From Dr Roy Spencer’s Global Warming Blog
by Roy W. Spencer, Ph. D.
SUMMARY
- Over the period 1900-2023, the average summer (JJA) daily high temperatures across the six southernmost large provinces of Canada (British Columbia, Alberta, Saskatchewan, Manitoba, Ontario, and Quebec) show no trend.
- The average of the 3 hottest days’ in each month month show a slight downward trend, while the 3 coolest days’ average temperature per month shows a slight upward trend.
- Recent years have generally averaged as warm as was experienced in the 1920s to 1940s, with 7 of the 10 hottest summers occurring in the 1930s.
- Results for the 6 provinces separately are also presented.
Introduction
Given media reports, it is likely that most Canadians think they have been experiencing unprecedented summer warmth in the last couple of decades. But this isn’t true.
Below I present analyses of daily high temperatures (Tmax) from all available stations in the 6 southernmost large provinces, based upon the daily Global Historical Climate Network (GHCNd) dataset. These are the 6 provinces that border the Lower 48, and contain 86% of Canada’s population.
I simply averaged together the relevant statistics (monthly average Tmax, average of the warmest 3 days in each month, and average of the coolest 3 days in each month) from all available stations. Each station had to have at least 90% of the days in a month reporting data for that month to be included in the analysis.
Since stations come and go over the years, and since there are some large terrain elevation variations in western Canada, I performed an elevation correction to these Tmax metrics, in all provinces, using the departure of each year’s station-average elevation from the all-year (1900-2023) station average elevation, using a lapse rate of 6.5 deg. C per km. Corrections for average changes in station-average latitude were not done, which might be necessary in the winter since there are large North-South gradients in air temperature then. Such corrections in the summer would likely be small, but I can revisit that nuance at a later time.
Results
I’ll start with the 6-province average Tmax temperature time series, along with the total number of stations available in each year. In all plots that follow, I list the linear temperature trends, but plot a 3rd order polynomial fit to the data which captures the dominant feature of relative warmth in the 1920s to 1940s and in the most recent decades, but relative coolness in the intervening decades. In all provinces the number of stations increases from 1900 to the 1970s, then decreases substantially in recent years.
As can be seen in the first plot (averages for all 6 major provinces), there has been no long-term linear trend in the average summertime Tmax (0.00 deg. C/decade), a small downward trend in the 3 hottest days per month (-0.02 deg. C/decade), and a slight warming trend in the 3 coolest days per month (+0.03 deg. C/decade). Relative warmth around the 1930s is evident, as well as warming in recent years.
I have also annotated 2021, which experienced the extreme heatwave in late June in western Canada. While that pushed the hottest 3-day average Tmax metric (red curve) to the highest average temperature of any year since 1900, the 3-month (all-days) average summer Tmax temperatures was very close to other years (3rd place, behind 1961 and 1919).
Notably, 7 of the 10 hottest summers occurred in the 1930s.
Individual Provinces
The results for the individual provinces follow. I present them without comment; my Canadian friends can peruse the results for their home province if they wish. These are presented from West to East:
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
I wonder if internationally-lauded central banker and now Canadian Prime Minister Mark Carney with his proclaimed proficiency with numbers can now conclude that Canada can put aside its concerns about global warming, and concentrate on helping citizens with cost of living crises.
(Nobody should hold their breath for this segue into reality & rationality from such a world governance acolyte)
Now can we send back to Canada the climate science crisis professor ( preacher) at Texas Tech.
The main reason that some, actually a minority, of Canadians believe that their summers are growing warmer is that they’ve been exposed to the non-stop climate alarmism peddled by the domestic media, especially the CBC. That entity, which receives its funding from a left-leaning government, never misses the chance to attribute any weather anomaly to manmade industrial activity, especially the use of fossil fuels. Except when any news or information appears that refutes its claims, it makes certain to suppress it or ignore it entirely. Meanwhile the Canadian public obviously hasn’t bought any of these doomsday scenarios since it rejected the much-reviled carbon tax in the most recent federal election and has shown few inclinations to make any big lifestyle changes to combat the mythical climate crisis. One classic example has shown that despite billions the federal government has poured into mass transit initiatives, Canadians are using their private vehicles even more now than ever before. So the reality is that the Canadian public, which is accustomed to wide variations of temperatures, is not going to lose any sleep over a few more degrees one way or another.
I’ve been to Canada quite a bit, in all seasons. I never recall being hot there but the coldest I’ve ever been was in Fort. McMurray in January. Before that, I thought Edmonton was cold.
I skied in Banff once, but it was also cold, too cold. Skiing in Colorado is generally better if only for comfort. Even the time I skied in Revelstoke in spring it was cold enough for me, but that is a nice mountain.
In Canada, it takes up until July 31 before it’s warm enough to swim in Lake Stuperior and on Aug 1 it starts to cool down again.
Global Warming? Bring it!
But in the UP in August, my Lab found it quite pleasant. Just another advantage of living South of the border.
I wonder how much warming since the 1980s is due to urban effects ?
And I’m pretty sure that Canada is one place that could use a little bit of warming , anyway. 😉
All of it, apart from the unknown effect of a chaotic atmosphere. In my worthless opinion, of course.
Interesting to note from the graphs that warming only seems to happen when the number of stations is reduced…
… probably leaving ones that are mostly in the more southern urban populated areas.
Which makes me question the endless discussions about statistical probability, error bars, sigmas, and all the other diversions supposed to convince us that adding CO2 to air makes thermometers hotter.
That’s a load of rubbish, of course.
This is the usual dumb stuff of averaging temperatures, not anomalies. It never works. The reason is that the stations in the mix cange over the years, so the temperatures drifts as the places sampled are colder or warmer places. It isn’t measuring whether any one region is getting warmer or colder.
In the case of Canada, there is continued development of northern regions. More northern regions enter the average over the years. The average gets colder because of those added cold places. I see that Roy has corrected for change in altitude, but not latitude, which is strange.
Yeah Roy Spenser does dumb stuff! Nick Stokes is the true expert Lmao.
Yes. Here is GISS saying it is the wrong thing to do. Here is NOAA.
Nope, It is GISS that is doing the wrong thing..
All part of the climate scam.
Well, GISS’s director is on record saying a 38% likelihood of an event occurring was “near certainty”.
NOAA says “Ocean currents can be caused by wind, . . . “. In an alternative universe, I suppose.
Are you more gullible than ignorant, or vice versa?
Chuckle. Of course they do. Narrative support is hardly surprising.
Note that Roy Spencer has now admitted the problems and produced revised results.
https://www.drroyspencer.com/2025/12/canada-summer-temperature-trends-1900-2023-part-deux
You could have posted the new results.
Now that you know this and since it is NOT unique to Canada, maybe you could start paying more attention to data that is hidden beneath averages. Things like diurnal variance, regional trends of Tmax and Tmin, and hemispherical differences.
A posted a link to the article. I assume people are capable of clicking on the link. I also assume the WUWT will be posting the updated results in due time.
The point of my comment was that as Nick Stokes pointed out, the original analysis was just bad. In this thread many spent a lot of time telling Nick he was wrong – so it’s amusing when the author confirms it was wrong.
I make no comment on the revised result – beyond the fact it seems more in keeping with other observations.
In the revised analysis all provinces show a positive trend in summer maximums with a trend of +0.06°C / decade, and minimums are warming three times faster. Compared with original claim of 0.00°C / decade warming for maximums.
I still think it’s a mistake to apply a linear trend over the entire 123 year period, when the changes are clearly not linear. And I wish he would actually show the uncertainty in the trend.
“Now that you know this and since it is NOT unique to Canada, maybe you could start paying more attention to data that is hidden beneath averages.”
Why do you keep pretending I don’t do that? Not in this comment section alone I’ve presented data on CET and Germany.
Do your own work and present it at WUWT.
I did it extensively here when Andy May did the same thing with SST. He averaged temperatures, where the number of data from cold places inceased over time, as it would with Canada. I showed with detailed calculation that Andy’s result was purely a result of the change in stations average climate, not any places themselves getting warmer or cooler.
It’s just a rookie error.
It’s a pity that you can’t provide a reproducible experiment that shows that adding CO2 to air makes thermometers hotter.
That means that all your “calculation” are completely meaningless. Maybe you play with your pencil too much?
Ignorant and gullible.
Roy’s calculation is about surface temperature. It tells nothing about CO2. And he got it wrong.
Nick, earlier you said –
As I mentioned before, climate is the statistics (including averages) of weather observations. “Changes in station’s average climate . . .” is the witless utterance of someone who doesn’t realise that they’re talking about the average of statistics which include averages.
As you say, just a rookie error. Typical of an ignorant and gullible person who believes that adding CO2 to air makes thermometers hotter.
Have you considered quitting while you’re still behind?
Surface temperature have nothing to do with CO2..
WOW well done Nick.. you finally got there. 🙂
And no, Roy’s calculation is correct, except for not accounting for a large amount of urban warming.
Average baverage blabberage. Please explain how you twerps manage to extract increasingly more detailed data with finer and finer decimal accuracy, with perpetually increasing confidence and bravado, from fewer and fewer and fewer stations.
Creative accounting meet pseudoscience?
Sounding a bit like Captain Queeg there, Nick.
“the stations in the mix cange over the years”
Except everyone still lives in the nearest 100 miles to the US border.
He hasn’t accounted for urban warming either, so the real temperatures probably drift downwards after about 1970.
Your comment drifts off into its normal irrelevance. !!
Edmonton is 500 km from the US border. Athabasca, with huge new development, is 140 km further north.
Well, gee. That’s good to know. Have you any more scientific gems to cast before us?
lol, two towns.. when the increase in population over that period is from 5 million to 35 million.
Do you really deny that a large majority of Canadian growth and development is close to the USA border… seriously !!
Here is a map of Canada’s population density, Edmonton at red arrow..
Athabasca’s “huge development” has a population of 2,800 and dropping in the small town and just less than 7000 in the region.. So funny !!
Athabasca Population 2025
I’ve been to some of the yellow part….nothing but forest, rocks, lakes, and peat bogs we call “loon shit”…if your bush plane or skidoo stops producing CO2, finding your remains is questionable…west of the Rockies was all canoed by French fur traders and their Indian guides who knew the next 40 miles of river, along with most of the US 300 years ago….so big and empty they mostly didn’t bother claiming it for France…
LOL Detroit Lake averages colder than Edmonton Alberta even though it over 600 KM south of Edmonton. Geology play an important factor in what the average temperature is, north or south does not rule always, have you not learned that yet. I know so called “climate Scientist: have not. It no surprise you haven’t. Hint Detroit Lake Minnestupid(my birth place) does not benefit from Chinook since it to far east. My home town only 37 miles north of Detroit lakes is even colder even though is on a short distance from Detroit Lakes.
Detroit Lakes is not in Alberta, or even Canada. The point is the northward drift of reporting weather stations in the various provincial averages.
Station count in BC dropped from 400 in 1990 to only about 150 in 2020.
The ones dropped will be a colder less accessible one.
Athabasca? LOL. From 2023-2024 it saw a population decrease of -0.21%.
Of course it could increase 100% to the positive and it would be completely irrelevant. The population of Athabasca, AB was 2894 in 2024.
Athabasca is an oil region. It has towns like Fort MacMurray, which grew from less than 1000 people, to over 70000 now. But it actually isn’t population that counts, here. It is number of weather stations reporting. That is what is skewing Roy’s analysis.
No, it is skewing YOUR calculation, because most of the colder ones would be the ones closing
And arguing against actual data .. which shows Athabasca only has a population of 2800 and falling.. WOW. !
And do you really think that imaginary population growth of yours isn’t going to have a large warming effect on the measured temperature in the urban area. WOW again. !
Except that’s not what happened! In the early ‘warm period’ on Spencer’s graph the few stations that existed were clustered near the border with very few further north. In the modern period the number of northern stations increased significantly.
See for example for British Columbia: https://services.pacificclimate.org/met-data-portal-pcds/app/
Station count in BC dropped from 400 in 1990 to about 150 in 2021.
Most of those removed will be in the less accessible colder areas.
No if you look at the source I provided you’ll see that most of the stations that were removed were near the US border, not in the north.
To reinforce that, from that link below left are stations reporting at times from 1900 to 1925. On right are stations reporting 2023-2024
Here is a plot of the climatic average of the stations reporting in each July in Alberta. That is, the average of the all time means of those stations. As you include colder places the average goes down. There is no weather information here, just the constant climate stae and whether the station is in the sample or not. Note that the variation is substantial; in 2025 stations in the sample had long term averages about 0.6C colder than in 1900. That’s because there are more stations further North.
And here was Roy’s graph. It goes up and down in much the same way, although there is actual warming superimposed:
You have obviously “lost the plot”, Nick. Climate is the statistics of weather observations, so calculating the “average” of those statistics (which include “averages”) just highlights your ignorance and gullibility.
Or are you just denying reality as a result of insanity?
Swenson,
What are you braying about?
I’m sure I don’t know. Why don’t you tell me? Are you one of those ignorant and gullible people who believe that adding CO2 to air makes thermometers hotter?
That would explain why you are making such a bizarre comment, I suppose.
Swenson,
For once you make sense –
“I’m sure I don’t know.”
Why should anyone try to play riddle with someone who used many sock puppets over more than 15 years to push his silly idea that the Earth’s surface temperatures come from its inner core?
Cheers.
I’m sure I don’t know that, either. Why do you try?
I should point out that that “. . . his silly idea that the Earth’s surface temperatures come from its inner core . . . ” is a fantasy you just created. I have said such a bizarre thing, which is why you were unable to quote me.
You believe, like many other ignorant and gullible people, that adding CO2 to air makes thermometers hotter!
Even Google AI agrees with me –
and Google AI even accepts that the atmosphere contains CO2 and water vapour!
Swenson,
You ask –
“Why do you try?”
What makes you think that I try, and do you realize that “Google AI” is a company, not a chat bot?
Chat bots may not help you with your “inner core” theory.
Cheers.
Seeing as how I don’t have an “inner core theory”, it is obvious that you are one of those ignorant and gullible cultists who believe that adding CO2 to air makes thermometers hotter!
Feel free to keep being dimwitted, and making stuff up. I don’t mind.
Swenson,
You pretend – “seeing as how I don’t have an “inner core theory””
Have you already forgotten:
https://www.drroyspencer.com/2022/11/canadian-summer-urban-heat-island-effects-some-results-in-alberta/#comment-1404739
or are you just feigning ignorance?
Notice the theme of Roy’s post.
Cheers.
Wee Willy Dimwit, you write –
and then proceed to link to an article about the Moons core, written by someone else! Pretty stupid, aren’t you? You seem to be losing your already tenuous grip on reality, completely. A major indicator of insanity.
Why should I? Why should I take notice of commands from a madman? Would you?
If Roy Spencer, like you, believes that adding CO2 to air makes thermometers hotter, the he is as ignorant and gullible as you, in that regard.
In the meantime, carry on. I find your antics quite humorous.
Where is your adjustment for Urban warming?
Or do you pretend it doesn’t exist. !
Congratulations you’ve finally recognized that the period from1920 to about 1945 was remarkably warm (by modern comparisons) to current day Alberta. The trend you’re missing is that from 1945 to about 1970 Alberta was in an unusual period of cooling.
Also there are no further increases in stations to the north. Environment Canada and Climate Change, the Canadian government department for monitoring Canada’s weather has decreased the number of stations dramatically (-50%) over the last twenty years.
Continued northern development of northern regions? Really? Where?
Edmonton is the only larger city living away from the U.S. border. 66% of the population lives within 100 km (60 miles) of the border. 90% of the population lives within 160 km (100 miles). Edmonton’s projected population by 2040 is 2 million people.
You want to talk about dumb stuff, the fractional increase in northern development contributing to anything climate related falls easily into that category. Please don’t believe the hype about Churchill becoming a major export port. The Bay still freezes for significant periods of time, meaning nothing moves out of there for extended periods, never mind the utter lack of infrastructure.
That’s all without acknowledging that Canada has reduced its surface stations by well over 50% in the last 20 years.
And most of those would have been in colder, more remote regions…
… making Nick’s comments and calculations even more meaningless and irrelevant.
It isn’t development changing climate. It is just causing more weather stations to be located in northern places, which brings down the kind of averages that Roy is calculating.
No evidence of that.. Urban areas in the south have expanded rapidly.
That is where most of the remaining surface stations are.
It is far more likely that the last 30-40 years of measurements is getting artificially raised by urban effects.
You yourself said that Edmonton had grown a lot. !
There are most certainly no new stations in colder regions.
The whole station count is decreasing, almost certainly losing colder more remote stations.
Any newer stations will be very much in the massive expansion of population near the US border.
As shown above in the case of BC it is the proportion of northern stations that is increasing as the number of stations near the US border have substantially decreased!
Can’t let go of anomalies for any reason can you. Anomalies basically create values that are one, if not two, magnitudes smaller than the absolute temperature. Creatively reducing the magnitude makes variance and its derivatives smaller by the same amount. God forbid you have to give up the fake accuracy and variance when looking at absolute temperatures.
Averaging absolute temperatures over a region, be it a city, county, or province, country, etc. provides a view of what that regions temperature is. The fact that there are cooler stations along with warmer stations doesn’t invalidate the average. The average is what it is. Notice that Dr. Spencer’s work only claims to show that certain categories of average temperatures have not changed appreciatively.
If stations are closed randomly, then one would expect that both colder and warmer stations have disappeared equally. That means the results of an average will not have been changed to a large extent. The CAGW advocates here like to claim that averaging stations cancel both random and systematic errors when averaging various stations. Are you claiming that this assumption is no longer valid?
Lastly, I recently obtained a weather station that will send temperatures and other weather data to Weather Underground. I am attaching a screenshot of the temperatures surrounding my area. These are not necessarily calibrated, but even assuming a ±3F uncertainty, there is a vast range of temperatures shown. There is a low of 21 and a max of 30. Adding 3 to 21 and subtracting 3 from 30 provides a range of 24 to 27 degrees. That is in an area about 10 mi by 15 miles. This is part of the variance in surface temperature that is always ignored by climate science. It is never propagated forward when averaging all the way up the GAT.
You want to average Tmax and Tmin? What is the variance of this distribution. Why is it dropped when doing the next average? Why do cold stations and warm stations matter when the variance is ignored anyway?
….duhohh….averaging temperatures are how anomalies are determined….it’s the reverse of a “never works” situation….temps are bedrock, anomalies are sand dunes….
Find the mouse but ignore the elephant. A consistent drop-off of nearly 2/3 of reporting stations across the region seems rather a significant consideration, but you make no mention of it. Curious.
Yes, just another example of how it was just as warm in the recent past, when less CO2 was in the air, as it is today, with more CO2 in the air.
This is the case all over the world.
Conclusion: CO2 has had no noticeable effect on the Earth’s temperatures.
I wouldn’t go quite that far, but the effect may be quite irrelevant, as the outcomes of changes to the inputs of chaotic systems are quite unpredictable.
Google AI –
From memory, he stated that the CO2 “insulation” was equivalent one seventh of an inch of polystyrene (whatever he meant), but I’ve never seen his reproducible experimental data, so I assume he just made stuff up.
I think he’s nuts, and he might well think the same about me. He’s obviously a bit strange, because he thinks that adding CO2 to air makes thermometers hotter. He doesn’t know how, or by how much, or where this miracle occurs, but who cares? He’s a “leading geophysicist” – leading the ignorant and gullible along the road to nowhere!
Petty-humbert is well known for “making things up”.
iirc.. it was he that pooped out the 2C warming meme.. later changed to 1.5C
I wouldn’t go quite that far”
Really? Where’s your evidence that CO2 *has* caused a noticeable effect on Earths temperatures?
I’m surprised at this from you since you say all the time: “that adding CO2 to air does not make thermometers hotter”
It seems a little contradictory.
Tyndall showed that adding CO2 to air makes thermometers colder. Is this “noticeable” in the temperature of thermometers on or near the surface?
I don’t know how much, in measurable terms. I don’t believe anyone knows.
Sorry if I sounded contradictory.
I can say without doubt that in my 65 years I have not noticed it. If no one told me the Earth was ”warming” since I was kid, I would not know it. Yet they tell my it’s panic time.
So have I in my 72 year I do know this is that the year‑to‑year swings make the so‑called ‘global warming’ look like a rounding error. Back in the 1960s, I’d ask my mother when summer was finally going to warm up in late June. Her answer? It might not — it could just be one of those cold years.
And this is coming from a woman who grew up in the 1930s, the hottest decade in U.S. history. Even she understood that weather changes constantly and nothing is ever ‘normal.’ But sure, tell me again so called “climate scientist” how a few warm years is proof of impending doom
The weather does change constantly, but it also has a reliable pattern, and temperatures have not varied more than about 2.0C (hot to cold) since the Little Ice Age ended in the 1850’s.
The Earth’s climate is cyclical. It warms for a few decades and then it cools for a few decades and then it repeats the process and variations take place within this narrow band of temperatures.
This cyclical movement is visible in all the unmodified, regional, historical, written temperature records.
It is amazing to me that so-called scientists have ignored this fact for years. It’s right in front of their eyes, as with the graphs in this article. I guess they have been mesmerized by Michael Mann and Phil Jones lies about the temperatures in the past. They think the Hockey Stick global chart represents reality, and then deny reality by dismissing the written temperature record.
Human psychology is complicated.
I agree, Mike.
Agreed. I have lived in Texas, mostly Central Texas, all my 72 years. I have sensed no noticeable change in Texas climate. Some summers are oppressively hot, while others are quite pleasant, often back to back. The hottest day I have experienced in my area was 114° F in September 2000. On the other hand, we have had days in July and August when the temperature did not even reach 90, “cold” by Texas Summer standards. Likewise, precipitation and drought are highly variable.
In most winters, there are many days that I can go outside in shorts and a T-shirt, and the first freeze, normally in late November, can be as late as December or even January. Other years, we have already had snow, a rarity, at Thanksgiving.
When I was a kid, one of the stupidest Christmas gifts for a child was a sled, which could never be used.
btw, I am a degreed meteorologist (BS) with a masters degree in soil science. I never practiced as an operational meteorologist, but I am within two weeks of completing 50 years as an environmental professional (26 yrs as a researcher and consultant to government and industry; 24 years as environmental program manager for one of the largest university systems in the United States). I have also spent much of my life in outdoor work and play, so I know what un-air-conditioned/heated life feels like.
So we may just have to forgive Nick for his youthful ignorance and lack of manners. Like most “kids” these days, he probably seldom gets out from behind his computer screen to experience actual nature and weather.
Here in Salt Lake City, the winters in general do seem a bit less bitterly cold, especially in February, than in the 1970s; the snow more often restricts itself to the mountains where it belongs. It is raining here now, while a few thousand feet up, they are getting a two-foot dump of the white stuff. Certainly nothing I would call a “crisis”.
Mike, I got you by 13 years, and have to say my observations match yours, in spite of the extra decade. To the climate hoaxsters, that would mean I am wronger than you…
Over the period from 1900-2020 the Canadian population grew from 5 million to over 35 million.
If that doesn’t affect urban temperatures, I don’t know what would.
Frankly, I find it quite amazing there hasn’t been a lot more warming measured at surface sites.
Nah, it’s well known to climate scientists that heat doesn’t make thermometers hotter. That’s due to CO2.
/sarc
It’s pretty obvious. If you move a thermometer closer to a fire, it gets hotter. Closer to the CO2 produced by burning stuff, you see. Also, bigger fires produce more CO2. That’s why you can feel the heat from further away.
Leading climate scientists agree, so it must be true.
/sarc off
Lytton in BC has often made headlines for setting temperature records, notably in one recent year, and often touted as an indicator of climate change. I have some knowledge of Lytton’s weather stations as one of my brothers was a met tech there in the mid 1960’s, back when humans still read thermometers and reported the weather.
Lytton has had 3 different station locations that I am aware of; the first that my brother reported from was beside the CPR station downtown and immediately above the Thompson river canyon. While he was there a new station was constructed beside Highway 1 across from the Lytton airport, a hundred meters higher in elevation and over half a km away.
When it became automated the new station property was sold and it was moved south to a spot above a long SW-facing gravel slope to the Fraser river which bakes nicely in afternoon summer sun–and produces an upslope wind in response that carries that hot air right over the monitoring location.
Lytton is situated in the deep narrow canyons of the Fraser and Thompson rivers and naturally records extreme weather; my brother recalls one day where he recorded ‘clear skies and rain’ as the strong up-canyon wind from the Fraser blew rain around the corner of the valley and into the rain gauge, but no clouds were visible from the station location. But the Lytton record supplies better information for location change than it does for climate change.
They pulled the same stunt in Detroit Lakes, Minnesota. The weather station used to sit in a low‑lying, tree‑sheltered area — you know, the kind of environment that naturally regulates temperature and has moisture‑holding ground nearby. Then they moved it up to the airport: higher elevation, no water‑retaining areas, basically open prairie. And just to make the data extra ‘reliable,’ it’s now sitting within feet of air‑conditioning units.
But sure, I’m supposed to believe the temperature increase there is all because of CO₂. LOL. If this were a one‑off, maybe. But as Anthony’s studies show, this kind of station relocation, bad siting, and outright removal is the norm, not the exception.
I’m reading John Vaillant’s book “Fire Weather” about the 2016 Fort McMurray fire. There is just a little bit of “global warming” stuff in the first 17 chapters. Yesterday I started Chapter 18. He has now gone off the rails about climate change. I will have a hard time finishing.
JH:
[also O/T]
Just finished Erica Gies’ book “Water Always Wins” (2022) about water issues around the world. Lots of interesting stories how different nations and local farmers deal with drought. Some are quite ingenius [google Peruvian “puquios” ]
But the book is marred by lots of climate crisis comments concerning extreme weather to which I, in pencil, annotate the margin with “Nope! see AR6 Table 12.12”, and then move on.
In doing so, I still learn things yet don’t give implied assent by remaining quiet while reading “misinformation”.
Thanks for the work on Canadian stations eh!! Here is a nice study on airport microclimates in Canada: https://www.mdpi.com/2674-0494/1/2/12#B21-meteorology-01-00012
The authors describe a 1970 reference on airport “microclimates”. Meteorological station microclimates is an issue WUWT has championed for years. Worth considering is all them kids generating climate scenarios are born well after e.g. 1970. The modelling community is unable to grasp the concept of a local signal in a meteorological record and blindly churns these tarnished records into their workflows. Also there is a high likelyhood of highest priority continuous operation of airport stations for the obvious navigation information needs. With the abundance of “adjustment and homogenization” applied to building a climate dataset, it is likely the airport “signal” biases those less more remote locations that need some infilling.
A few observations, some of which I raised on Spencer’s original post.
I think the headline is a bit misleading. It says this is about summer daily high temperatures, which would imply the average of all summer days. But Dr Spencer is only looking at the hottest days in each summer.
I’m dubious about the metric Spencer uses here – the average of the warmest three days in each calendar month. It’s not really clear what the purpose of this is. If you are only interested in the hottest part of any summer, why split it between arbitrary months?
My main objection is that this could lead to arbitrary results, e.g. if there is a hot six day period in one summer, and those six days all fall in July, only three of those days count. But if the same temperatures fall split evenly between July and August, all the days count. In the long term I doubt this makes much difference, but it does feel a bit random.
But the biggest issue, as Nick Stokes says, is that this is just averaging all stations, regardless of changes in location. For one thing this means you bias the temperatures to where stations are, and so a heatwave over a location with a lot of stations will give a higher average than if the heatwave was over a sparsely populated area. And then there’s the problem of changes in distribution of stations over time.
It would be interesting to see the results of this method applied to areas away from North America.
Here’s my graph using the same 3 hottest day in each month, applied to the daily CET TMax values.
The trend since 1900 has been +0.19 ± 0.08°C / decade, but clearly this is not a linear increase. The trend since 1960 has been +0.47 ± 0.20°C / decade.
CET….. That’s funny!
A load of sites that have been proven to be TOTALLY UNFIT and CORRUPTED for the purpose measuring basically anything.
Let’s look at a nearby site that is as untainted as possible. (use blue data)
Average temperature in the 1930s, 40s higher than than 2000-2020
Following on from Spencer’s TMin data for Canada, here’s a comparison of the same 3 days in each month, method applied to CET.
It’s obvious that there is far less variation in the minimums than the warmest maximums, and in contrast to Spencer’s Canada data, maximums have been warming noticeably faster than minimums.
Linear regression since 1960
TMax: +0.47 ± 0.20°C / decade
TMin: +0.27 ± 0.07°C / decade
So about 4.7°C in a century? Maybe you might wonder if plain old linear regression is the best choice for forecasting.
“Maybe you might wonder if plain old linear regression is the best choice for forecasting.”
It isn’t. I’m not forecasting. I specifically pointed out that the trend since 1900 was not linear. I was trying to draw a comparison with Dr Spencer’s linear trend since 1900, which I’m sure he wasn’t using to make a forecast either.
Any other questions?
“CET….. That’s funny!”
Yes, very funny how CET is regarded as a good proxy for global temperatures by “skeptics” when it suits them, but is regarded as completely corrupt when it doesn’t.
“A load of sites that have been proven to be TOTALLY UNFIT and CORRUPTED for the purpose measuring basically anything.”
Is 3 considered to be a load in your part of the world? Any actual evidence for them being corrupted and unfit apart from your wishful thinking? Writing everything in all caps doesn’t count as evidence.
Yes, MetOffice sites have been surveyed.
Around 70% of them are Class 4 and 5 sites…. ie JUNK data.
Not only that , but a large number of them are producing “data” even though they have been closed for several to many years… ie TOTALLY FABRICATED
Only climate hysterics use data from sites that are known to be junk and totally fabricated.
And statisticians who have no use for scientific concerns, only how many numbers are available.
Here’s my approximation of the Spencer technique using the GHCN daily stations for Germany. Same method, warmest 3 days of each summer month.
I restricted this to just the stations which have a start date before 1900 and data up to 2024. 32 in total. This reduces the problems of moving distributions, though not every station will have data for every year. I made no adjustment for elevation.
Trend since 1900 is +0.24 ± 0.06°C / decade
Trend since 1960 is +0.49 ± 0.16°C / decade
Here’s the comparison between tmax and tmin, and as with England it’s clear that maximums have been warming faster than minimums.
Trends for tmins from 1900 are +0.15 ± 0.03°C / decade. Since 1960, +0.24 ± 0.08°C / decade.
“It says this is about summer daily high temperatures, which would imply the average of all summer days.”
Apologies. I failed to note he does show all summer days as the black line.
I would take it more seriously if corrections for average changes in station-average latitude were done, something Spencer mentioned in the post.
Look at it this way,
LIG stations were lucky to be ±2°F. A changing of the number of stations would need to exceed this to be significant. ASOS are ±1.8°F. CRN type stations are about ±0.54°F.
So you think random daily instrument uncertainties won’t cancel when averaging, but systematic changes in station distributions will cancel?
SO, one can infer that the global temperature anomaly, e.g. from the Hadley Center is upside down and backward. Figures.