucky sometimes not and famines and food shortages were the result if the weather didn’t play ball. They became past masters at methods of storing.

Where problems ensued was when the nature of the seasons and years had substantially altered for more than one year and new crops had to be found that would be successful in the changed conditions.


Reply to  tonyb
February 16, 2020 2:05 am


February 15, 2020 9:37 pm

All I can say is that this is all quite terrifying.

donald penman
February 15, 2020 11:48 pm

I think that temperature is affected by topography also because cold air sinks into valleys from higher ground being more dense, some of differences attributed to UHI may be caused by this.

See - owe to Rich
February 18, 2020 1:52 am

Tony, I think you have a “1-off” in your graphs, because they appear to show that the two hot summers in the noughties were 2004 and 2007, whereas they were 2003 and 2006 (which I fondly remember – bring on the warming – GWIG – Global Warming Is Good).


February 28, 2020 6:22 pm

Hi Tony
A very quick and very dirty analysis of the CET data sort of confirms your take on the matter.

%d bloggers like this: