Record Temperatures & UHI

From NOT A LOT OF PEOPLE KNOW THAT

By Paul Homewood

Returning to the “Record” temperature at Kew Gardens last week, it’s worth taking a closer look at UHI:

https://data.london.gov.uk/dataset/londons-urban-heat-island-during-a-warm-summer-vdjgm

Although Kew Gardens itself is open space, it is surrounded on all sides by London’s built up environment. Kew simply cannot escape the Urban Heat Island effect of that entirely.

One study, above, found that average daily max temperatures were as high in and around Kew as elsewhere in London, during the summer of 2006. Kew was a good 2C hotter than the leafy southern extremities of Bromley and Croydon.

What is just as significant though is the temperature distribution during the July heatwave that year:

Note how the hottest parts were to the west, suggesting that the winds south-easterly winds had carried the UHI effect to the northwest.

The weather set up was very similar last week,and it is no coincidence that, along with Kew, three other west London sites, Heathrow, Northolt and Teddington were the four hottest spots in the country:

What we saw last week was not “climate change”, it was urban heat island effect.

If anybody is in any doubt about the unsuitability of Kew as a climatological station, a few years ago the BBC/Met Office laughably tried to present Heathrow as a reliable site, because its temperatures closely tracked Kew!

https://www.bbc.co.uk/news/uk-44980493

AS GOOD AS HEATHROW!!

That’s hardly a commendation!

The climate data they don't want you to find — free, to your inbox.
Join readers who get 5–8 new articles daily — no algorithms, no shadow bans.
5 12 votes
Article Rating
127 Comments
strativarius
June 3, 2026 2:47 am

For the very best in London weather it has to be Tooting, South West London. Unless the weather is coming from the north….

How is temperature measured?

To get the spike…

Nick Stokes
June 3, 2026 3:03 am

One study, above, found that average daily max temperatures were as high in and around Kew as elsewhere in London, during the summer of 2006.”

These were not measured temperatures. They were the output of a GCM, LondUM
LondUM (2011). Model data generated by Sylvia I. Bohnenstengel (*), Department of Meteorology, University of Reading 



strativarius
Reply to  Nick Stokes
June 3, 2026 3:09 am

Ever been to Kew, Nick? I have many times. The butterfly house is particularly warm as are the other glasshouses and walled gardens….

Reply to  strativarius
June 3, 2026 3:16 am

I guess the temperature station isn’t located within the butterfly house or the walled garden though?

strativarius
Reply to  TheFinalNail
June 3, 2026 3:27 am

Neither is the cafeteria.

In 2021 the Met Office assessed the site as CIMO Class 4 with an error margin due to siting of 2°C.
https://tallbloke.wordpress.com/2024/10/24/kew-gardens-dcnn-5260-open-to-question-plus-further-notes-on-st-james-park-london/

June 3, 2026 3:07 am

I just looked at the Met Office web site and Kew Gardens is still listed as a Grade 2 site. I thought someone said here recently that it had been downgraded.

Nick Stokes
Reply to  Oldseadog
June 3, 2026 3:53 am

Yes, it is Grade 2.

Leon de Boer
Reply to  Nick Stokes
June 3, 2026 4:16 am

Depends who you ask class 2,3 and 4 are possible answers.

The link above will give some background with questions and concerns about the site going back years.

strativarius
Reply to  Leon de Boer
June 3, 2026 4:43 am

In 2021 the Met Office assessed the site as CIMO Class 4 with an error margin due to siting of 2°C.
https://tallbloke.wordpress.com/2024/10/24/kew-gardens-dcnn-5260-open-to-question-plus-further-notes-on-st-james-park-london/

Reply to  Leon de Boer
June 3, 2026 7:01 am

The requirements for Class2 are:
Class 2 requirements demand the following conditions for the measurement point:

  • Terrain: The land should be flat and horizontal with a slope inclination of less than \(\frac{1}{3}\) (19°).
  • Ground Cover: The surrounding ground must be covered with natural, low vegetation (less than \(10 \text{ cm}\)) that is representative of the region.
  • Distance to Heat and Reflective Surfaces: The sensor must be located more than \(30 \text{ m}\) from artificial heat sources or reflective surfaces (e.g., concrete, buildings, or car parks). Artificial sources affect the measurement if they occupy more than \(10\%\) of the surface within a \(30 \text{ m}\) radius.
  • Distance to Water: Sensors must be located more than \(30 \text{ m}\) from any expanse of water (unless the water is a significant regional feature).
  • Shading: The installation must be away from all projected shade when the sun is higher than 7°. 

Kew seems to meet all those criteria.

Nick Stokes
Reply to  Phil.
June 3, 2026 2:43 pm

I think what happened here is that there was a brief period when the lawn was being used for children’s entertainment, with various transient structures. That put the class number up for a while. But that seems to have stopped.

Reply to  Nick Stokes
June 3, 2026 5:51 pm

That makes sense Nick.

Reply to  Leon de Boer
June 3, 2026 2:17 pm

Posted this a couple of days ago, showing just how much the CET and London regions is affected by urban heat

UK-3-x-3-new-1749654188.1402
Reply to  bnice2000
June 4, 2026 5:48 am

Those data are pretty unarguable.

June 3, 2026 3:13 am

This article completely misses the point that the new May record is still a new May record! It beat the old record on two consecutive days; first by +2.0C then by +2.3C the very next day.

The Kew Gardens Stevenson screen is located in the same place as it was when the previous May records were set. All the other siting criteria remain the same too. No newly-documented warming influences have been identified.

And speaking of UHI in the context of the US, as has been mentioned here many times, the “properly sited… state-of-the-art” (WUWT’s own description) USCRN sites have been measuring US surface temperatures for over 21-years now and the ‘pristine’ sites are warming at a faster rate than nClimDiv, which uses all sites and adds adjustments.

Check it for yourself.

Mr.
Reply to  TheFinalNail
June 3, 2026 4:03 am

That the world warms up by a few poofteenths as it emerges from the “Little Ice Age” shouldn’t be a surprise or cause for chicken little carry-on though.

And to put it all down to man-made CO2 is a level of absurdity that only far- left grifters could invent.

Reply to  Mr.
June 3, 2026 4:34 am

That the world warms up by a few poofteenths as it emerges from the “Little Ice Age” shouldn’t be a surprise or cause for chicken little carry-on though.

There is zero global warming trend over the first ~85-years of the global temperature record, between 1850 and the mid 1930s.

Was the “recovery from the ‘Little Ice Age‘” having a wee rest?

hadcrut4gl
Reply to  TheFinalNail
June 3, 2026 4:40 am

if you’re going to talk about a temperature record it might be useful to actually show a temperature record,

Reply to  Phil R
June 3, 2026 2:18 pm

They all show the same thing, mate.

Reply to  TheFinalNail
June 4, 2026 5:48 am

False. See bnice’s post above.

strativarius
Reply to  TheFinalNail
June 3, 2026 4:42 am

Why are you looking at ‘nanoseconds’? It couldn’t be cherrypick, surely.

comment image

Reply to  strativarius
June 3, 2026 7:07 am

Like your ‘cherrypick’ which fails to include the last 75 years?

strativarius
Reply to  Phil.
June 3, 2026 7:08 am

You are funny.

MarkW
Reply to  Phil.
June 3, 2026 8:10 am

It does show the last 75 years. It’s 0.1% of the final pixel.

Reply to  MarkW
June 3, 2026 10:25 am

No, the graph x-axis is ‘Years before Present’ which means years before 1950!

MarkW
Reply to  Phil.
June 3, 2026 10:54 am

Read it again, it says millions of years before present.
Which makes the last 75 years just a tiny fraction of the last pixel. Once you average 75 into 1 million, you will realize just how dumb your sound.

strativarius
Reply to  MarkW
June 3, 2026 11:26 am

What a plank!

Reply to  MarkW
June 3, 2026 2:22 pm

Read it again. Zero on that chart is, by convention, 1950.

Reply to  TheFinalNail
June 3, 2026 6:54 pm

CO2 is still VERY low..

So are the temperatures, even compared to the rest of the Holocene.

MarkW
Reply to  TheFinalNail
June 4, 2026 10:38 am

True and completely irrelevant.

Since 75 years is such a tiny fraction of a million years, it’s too small an amount to show up.

Maybe in 100,000 years there might be enough data to create a single pixel to the right of the zero mark, but for now, there is none.

Your complain is as usual both wrong and irrelevant.
Much like you are.

Reply to  MarkW
June 4, 2026 7:34 am

“The problem was tackled by the international radiocarbon community in the late 1950s, in cooperation with the U.S. National Bureau of Standards. A large quantity of contemporary oxalic acid dihydrate was prepared as NBS Standard Reference Material (SRM) 4990B. Its 14C concentration was about 5% above what was believed to be the natural level, so the standard for radiocarbon dating was defined as 0.95 times the 14C concentration of this material, adjusted to a 13C reference value of −19 per mil (PDB). This value is defined as “modern carbon” referenced to AD 1950. Radiocarbon measurements are compared to this modern carbon value, and expressed as “fraction of modern” (fM). “Radiocarbon ages” are calculated from fM using the exponential decay relation and the “Libby half-life” 5568 a. The ages are expressed in years before present (BP) where “present” is defined as AD 1950.”[16]

Large scale atmospheric testing of nuclear weapons after 1950 altered the ratio of C14/C12 which is a reason for the choice of 1950.

Reply to  Phil.
June 4, 2026 7:55 am

Also the lack of a vertical scale makes the graph useless. The CO2 value for 0 YBP is shown as significantly less than the Triassic average of 210 ppm which is ridiculous, Cretaceous is shown as 340 ppm which is around where it should be for 0 YBP and the current value would be ~100 ppm above that!

Reply to  Phil.
June 6, 2026 9:31 am

The problem is it’s a fake graph by a blogger which bares little relationship to the two papers cited! In fact the authors of the papers concerned have said that it has no relation to their work.

MarkW
Reply to  Phil.
June 4, 2026 10:39 am

See my response to TFN, then hang your head in shame.

Sparta Nova 4
Reply to  Phil.
June 3, 2026 12:52 pm

Read the text. You are wrong in the year.

Reply to  Phil.
June 3, 2026 8:20 am

That’s hilarious

I’ll bet you don’t know why it’s hilarious

Reply to  strativarius
June 3, 2026 2:20 pm

Your chart mostly covers a period when humans and human civilisation didn’t exist.

As if that didn’t matter.

MarkW
Reply to  TheFinalNail
June 4, 2026 10:40 am

The chart shows that throughout history, CO2 has never had any impact on climate.
That matters.

Reply to  TheFinalNail
June 3, 2026 5:51 am

Do you really believe that there were proper stations installed all over the globe during the period of 1850 to 1930?

MarkW
Reply to  Jim Gorman
June 3, 2026 8:12 am

Heck, we don’t have that today.
The satellite measurements, which didn’t start until a few decades ago, still only cover about 90% of the earth’s surface.

Reply to  Jim Gorman
June 3, 2026 2:23 pm

Do you really believe that there were proper stations installed all over the globe during the period of 1850 to 1930?

No. And no one says there were. Read the literature.

Reply to  TheFinalNail
June 3, 2026 6:50 pm

Here is where surface data was measured before 1920…

Anything “global” fabricated from them is FAKE.

And much of the data for the oceans before 2005, is also “just made up” according to Phil Jones from Hadley CRU.

tempmap
MarkW
Reply to  bnice2000
June 4, 2026 10:42 am

According to their interpretation of the law of large numbers, if we put 100,000 temperature stations in New York’s Central Park, we can know the temperature of the entire planet to within a few tenths of a degree.

Or if you want to be cheap, you can just use one station and have it take 100,000 readings in order to get the same result.

Reply to  MarkW
June 4, 2026 2:51 pm

WOW! You just nailed it to the wall! Climate science in s nutshell.

MarkW
Reply to  TheFinalNail
June 4, 2026 10:41 am

Even though you know the data is crap, you get yourself all bent out of shape over an alleged few tenths of a degree warming.

Robertvd
Reply to  TheFinalNail
June 3, 2026 8:30 am

But somehow most of the ice melt like in Glacier Bay was during those years.

Reply to  TheFinalNail
June 3, 2026 9:08 am

“There is zero global warming trend over the first ~85-years of the global temperature record, between 1850 and the mid 1930s.”

And you really choose to imply the temperature instrumentation and temperature recording station sitings (locations) from 1850 to mid-1930s compared to those from mid-1930s to 2026 have comparable accuracy (say to be even within +/- 1 deg-C)???

As has been quipped: “Absence of evidence is NOT evidence of absence.”

Good grief!

MarkW
Reply to  ToldYouSo
June 3, 2026 10:56 am

Prior to electronic recorders, only the daily high and low, estimated to the nearest degree were recorded.
And that’s before adding in any inaccuracies in the thermometer itself or the area surrounding the thermometer.

Reply to  MarkW
June 3, 2026 2:27 pm

And nobody could count or do averages until we had electronic recorders?

MarkW
Reply to  TheFinalNail
June 4, 2026 10:45 am

If you don’t have the data, you don’t have the data. Doing averages is meaningless if you don’t have sufficient daily data.

I’m guessing you actually are ignorant enough to believe that averaging the high and low for a day tells you anything meaningfull about what the average temperature was for that day.

Reply to  ToldYouSo
June 3, 2026 2:27 pm

And you really choose to imply the temperature instrumentation and temperature recording station sitings (locations) from 1850 to mid-1930s compared to those from mid-1930s to 2026 have comparable accuracy (say to be even within +/- 1 deg-C)???

No. There are higher uncertainty intervals. That is all taken into account. Read the literature. Maybe consider arguing about the methods used in the freely-published scientific literature, if you are so certain, rather than on blogs?

No one else has been able to do it.

Get yourself a Nobel Prize!

Reply to  TheFinalNail
June 3, 2026 2:44 pm

There are higher uncertainty intervals. That is all taken into account. Read the literature. 

Yeah, taken care of with the prescription that averaging a large number of stations and dividing the uncertainty by the (√# of stations) somehow eliminates all the uncertainty. That assumption is a gross improper use of the CLT.

Reply to  Jim Gorman
June 4, 2026 5:47 am

Who said the uncertainty was eliminated. HadCRUT publishes the uncertainties for every month in its record with every new monthly update.

Reply to  TheFinalNail
June 4, 2026 8:18 am

Who said the uncertainty was eliminated. 

I said it from reading the documentation. Find me one reference where there is a measurement uncertainty budget and where combined uncertainty is propagated through all the subsequent calculations.

Reply to  Jim Gorman
June 4, 2026 2:01 pm

It’s typical climate science trash. They don’t even bother to propagate instrument measurement uncertainty onto homogenization temperatures. It’s the same old meme: all measurement uncertainty is random, Gaussian, and cancels.

MarkW
Reply to  TheFinalNail
June 4, 2026 10:48 am

The vast majority of so called climate scientists.

Reply to  TheFinalNail
June 4, 2026 1:55 pm

HadCRUT publishes SAMPLING error estimates, not INDIVIDUAL STATION measurement uncertainty intervals. As usual for climate science, it is assumed that all measurement uncertainty is random, Gaussian, and cancels. Therefore it’s impossible for an objective observer to judge if the result of analysis of the data actually provides anything physically useful.

Reply to  TheFinalNail
June 3, 2026 3:31 pm

“That is all taken into account.”

So easy to say, so impossible to believe yet alone prove.

“Get yourself a Nobel Prize!”

Interesting attempt at a backhand insult . . . however, I don’t consider being awarded a Noble Prize as any great honor, since the award of such to terrorist Yasser Arafat in 1994,
followed by award of such to poser Al Gore in 2007,
followed by award of such to an unaccomplished Barack Obama in 2009.

Even a scientist and philosopher that I greatly respect, Richard Feynman, reluctantly accepted his 1965 Nobel Prize in Physics, famously calling the award a “pain in the neck” and “useless”.

Reply to  ToldYouSo
June 3, 2026 6:56 pm

Most surface station data from around the globe, shows that the 1930,40s was a similar temperature, or a bit warmer, than the first 2 decade of this century

Reply to  ToldYouSo
June 4, 2026 5:53 am

So easy to say, so impossible to believe yet alone prove.

Here’s a link to the published paper. Go and find the faults in their methods and publish your rebuttal in the peer reviewed literature, rather than expressing your opinion on a blog.

Interesting attempt at a backhand insult . . . 

Not intended as an insult; rather as a reference to the fact that no peer-reviewed comment or rebuttal to that paper has ever been published, anywhere.

Be the first!

Reply to  TheFinalNail
June 4, 2026 8:05 am

find the faults in their methods and publish your rebuttal in the peer reviewed literature, rather than expressing your opinion on a blog.

I don’t need to publish my findings at your demand. I only need to show you how this paper has failures. For your perusal, I have put several quotes from the paper here.

You should note that every reference here discusses systematic error/bias. The systematic errors/bias are deduced by assuming all measured data is 100% accurate with no need to include total combined measurement uncertainty as an important part of uncertainty analysis. Bias is defined by many international measurement agencies.

Bias is determined by calibration of individual devices and provides correction factors. Bias in this paper is considered as differences between data series. The fault here is that each series could be 100% accurate and the differences are simply related to inadequate identification of what caused the differences. That is not science, it is using statistical tools to achieve a distinction that is probably meaningless.

In order to obtain an analysis of systematic error by comparing different data sets, one must assume that measurement uncertainties are negligible. The basis for that is shown nowhere in the paper. In essence, every measurement is considered to be XX±0.0.

I could find no assessment of the measurement uncertainty or the development of an uncertainty budget anywhere within the paper. Without that, the variation in measurement quality cannot be assessed. through a correct propagation of uncertainty. That is a serious oversight in the paper.

This analysis included a new method for bias-adjusting station records, a process that is commonly known as homogenization, and combined estimation of homogenization adjustments with an independently developed spatial analysis method. 

A comparison to modern “instrumentally homogeneous” data sets by Hausfather et al. (2017), found that HadSST3 (Kennedy et al., 2011a, 2011b) and COBE-SST-2 (Hirahara et al., 2014) underestimated recent warming. 

The pervasive systematic errors, which have complex temporal and spatial correlations, are represented using a 200-member ensemble generated by varying uncertain parameters and choices in the bias adjustment scheme.

The error model for the noninfilled data set describes the estimate of temperature anomaly comment image at spatial location s and time t as a sum of the true temperature anomaly T(s, t) and three error terms: a bias term εb(s,t) representing biases with large-scale spatial and temporal structure; a partially correlated error term εp(s,t) for errors with typically local structure; and an uncorrelated error term εu(s,t) describing errors that are independent between spatial and temporal locations.

A detailed description of the land air temperature ensemble sampling method can be found in Morice et al. (2012). The sampling approach is designed so that the effects of known sources of residual systematic error in station anomaly series can be quantified for regional statistics and time series.

Mr.
Reply to  Jim Gorman
June 4, 2026 9:25 am

The pervasive systematic errors, which have complex temporal and spatial correlations, are represented using a 200-member ensemble generated by varying uncertain parameters and choices in the bias adjustment scheme.

The error model for the noninfilled data set describes the estimate of temperature anomaly comment image at spatial location s and time t as a sum of the true temperature anomaly T(s, t) and three error terms: a bias term εb(s,t) representing biases with large-scale spatial and temporal structure; a partially correlated error term εp(s,t) for errors with typically local structure; and an uncorrelated error term εu(s,t) describing errors that are independent between spatial and temporal locations.

A detailed description of the land air temperature ensemble sampling method can be found in Morice et al. (2012). The sampling approach is designed so that the effects of known sources of residual systematic error in station anomaly series can be quantified for regional statistics and time series.

I think Frank Sinatra had a similar hit song with this same theme –

“I Did It My Way”

Reply to  Jim Gorman
June 4, 2026 2:17 pm

This is all SAMPLING error, be it spatial or temporal. It is *not* MEASUREMENT uncertainty.

TFN has never understood the difference.

Statistical descriptors only describe the relationship the data elements have to each other. They tell you NOTHING about the quality of the individual data elements. The relationship the data elements have with each other is *NOT* measurement uncertainty. It’s why the SEM is not measurement uncertainty, it is SAMPLING uncertainty. Mathematicians and statisticians never seem to understand the difference.

Reply to  TheFinalNail
June 4, 2026 12:40 pm

” . . . rather than expressing your opinion on a blog.”

Ummmm . . . would that be your opinion as expressed here on the WUWT blog?

“. . . as a reference to the fact that no peer-reviewed comment or rebuttal to that paper has ever been published, anywhere.”

As I previously pointed out to you June 3, 2026 9:08 am:
“Absence of evidence is NOT evidence of absence.”
I guess you didn’t understand that.

ROTFL!

MarkW
Reply to  TheFinalNail
June 4, 2026 10:47 am

You claim that the uncertainty intervals are taken into account.
However when someone tries to claim that we have warmed a few tenths of a degree over the last 100 years, when the uncertainty intervals for records taken 100 years ago are several degrees, it shows that either you are lying, or they truly don’t know what they are doing.

Reply to  MarkW
June 4, 2026 12:37 pm

The real problem is that the so-called bias is used to determine a correction factor to make the series appear the same.

That is a totally different thing from determining the systematic uncertainty in the measurements. That can only be done through calibration, not comparing one to another.

Reply to  MarkW
June 4, 2026 2:24 pm

They honestly don’t care about measurement uncertainty. The ends justify the means. They will cheerfully make any assumption (guess) necessary to get the answer they have predetermined.

Reply to  TheFinalNail
June 3, 2026 1:50 pm

There is zero global warming trend over the first ~85-years of the global temperature record, between 1850 and the mid 1930s.”

Well, according to Phil Jones (you know of him, don’t you?) says after the end of the Little Ice Age the temperatures warmed three different times, all three warmings being of the same magnitude. The first warming was from 1850 to the 1880’s; the second warming was from 1910 to the 1930’s; and the third warming was from 1980 to the present day, with cooling periods after each warming.

All three warming periods reached a high temperature within just a few tenths of a degree of each other.

The bogus Phil Jones Hockey Stick chart you use to make your claim of no warming since 1850, show both the 1880’s and the 1930’s, as being similar in temperatures, within a few tenths of a degree of each other.

The Bogus part of the Hockey Stick chart is showing today’s temperatures as being hotter than the 1880’s, and the 1930’s. Today’s temperatures are Not hotter than the hot periods of the recent past.

This is what Phil Jones wanted to hide when he created this distortion of reality. You can’t have a decent climate crisis if current temperatures aren’t any hotter today than in the past. People would say, “where do you get the idea that CO2 warms the atmosphere enough to measure?”

So Phil fixed all that by demoting past warm periods so people like you can pretend we are experiencing unprecedented heat today.

If Climate Alarmists didn’t have the bogus Hockey Stick chart, they would have nothing at all.

I think what surprises me the most about all this human-caused climate change insanity since the 1970’s, is how the bogus Hockey Stick chart has supplanted reality in the minds of even smart, well-informed people.

The historical, written temperature records from all over the world show it was just as warm in the recorded past as it is today. No unprecedented heat.

The bogus Hockey Stick chart erases all the warm periods of the past and makes it appear that temperatures have gone higher and higher, decade after decade, and today is the hottest time in human history.

This is completely at odds with the written temperature records.

How does one get a chart with a “hotter and hotter and hotter” temperature profile out of written temperature records that have no such temperature profile?

Answer: You change the written record into Big Lies to sell a political agenda.

The Hockey Stick chart is a lie. The data does not support it. It is a figment of Phil Jones’ fevered imagination.

Believing in a Hockey Stick is just fooling yourself.

Reply to  Tom Abbott
June 3, 2026 2:33 pm

The first warming was from 1850 to the 1880’s; the second warming was from 1910 to the 1930’s…

But if you take 1850-1930s as a whole, there is net zero warming, globally.

If the observed modern global warming is just a linear continuation of an imagined “recovery from the Little Ice Age”, then how come we have an 80-year period of zero warming at the start of the ~175 year period of record then a big spike over the final ~80 years of it?

Reply to  TheFinalNail
June 3, 2026 6:43 pm

There is absolutely no way they have the data to fabricate that nonsense..

Data from land sites all around the globe shows the 1930s,40s was as warm as the start of this century.

Even Tree ring data from Biffa shows it.

Briffa-Tree-data-1900
Reply to  TheFinalNail
June 3, 2026 6:46 pm

The handle of the hockey stick is derived main from tree rings from trees where low CO2 was a constraint, making the tree ring data before about 1900 meaningless.

If you use non-tree ring proxies you get a much more realistic graph.

Lanser_holocene_figure11
Reply to  TheFinalNail
June 3, 2026 2:13 pm

HadCrut , GISS etc are FAKE. There were basically zero measurements for the oceans most of the period covered, and the land surface was measured, sparsely, in only a few select countries.

The map below shows where temperatures were measured up to 1920

tempmap
Reply to  bnice2000
June 4, 2026 6:12 am

Over the past 20 years UAH has been warming at a faster rate than both HadCRUT and GISS.

Is this another case of the surface data producers ‘hiding the incline’?

Or maybe you think UAH is FAKE?

Reply to  Mr.
June 3, 2026 6:20 am

Indeed . But the comment is pointing out that the temperatures have gone up. This is a different discussion than “why”…

Mr.
Reply to  Hysteria
June 3, 2026 6:29 am

Nah. It always devolves into the claims about manmade CO2 being the climatic “control knob”, and causing a “CRYsis”.

MarkW
Reply to  Hysteria
June 3, 2026 8:13 am

The temperature appears to have gone up. The problem is that the quality of the data is so bad, that we don’t know if the temperature is actually going up, or if it is just an artifact of the data.

Sparta Nova 4
Reply to  MarkW
June 3, 2026 12:56 pm

The real problem is there is no definition of optimum climate, let alone a definition with metrics that are measurable and testable by anyone.

We could have departed the climate optimum (I will not claim 1850 was the optimum) or we could be progressing towards the climate optimum.

All of that assumes a definition of a global climate optimum is valid or has any meaning at all. It is kind of hard to average the Sarah with Antarctica and get a meaningful result.

MarkW
Reply to  Sparta Nova 4
June 4, 2026 10:51 am

Historically life thrived when the Earth was warmer than it is today and struggled when it was colder.

Reply to  TheFinalNail
June 3, 2026 4:20 am

I don’t disagree that the site of the Stevenson screen hasn’t moved since previous records were set. In fact the UHI is included in the definition of a heatwave in London being 3 degrees Celsius higher than the definition of a heatwave elsewhere, specifically > 28 degrees Celsius in London cf >25 degrees Celsius elsewhere, including Birmingham, Manchester, Leeds and other major conurbations, for three consecutive days.
As I understand it the heat wave experienced across Western Europe was caused by an area of high pressure at altitude which increased the pressure below it by the downward flow of air, thus increasing the temperature through an adiabatic process.
When I was growing up in the 1960’s the national paper I looked at always listed the warmest places across different parts of the country, the one site that was invariably the warmest was Heathrow. Some of the issues with the siting of older weather stations in the U.K. was because the meteorological office came under our Ministry of Defence, which means that the measuring equipment was placed on RAF stations in the days of propeller powered aircraft which didn’t require huge runways. Of course those weather stations have remained pretty much where they were originally located, despite the changes to the aircraft type. Also, when Heathrow was chosen for the weather station it wasn’t the main London airport, which was actually Croydon.

Reply to  JohnC
June 3, 2026 7:26 am

Airfields typically require a local temperature measurement so that pilots can calculate their take-off distance. It was standard practice when I flew a Cessna to calculate the take-off distance using temperature, altitude, weight of fuel etc. Once when flying out of Los Alamos I had to rent a more powerful plane because of the altitude and temperature and the shortness of the runway.

Reply to  Phil.
June 3, 2026 8:02 am

The Met office weather station at Heathrow is no longer used for pilot information, apparently they now use their own network of weather stations.

MarkW
Reply to  Phil.
June 3, 2026 8:18 am

When coming back from a business trip in Milan, my flight had a layover in Madrid where we picked up passengers.
It was a hot day in Madrid and after the pilots had run the calculations they had to ask for volunteers to wait for a later flight as we had to reduce weight.
By the time enough people had volunteered and unloaded, we had missed our window for flying across the Atlantic and we had to wait 3 hours for the next available window.
By the time our window opened, it had cooled off enough outside that they were able to let the passengers who exited, back on.

Mr.
Reply to  MarkW
June 3, 2026 8:51 am

that’ll teach you for flying “ShitShow Airlines” 🙂

MarkW
Reply to  Mr.
June 4, 2026 10:52 am

I believe it was Delta.

Reply to  TheFinalNail
June 3, 2026 5:00 am

The Kew Gardens Stevenson screen is located in the same place as it was when the previous May records were set. All the other siting criteria remain the same too. No newly-documented warming influences have been identified.”

The site itself is *NOT* the microclimate associated with the site. If population and land use in the surrounding area affects the measuring station then it is MICROCLIMATE of the station that must be considered, not the latitude/longitude of the station itself.

Your assertion is basically a non sequitur, From Merriam-Webster: “logic : an inference (see inference sense 1) that does not follow from the premises”

The fact that “no newly-documented warming influences have been identified” is pretty damning for the reliability of climate science.

Reply to  Tim Gorman
June 3, 2026 7:13 am

The CIMO requirement identifies the surrounding microclimate as the surrounding 30m (see posting above), which in the case of Kew is a lawn!

Reply to  Phil.
June 3, 2026 8:10 pm

Right in the middle of the MASSIVE London urban heat dome.

MarkW
Reply to  bnice2000
June 4, 2026 10:56 am

A growing massive London urban heat dome.

Reply to  Phil.
June 4, 2026 2:35 pm

Does the color of that lawn change with the weather (wet/dry) or season? If so the microclimate for the station changes also. That natural variation becomes part of the measurement uncertainty budget – which for climate science is equal to +/- 0.

Reply to  Tim Gorman
June 5, 2026 6:32 am

If it does and that is typical of lawns in the area then it meets the requirement of the Class 2 rating.

Ground Cover: The surrounding ground must be covered with natural, low vegetation (less than 10cm) that is representative of the region.”

Reply to  Phil.
June 5, 2026 6:48 am

Meeting Class 2 rating doesn’t mean that there is no measurement uncertainty due to changes in the ground cover due to weather and/or season.

Reply to  Tim Gorman
June 5, 2026 9:15 am

The purpose of the weather station is to measure the temperature at that location, if the temperature changes because the lawn goes yellow then the instrumentation should measure that, how does that generate an ‘uncertainty’.

Reply to  Phil.
June 5, 2026 1:52 pm

Because the station is supposed to be measuring AIR temperature, not heat from the ground. ALL stations are contaminated with temperature effects that have nothing to do with the air temperature 6′ from the ground. Direct sunlight heats the station enclosure, part of that heat is transferred to the airflow of the atmosphere through the station sensor thus the measurement of the atmosphere temperature is inaccurate no matter how closely calibrated the actual sensor is. The station enclosure paint degrades over time due to UV effects, thus changing the absorptivity value for sunlight, i.e. the amount of heat the enclosure absorbs and transfers into the station enclosure. Insect detritus inside the enclosure or in the air intake will change the temperature profile of the airflow across the sensor – resulting in an inaccurate measurement of the actual atmospheric temperature. Component drift in the measurement and detection equipment from heating during use will affect the accuracy of the measurement no matter how accurately the actual sensor is calibrated.

If climate science were to do an ISO measurement uncertainty budget then all of this, plus MORE, would be identified and quantified. But climate science just refuses to do so. They would rather just use the meme of “all measurement uncertainty is random, Gaussian, and cancels”.

BTW, what happens to most materials when they are heated? Do most materials see some samples shrink and some expand – in equal measure – so the assumption of a Gaussian distribution is valid?

Reply to  TheFinalNail
June 3, 2026 7:59 am

Which type of thermometer was in use when the previous record was set, LIG, PRT? What was the duration? When did the site switch from manual readings to auto-reporting?

MarkW
Reply to  Right-Handed Shark
June 3, 2026 8:21 am

Whether the site itself has moved is only a part of the issue.
What has changed in the immediate vicinity of the site, plus what has changed in the neighborhood surrounding the site.

Like most alarmists, TFN likes to pick only the portion of the data that supports his belief system and ignore the rest of it.

Reply to  MarkW
June 4, 2026 10:08 am

“What has changed in the immediate vicinity of the site, plus what has changed in the neighborhood surrounding the site.”

Look at Google Earth,it’s hardly changed since 1945!

MarkW
Reply to  Phil.
June 4, 2026 10:53 am

We had satellites in 1945?

Reply to  MarkW
June 4, 2026 12:02 pm

No we did have reconnaissance aircraft though and the image from 1945 is shown on Google Earth.

Reply to  Right-Handed Shark
June 3, 2026 8:22 am

Not only that, where are the calibration records?

Robertvd
Reply to  TheFinalNail
June 3, 2026 8:05 am

And all that hot air came as a bubble from the Sahara where the Sun is at its highest most energetic point of the year. The same weather pattern that brought the hot bubble also brought the much colder weather as normal to places like the west part of Russia , Ukraine, Turkey, Syria
Lebanon, Israel and even Egypt. But not a word about it in the news.

Reply to  Robertvd
June 3, 2026 2:41 pm

But not a word about it in the news.

It was all over the news I saw.

And what… are we saying that May 2026 was the first May in ~145 years of temperature measurements at Kew Gardens that atmospheric conditions permitted warm air from North Africa to arrive in southern England?

Hardly.

MarkW
Reply to  TheFinalNail
June 3, 2026 8:08 am

If most of your stations are junk, then the results are junk. Whether the average is going up down or sideways, it’s still junk.

Reply to  MarkW
June 3, 2026 11:25 am

If most of your stations are junk, then the results are junk

Why is that so hard to understand and to admit? Way too many folks in climate science have no idea what that means. If I keep a micrometer in the bottom of my toolbox in the back of my pickup, any measurements I take are junk. I simply can’t justify them being anything but junk no matter how many statistics I throw at it.

Reply to  MarkW
June 3, 2026 2:44 pm

If most of your stations are junk, then the results are junk. Whether the average is going up down or sideways, it’s still junk.

If” is doing a lot of heavy lifting there, MarkW.

As mentioned above, the fact that the USCRN stations are waring at a faster rate than those used by nClimDiv is a major problem for your hypothesis, and one which WUWT seems curiously reticent to address.

Reply to  TheFinalNail
June 3, 2026 5:20 pm

Since nClimDiv is an adjusted, fake series of numbers.. that depends ONLY on the adjustments made…

… your comment is utterly meaningless.

MarkW
Reply to  TheFinalNail
June 4, 2026 10:55 am

I take that back.
Since most stations are junk, the results are junk.

Does that satisfy you?

Reply to  TheFinalNail
June 4, 2026 2:44 pm

It’s been addressed here. You.just refuse to read anything that questions your preconceived belief.

Even USCRN is a SHORT record. 30 years is simply not long enough to capture temperature cycles adequately. Meaning it’s actually impossible to know if CLIMATE is actually changing ir if it is merely weather that is changing cyclically. Nyquist you know!

gyan1
Reply to  TheFinalNail
June 3, 2026 1:14 pm

“This article completely misses the point that the new May record is still a new May record”

The point you completely miss is that because of the extreme variability of weather, temperature records are often broken somewhere both hot and cold. The only meaning they have is validation of how crazy weather is. The easily manipulated have been duped into believing that the wide swings of natural variability is something new rather than a fundamental feature of climate and weather.

Reply to  gyan1
June 3, 2026 2:47 pm

The point you completely miss is that because of the extreme variability of weather, temperature records are often broken somewhere both hot and cold. 

How many monthly temperature records at 145-year old weather stations are broken by +2.0C then +2.3C over consecutive days?

gyan1
Reply to  TheFinalNail
June 3, 2026 3:07 pm

“How many monthly temperature records at 145-year old weather stations are broken by +2.0C then +2.3C over consecutive days?”

Every time an anomalous heat dome sets up. CO2 is a negligible factor when that happens.

Reply to  TheFinalNail
June 3, 2026 2:09 pm

As has been mentioned MANY times, nClimDiv is a product OF ADJUSTMENTS..

It has no other meaning except how well or badly they did those adjustments.

MrGrimNasty
June 3, 2026 4:17 am

The London conurbation (and other metropolitan areas to a lesser extent) is so massive it not only creates its own climate, it affects a large area of the surrounding country depending which way the wind blows.

You could end up twisting yourself in philosophical knots trying to imagine what the climate would be if measured in a landscape that no longer exists.

MrGrimNasty
June 3, 2026 4:28 am

Story Tip(UK)

Brighton council has been waging war on stoves, ASA upheld complaint about the lack of evidence backing up their insinuation.

https://www.dailymail.com/news/article-15866507/Woke-Labour-run-councils-adverts-against-wood-burning-stoves-banned-providing-no-evidence.html

Being coastal, most of the time Brighton’s air quality depends on wind direction, oh, PMs include salt spray!

They make a lot of daft claims about air quality to justify their war on traffic too.

strativarius
June 3, 2026 5:14 am

The Wednesday funny. Remember the Hood in Thunderbirds? Mayor of London Sad Khan has just been on his ‘second’ hajj.

comment image

Reply to  strativarius
June 4, 2026 5:57 am

What on Earth has befallen Sadiq Khan recently? He has aged 30 years, and seems to be starving to death.

Reply to  Graemethecat
June 4, 2026 7:36 am

Exactly my thoughts when seeing this photo.

Robertvd
Reply to  Graemethecat
June 7, 2026 5:59 am

The Will of his God.

June 3, 2026 6:18 am

Re UHI. I always smile to myself when a UK weather forecaster says, dead pan “it will be x degrees in the countryside, and a couple of degrees warmer in the cities”

Sometimes they even add the magic words “due to the urban heat island effect”

June 3, 2026 7:04 am

UK might have experienced heat spikes this Spring.
But 80 Degrees north in the Arctic during the month of May temperatures have definitely been running a fair bit colder than average. https://ocean.dmi.dk/arctic/meant80n.uk.php
Looks like cold is very much locked up into the higher reaches of the Arctic currently, which might explain why less cold air from the north wasn’t able to oppose the hot air that migrated into our shores from the Sahara. Mostly, temperature is influenced by prevailing synoptics more than anything else

MarkW
Reply to  Neutral1966
June 3, 2026 8:23 am

Looking at all the available data.
That’s just not how climate science is done.

June 3, 2026 7:48 am

The nearest ‘clean’ station to Kew is Chertsey, Surrey.. or at least it used to be before they surrounded the sensors with a few hundred solar panels. If you check Chertsey now, you are given “We are showing you the observations for the nearest location to Chertsey (6.9 miles, 9 m higher).” HEATHROW. Leafy, rural Surrey is now as hot as Europe’s busiest airport.

Clipboard-1
Reply to  Right-Handed Shark
June 3, 2026 2:24 pm

Yep Chertsey is now just another of the Met’s many “totally unfit for purpose” sites.

Chertsey-2
June 3, 2026 8:56 am

Let’s see . . .observed temperature variation of right around 2 deg-C just over the area of London, as scientifically documented in the above article.

Based on the IPCC’s warnings about a 1.5 deg-C rise in temperature and “tipping points”, shouldn’t the greater part of London now be uninhabited desert?

Robertvd
Reply to  ToldYouSo
June 3, 2026 2:42 pm

That’s why these days it is mostly inhabited by people from warm places. The indigenous tribe has more or less moved away. Even the major is a Mecca loving man.

Sparta Nova 4
June 3, 2026 12:47 pm

I wonder how much of this would exist were it not for the climate alarmism mountain to climb.

Bob
June 3, 2026 4:08 pm

I would give the Met one week to stop using class 4 and 5 stations. If they didn’t I would fire all the top managers politicians included.