Guest essay by Larry Hamlin
The L A Times ran an article addressing the year 2023 Northern Hemisphere summer temperatures which falsely claimed that:
“But in Britain and the United States, global records go back to the mid-1800s, and the two countries’ weather and science agencies are expected to concur that this summer has been a record breaker.”
Despite all the climate alarmist politically driven ignorance-based hype about “record” year summer 2023 temperatures the reality of the year 2023 summer temperatures in the U.S. and other global locations are, in fact, disputed by NOAA’s measured data.
NOAA’s year 2023 U.S. temperature data records covering the NOAA defined 3-month summer June through August period actually shows that U.S. 2023 summer temperatures were far below “record” summer maximum temperature levels regardless of whether one is looking at NOAA’s national, regional, state, county or city summer temperature data.
Looking first at NOAA’s National Contiguous U.S. Maximum Temperature for year 2023 (shown below) we see a maximum temperature of 85.72 F which represents the 109th highest maximum summer temperature of the 129 maximum summer temperatures identified. There are 20 years in which the Contiguous U.S. Maximum Temperature was higher than in 2023 with the highest ever being in 1936 at 87.92 F.
The year 2023 is not even close to being a “record” maximum highest summer temperature for the Contiguous U.S.
Looking next at NOAA’s West Regional Time Series for year 2023 (shown below for the West Region) we see a maximum temperature of 86.6 F which represents the 73rd highest maximum temperature of the 129 maximum summer temperatures identified. There are 56 years in which the West Regional Maximum Temperature was higher than 2023 with the highest ever being in 2021 at 91.2 F.
There are 7 other NOAA Regional Areas which are the Ohio Valley, Upper Midwest, Northeast, Northwest, South, Southeast, Southwest and Northern Rockies Plains of the U.S. with each of these showing that the year 2023 maximum summer temperature is not the “record” highest with these “record” highest maximum years being 1936, 1988, 1949, 2021, 2011, 2020 and 1936 respectively.
The year 2023 is not even close to a “record” maximum highest summer temperature for any of NOAA’s U.S. Regions.
Looking next at the NOAA’s Statewide Time Series Maximum Temperature for year 2023 (shown below California) we see a maximum temperature of 87.9 F which represents the 77th highest maximum temperature out of the 129 maximum summer temperatures identified. There are 52 years in which California’s Statewide Maximum Temperature was higher than in year 2023 with the highest ever being in 2021 at 91.9 F.
Of the 48 States in the Contiguous U.S. 47 did not see a “record” maximum highest summer temperature in 2023. Only Louisiana has a “record” maximum highest summer temperature in year 2023.
California’s year 2023 summer temperature was not even close to a “record” maximum summer temperature.
47 of the 48 Contiguous U.S. States did not have a “record” maximum summer temperature.
Looking next at NOAA’s County Time Series Maximum Temperature for year 2023 (shown below for Los Angeles County) we see a maximum temperature of 85.0 F which represents the 49th highest out of 129 maximum summer temperatures identified. There are 80 years in which the Los Angeles County Maximum Temperature was higher than in year 2023 with the highest ever being in 2006 at 89.7 F.
There are 58 California counties listed in NOAA’s County Time Series and none these counties had a “record” maximum highest summer temperature in 2023.
Looking next at NOAA’s City Time Series Maximum Temperature for year 2023 (shown below for Los Angeles) we see a maximum temperature of 72.8 F which represents the 47th highest summer temperature out of the 79 identified. There are 32 years in which the Los Angeles City Maximum Temperature was higher than in year 2023 with the highest being in 2006 at 76.8 F.
There are 9 California cities (including Death Valley) listed in NOAA’s City Time Series and none of these cities had a “record” maximum highest summer temperatures in 2023.
NOAA’s temperature data as identified and addressed in the above discussion for year 2023 clearly indicates that the U.S did not have “record” breaking summer temperatures in year 2023 – not at the national level, not at the nations regional level, not at the nations state level, not at the level of California’s 58 counties, and not at the level of 9 major California cities.
Despite these outcomes in California and the U.S. climate incompetent alarmist politicians and news media will continue to falsely hype “record breaking heat” as being present across the nation and its states, regions, counties, and cities based on the grossly and completely invalid misapplication of a global wide temperature averaging outcome that cannot define temperature outcomes at specific global regional, national, country, state, county, or city locals.
Additionally, given the 2023 summer temperature outcomes of the U.S. addressed above it seems unlikely that the Global Region of North American had “record” summer temperature outcomes as hyped by the propaganda driven alarmist media. This outcome is further addressed below.
Alaska highest Maximum Summer Temperatures Peak was 65.3 F in 2004 with the year 2023 maximum summer temperature being only 60.6 F in 2023 making it the 85th out of 99 total yearly maximums as shown below.
The highest reported temperature measured in Canada this year was 41.4 C in British Columbia’s South Coast on August 14, 2023.
Canada’s highest ever reported temperatures are shown below.
This data clearly supports that Canada did not have “record” high maximum summer temperatures in year 2023 which is consistent with the same outcome for Alaska whose territories occupy the same Northern Hemisphere global latitudes as Canada.
Given that neither the U.S., Alaska nor Canada had “record” high maximum summer temperatures in 2023 it seems extremely probable that neither did the entire North America Global Region.
But despite this reality in the inane ignorant based climate alarmist political propaganda world the alarmism hype will continue unabated.
“The year 2023 is not even close to being a “record” maximum highest summer temperature for the Contiguous U.S.”
The endless misrepresentation that goes on here. The LA Times did not say anywhere that it was a record for the US. The UK and US indices agreed that it was a record for the Northern Hemisphere summer average, and that is what the LA Times reported, correctly.
The endless misrepresentation always starts with Nick’s comments.
There are a multitude of unadjusted measurements that show the 1930s,40s was warmer in the NH.
You KNOW that HadCrud, GISS, GHCN etc are FAKE and do not remotely represent actual NH or Global temperatures…
… but fakery is all you have.
Please stick to topic. The headline here
“L A Times Article Falsely Asserts U.S. Had “Record” High Summer Temperatures in 2023”is itself false. The L.A.Times did not assert that.
They say
“Scientists have used tree rings, ice cores and other proxies to estimate that temperatures are now warmer than they have been in about 120,000 years. “
Which is of course total BS.
It has been warmer than now for most of the last 10,000 years
Why aren’t you correcting that LIE, Nick ?
Nice to see that you agree that it wasn’t the warmest in the USA.
Now if only you would you use non-urban, non-faked NH data, you would see that there statement about the NH is also a fallacy.
But you use fallacy all the time. Its your “thing”
UAH shows 2023 in 4th place on a year-to-date basis. (45 years)
UAH shows world oceans in 6th place on a year-to-date basis
UAH NH also shows 2023 in 6th place on a year-to-date basis.
Two warmer than normal months carry this whole story….. Whoopie doo !
Yes, I’m sure you can come up with an endless list of things that the L.A.Times did not say.
They did say..
“Scientists have used tree rings, ice cores and other proxies to estimate that temperatures are now warmer than they have been in about 120,000 years. “
Which is of course total BS.”
Where is you comment denouncing that piece of nonsense.
Do you agree that this “warmest summer” fakery relies on just 2 warm months?
So is totally meaningless.
Note, Nick has NOT denounced this LIE. !!
First off, only the totally clueless, or those dedicated to pushing a lie, would use tree rings as a proxy of temperature.
Secondly, the world was warmer than it is today for about 90% of the last 12000 years. The Holocene Optimum which lasted from around 12K years ago to about 5K years ago, was as much as 3 to 5C warmer than today.
After that, we fall back into the most recent glacial phase of the Quaternary Ice Age.
I for one am glad we do not have ice sheets down into the mid-west anymore.
use tree rings as a proxy of temperature
I’ve never understood that. There are a other factors that affect tree growth.
The Grand Solar Minimum has started the Beaufort Gyre is about to dump its fresh water into the North Atlantic possibly shutting down heat transport from the tropics to the north. The interglacial periods usually last about 10,000 years, it has been 11,700 years, so a new glacial period might be on the way.
“First off, only the totally clueless, or those dedicated to pushing a lie, would use tree rings as a proxy of temperature.
Secondly, the world was warmer than it is today for about 90% of the last 12000 years. The Holocene Optimum which lasted from around 12K years ago to about 5K years ago, was as much as 3 to 5C warmer than today.”
So we can’t use thermometers. Tree rings are out. So how do you know all those things about the last 12000 years?
There are other proxies that aren’t proxies for CO2 !
“Secondly, the world was warmer than it is today for about 90% of the last 12000 years.”
Bollocks, complete and utter. MIUMark at it again. How many is that this week? There was the poverty line one… oh and only one oil disaster……
Simon is wrong, as usual:
https://wattsupwiththat.com/2016/07/31/a-warm-period-by-any-other-name-the-climatic-optimum/
What a clown. You post a graph that has no years listed and … wait for it … was by the fraudster Tim Ball. Here is a real graph that shows MarkW is a misinformed at best, a liar at worst. I’m picking it is option 2.
Its the Slimon hockey stick!
We are domed!
Hurrah!!!!…. The proud boys poster is here. Nice to have our resident racist pop by.
Racist. n. A person who has just won a political argument with a Leftist.
Poor Simon..you would be proud if you were a boy..
But you have never reached that level of maturity and can’t decide what you are…. cockroach or amoeba.
What exactly is your evidence that the proud boys are “white supremicists”, Slimon?
A lot of them were feddies, BTW.
The DNC says they are racists, and Simon always believes what he is told to believe.
Oh, and I forgot to ask: where are the uncertainty limits on your hockey stick?
I don’t see any.
One thing that is a constant with those on the left. The minute they feel they are losing, they start accusing everyone who disagrees with them of being a racist.
Yep, a graph based on Michael Mann’s fraudulent tree-ring data.
Tree rings , which are a great indicator of CO2 and water levels, but not of temperature.
Hilarious, isn’t it ! 🙂
Sorry, I couldn’t resist reposting the photo you hate so much:
https://www.cambridge.org/core/journals/quaternary-research/article/abs/latequaternary-vegetation-and-climate-near-the-arctic-tree-line-of-northwestern-north-america/3CD9A82242E6D90FF3AF8F8D935EA44B
So Graeme…. do you have any evidence that globally the last 10000 years were warmer than today? I know you have this picture you trot out, and good on you…. but I’m looking for global. I have one.
I do indeed have one: It took me 5 seconds to find it.
https://wattsupwiththat.com/2018/11/03/yet-another-study-illustrates-that-the-medieval-warming-period-was-not-regional-but-global/
There’s a search function on WUWT. You should learn to use it.
Why do you even bother? Even a a troll you’re hopeless.
G: you are a heretic preaching to a religious believer. He’s not a troll, he’s a cultist.
Oh dearie me,, Some moron has spliced FAKE temperature measurements on their agenda driven FAKE Holocene chart.
So funny !!
Simon displaying his ignorance as usual
Yes, most of the last 10,000 years WERE warmer than current temperatures.
“Yes, most of the last 10,000 years WERE warmer than current temperatures.”
BS from the BS master.
More evidence of your dishonesty:
https://wattsupwiththat.com/2016/07/29/a-review-of-temperature-reconstructions/
Oh, and by the way, you whined that the first graph I posted had no years. Yet another lie. The reason is simply that the x-axis had been truncated to fit the frame.
Still no answer to how trees grow under glaciers??
I’m sure you can come up with a simpleton answer.
Well come on bnasty…. where is your evidence the BS you spout is true? And while I’m talking about your BS, isn’t it an olympic level irony that your get out of jail card is to say scientific data is fraudulent, yet the only data you cite is nonsense?
Large amounts of evidence posted just below.
You have just proven yourself to be a totally mindless zealot incapable of learning even the most simple FACTS.
You are an insignificant and petulant child, with zero to add to any rational discussion.
“You have just proven yourself to be a totally mindless zealot incapable of learning even the most simple FACTS.”
Translation…. I have nothing I can produce that can back up what I say…. so I’ll just resort to personal abuse.
The fact is that the dozens of posts that have presented data refuting Simon, are still unread. Simon refuses to look at anything that is not from a party approved site.
The fact is none of them were global which is the most significant point…… Now do you have one?
See my post above. the MWP was certainly global.
Don’t try repeating the lie that the warming was only regional.
Everyone know that Mickey Mann’s fabrication was fraudulent..
That you don’t seem to be aware of that shows just how incredibly IGNORANT you choose to be.
You don’t understand, Mickey Mann says that his work has never been refuted, and that’s good enough for Simon.
As usual, Simon defines nonsense as any data that disproves the nonsense he is pushing.
It is almost as though Nick KNOWS that the “warmest summer” comment is based on FAKE and MANIPULATED cooling of past data. !
Isn’t that right, Nick !
Noise
So you agree.. Maybe you aren’t as much of a mindless zealot as I thought. !
I agree with your characterization of your posts.
“Scientists have used tree rings, ice cores and other proxies to estimate that temperatures are now warmer than they have been in about 120,000 years. “
It was warmer than now just 8,000 years ago it is called the Climate Optimism, ever read about it?
It’s good to know that Nick agrees that the Danish have re-settled Greenland and started farming grains to feed their domestic animals again as they had done for hundreds of years before abandoning it due to cold.
Yes. Climate optimism has returned. Here’s Igaliku
Nice photo to show they remembered to install plenty of white Stevenson screens.
Geoff S
I’m going out on a limb and guess that Nick actually believes that this picture is relevant.
________________________________________________________
What they said is:
But in Britain and the United States, global records go back to the mid-1800s,
and the two countries’ weather and science agencies are expected to concur
that this summer has been a record breaker.
Did you not notice the word “global” in that paragraph?
Are you suggesting that the UK and USA have “global records” i,e. temperature measurements worldwide that no one else has had access to? Go on do tell us what
“But in Britain and the United States, global records go back to the mid-1800s”
means.
Are these records global or not? If they are why mention UK and USA?
No, just that in the context of the article they are clearly talking about the specific global data sets, such as HadCRUT, GISS and NOAA.
The USA has NOAAGlobalTemp.
The UK has HadCRUT.
Everyone has access to these datasets.
Yes. They are global. They mention the UK and USA because those are the countries providing those datasets.
And they are both manically adjusted , as well as being based on data from surface sites that are TOTALLY UNFIT for measuring changes in “climate” over time.
Not useful for use as anything but climate propaganda.
Again, this is just an opinion from a ‘shouty’ person on the internet. Quite funny.
Again, the Flaming Nitwit thinks urban warming and airports are “the way” to measure climate.
No accounting for its incredible ignorance and lack of any rational thought process. !
Scientific rule #1: Always trust government agencies to provide the truth for the public to see.
Scientific rule #2: Once established by the government, scientific data never changes.
No, I didn’t so you get to score a run.
Nick is implying that the United States climate is significantly different from the rest of the world, an outlier of sorts. Long time ago here on WattsUpWithThat, Richard Verney’s post – you can read it here – said pretty much the following:
The United States isn’t the world, but it spans North America with a good variety of geography and topography which makes it a good sub-set of a land mass. World-wide, the US has the best sampling of weather data of any significant land surface. If the US does not show warming, a strong explanation is needed as to why it is an outlier compared to the rest of the world as a whole.
Taken from (Richard Verney, WUWT, July 2017)
“Nick is implying that the United States climate is significantly different from the rest of the world”
I’m not responsible for that. I just point to the respective measurements. That is just how it is.
Respective measurement (real ones) show that a very large proportion of the NH had very similar temperature patterns to the USA.
Seems you are not taking responsibility for your own ignorance.
The US is only 2% of the global surface area. It would hardly be surprising if its temperature timeseries did not match that of the globe.
Except most untainted, unadjusted data from the NH DOES match US temperatures
Only it doesn’t. It matches GISS, NOAA and HadCruT.
Remember Berkeley Earth?
That was funny.
The Berkeley Earth database shows thermometer readings in the 1700’s with uncertainty values in the hundredths digit!
And we are supposed to believe anything taken from that data?
“The Berkeley Earth database shows thermometer readings in the 1700’s with uncertainty values in the hundredths digit!”
They do not. Monthly uncertainties for the 1750s are generally multiple degrees – for annual averages closer to 1°C.
“And we are supposed to believe anything taken from that data?”
Tell that to those who keep spamming the “US was hotter in the 30s” graphs using Berkeley data. They seem to believe the data when it suits them.
Malarky. Two years ago I directly asked BE about the uncertainties they showed for two stations in the 1700’s. The answer? “Well, that will be taken care of when the data is actually analyzed.”
Really? The way it is handled? IT’S IGNORED! Just like you do! “All uncertainty is random, Gaussian, and cancels”. Therefore it can be ignored!
And it *was* hotter in the 30’s than today, at least according to the recorded temperatures.
Like usual, you want your cake and to eat it too. “BE is accurate when bellman wants to use it but it isn’t accurate when someone else uses it”!
“Malarky.”
There’s that tell again. Every time Tim is caught out in a fabrication, he just shouts “malarky” or some such, and then tries to change the subject.
“Like usual, you want your cake and to eat it too. “BE is accurate when bellman wants to use it but it isn’t accurate when someone else uses it”!”
A bit of projection from the man who keeps claiming BEST is fraudulent, but is happy to use it to claim that “it *was* hotter in the 30’s than today”.
I think BEST is as accurate both when I use it and when others use it. The problem with all the “it was hotter in the 30’s” graphs are not the accuracy of the data – it’s the way it has been selected to make a point that does not stand up to scrutiny. It was hotter in the 30’s if you limit yourself to just the US data, only look at maximum temperatures, and only look at summer temperatures.
No fabrication. But I’m not going back to search emails from two years ago to try and convince you. The answer *was* that the uncertainty figure in those old records were not meant to be used – of course what uncertainties the actual analyses use was never given. I can only assume that the measurement uncertainties, as usual in climate studies, were assumed to be random, Gaussian, and would cancel. Therefore they could be ignored.
It *was* hotter in the 30’s than today. The fact that you can’t accept that fact is quite telling.
The 20’s and 30’s were hot all over the globe. If I can find them I’ll post graphs from China that shows it was hotter there as well as in the US. Between the two the land area involved is a significant percentage of the global land area.
In the meantime:
from oxfordre.com: “For example, historians Meng Wentong and Hu Houxuan interpreted the archaeological evidence and oracle bone scripts and suggested that the climate of northern China in ancient times (ca. 2,000 to 3,000 years ago) was warmer than that of the 20th century (Meng, 1930, 1934; Hu, 1944).”
from “Temperature variation through 2000 years in China: An uncertainty analysis of reconstruction and regional difference”
“ Considering the past 2000 years, the winter half-year temperature series [Ge et al., 2003] indicate that the three warm peaks (690s–710s, 1080s–1100s and 1230s–1250s), have comparable high temperatures to the last decades of the 20th century.”
for SE China: “The 500–years regional coherent temperature series shows maximum temperature during the decades 1490s, 1620s, 1670s and 1930s, whereas minimum values were found in the 1650s and 1830s.”
“the warming during the last decades of the 20th century is only apparent in the TB region, where no other comparable warming peak occurred. For the regions of NE and CE, the warming peaks during 900s–1300s are higher than that of the late 20th century, though connected with relatively large uncertainties.”
Now, come back and tell us that China and the US aren’t significant land areas.
“No fabrication.”
The fabrication was your claim that BEST showed uncertainty values of hundredths of a degree. By well evaded.
“It *was* hotter in the 30’s than today. The fact that you can’t accept that fact is quite telling.”
More telling is the way you claim this as a fact with no uncertainty, whilst claiming that global temperatures have an uncertainty in the tens of degrees.
“The 20’s and 30’s were hot all over the globe.”
and there go those goalposts. First it’s in the 30s, then it’s in the 30s and 40s, and now it’s in the 20s as well.
I’m sure you can find that everywhere in the world had at least one summer between 1920 and 1950 that was hotter than some summers in recent years. It does not mean that the globe was warmer during those 30 years than it is now.
“Between the two the land area involved is a significant percentage of the global land area.”
Which still means nothing if they happened to be the hottest parts of the globe in that particular decade. And even less if China was warm in the 20s, but the US in the 30s.
“For example, historians Meng Wentong and Hu Houxuan interpreted the archaeological evidence and oracle bone scripts and suggested that the climate of northern China in ancient times (ca. 2,000 to 3,000 years ago) was warmer than that of the 20th century”
So your evidence that it was hotter in China in the 20s – 30s, are three papers written in the 30s and 40s saying it was colder than it was 2000 years ago?
“ Considering the past 2000 years, the winter half-year temperature series [Ge et al., 2003] indicate that the three warm peaks (690s–710s, 1080s–1100s and 1230s–1250s), have comparable high temperatures to the last decades of the 20th century.”
So now you are interested in Winter temperatures.
Again, though it was warmer in the 13th century than at the end of the 20th century is not saying it was warmer in the 1930s.
“for SE China”
You missed Central East China
“The 500-year regional coherent temperature series shows temperature amplitude between the coldest and warmest decade of 1.8°C. Three extended warm periods were prevalent in 1470s–1610s, 1700s–1780s, and after 1900s. It is evident that the late 20th century warming stands out during the past 500 years.”
and Tibet
“After 1850, the temperature increased rapidly, with the last 20 years of the 20th century experiencing the warmest conditions of the past 1000 years in the TB region.”
“Now, come back and tell us that China and the US aren’t significant land areas.”
You’re citing a paper that looked at different regions of China, only one of which shows anything like temperatures being as warm in the 1930s as they were at the end of the 20th century.
It concludes:
“The fabrication was your claim that BEST showed uncertainty values of hundredths of a degree. “
Again, no FABRICATION! You are welcome to not believe it, there’s a lot of fact that you don’t believe. But It happened. When I questioned it the answer was “it doesn’t matter, we fix it later”. In other words they *IGNORE* measurement uncertainty – just like you do!
“Which still means nothing if they happened to be the hottest parts of the globe in that particular decade. “
It’s called GLOBAL warming for a reason! If CO2 is well distributed then so are its impacts. Now, come back and tell us that CO2 is *NOT* well distributed so “GLOBAL” warming only occurs in part of the globe.
“You’re citing a paper that looked at different regions of China, only one of which shows anything like temperatures being as warm in the 1930s as they were at the end of the 20th century.”
Not every place in the US saw record heat in the 30’s either. So what’s your point? The 20’s and 30’s WERE hotter in the US than today.
“In particular during the warming period of 900s–1300s, the maximum temperatures in some regions exceeded the highest level of the 20th century.”
Guess that doesn’t mean anything to you. Your lac of reading comprehension is showing again!
“Again, no FABRICATION!”
Don’t call it a fabrication, just say it was a lie then. But you said BEST claimed the uncertainty of global anomalies from the 18th century was stated to be hundredths of a degree, when they actually claim they are several degrees.
“It’s called GLOBAL warming for a reason!”
That reason being that on average the globe is getting warmer.
“Not every place in the US saw record heat in the 30’s either”
Yet you still keep insisting that the entire world was warmer in the 30s.
“Guess that doesn’t mean anything to you.”
It means what it says, and doesn’t mean what it doesn’t say. What it doesn’t say is it was warmer in the 1930s
“Don’t call it a fabrication, just say it was a lie then. But you said BEST claimed the uncertainty of global anomalies from the 18th century was stated to be hundredths of a degree, when they actually claim they are several degrees.”
You lack of reading skills is showing again. And it also shows you know nothing about the BEST.
I didn’t say BEST showed anomalies with measurement uncertainty in the hundredths digit – I *SAID* they showed temperatures taken in the 1700’s with uncertainties in the hundredths digit.
Since the uncertainty of the measurements propagate into anomalies calculated from the measurements the anomalies inherit the uncertainties of the measurements.
And BE said they use other values for the uncertainties of the actual measurements, that what the data base shows is not used!
And the truth is that they *aren’t* used – but they should be! And the measurements should be shown with the appropriate level of uncertainty! The truth is that they use the same meme you always do: “all measurement uncertainty is random, Gaussian, and cancels.” Thus you don’t have to worry about the measurement uncertainty and how to propagate it. You just ignore it.
Before you say people are liars you need to make sure of your facts. And that includes having someone else read what you are responding to you and then having them explain it to you! You simply aren’t capable of comprehending what you read.
“That reason being that on average the globe is getting warmer.”
EXACTLY what does that statement mean as far as climate is concerned? Is it from warmer minimums? Is it from warmer maximums? Is it from a combination of the two? If it’s from minimums (which is backed up by ag studies showing longer growing seasons) then what’s the impact on the global climate? It certainly isn’t causing a lack of food as the global warming advocates claim will happen. If it’s from maximums then where is the impacts? The New York and Miami aren’t underwater yet. The Arctic isn’t ice free yet. It certainly hasn’t caused food shortages or expansion of desertification. Nor has it resulted in more extreme weather.
If a proper propagation of measurement uncertainty is included with the global average temperature then it is UNKNOWN if the GAT is warming. It’s a subjective guess drawn from a cloudy crystal ball. And the future being prognosticated by the global warming advocates seems to be having about the same accuracy as a circus fortune teller looking into a cloudy crystal ball!
For a reply bellman will pull a few quotes out-of-context and make up some noise about them a la Nitpick Nick Stokes.
My mistake. You did say “thermometer readings” rather than global anomalies. I apologise for suggesting you were claiming otherwise.
But that does illustrate my point. Even if they are claiming the uncertainty of individual readings has an uncertainty in the hundreds of degrees, when used to create a monthly global average, the uncertainty is in multiple degrees. It’s the lack of coverage that is the main cause of uncertainty.
If global warming was being caused by a well-mixed, globally homogenous gas like CO2 then coverage really wouldn’t be a problem. The warming should show up *EVERYWHERE* all the time.
The growth of CO2 in the atmosphere has been consistent for a hundred years and yet we’ve been through at least two global warming cycles and one global cooling cycle. And the current warming cycle has yet to reach the magnitude that was seen 100 years ago.
“Even if they are claiming the uncertainty of individual readings has an uncertainty in the hundreds of degrees, when used to create a monthly global average, the uncertainty is in multiple degrees.”
This is so garbled I can’t make any sense of it. I suspect you are saying they are claiming uncertainty in the HUNDRETHS (i.e. .0x decimal point). And if you are saying that the measurement uncertainty grows as you combine more and more single measurements of different things then you are correct. The SEM is *NOT* the measurement uncertainty of the average. The propagated uncertainty of the elements is the uncertainty of the average.
No, other countries in the NH experienced heat waves as well.
Lots of placers have heat waves – it doesn’t mean that it’s hotter globally.
Where in the mid-Northern Hemisphere and above does it actually get “hot” in the Winter?
“hotter”, not “hot”.
Equivocate you much?
The Berks use data that is even more tainted than the GISS data.
They can and do manufacture any result they want to get.
Just because an area is ‘anomalously’ warm, it doesn’t mean that it was necessarily hot. Take a look at Antarctica:
https://scitechdaily.com/nasa-summer-2023-was-earths-hottest-since-global-records-began-in-1880/
When all a painter has in his palette is a tube of red paint, the whole picture is painted in red tones.
That’s what media graphics does with temps representations these days.
Did someone say it did?
““But in Britain and the United States, global records go back to the mid-1800s, and the two countries’ weather and science agencies are expected to concur that this summer has been a record breaker.””
“While the world’s air and oceans are setting records for heat”
Heat is usually measured by *temperature*.
It appears that the LA Times did. And you are on here trying to defend their article!
I was just pointing out that “anomalously warm” needn’t mean “hot”. Thought most people would understand that. Not here, maybe.
“Earth is sweltering through its hottest Northern Hemisphere summer ever measured, with a record August capping a season of brutal and deadly temperatures,
Brutal and deadly temperatures indicate HIGH temperatures, not anomalously warm temperatures.
Stop blowing smoke!
“The world’s oceans — more than 70% of the Earth’s surface — were the hottest ever recorded, nearly 70 degrees,”
I suppose this isn’t implying *hot*? Even though that is the actual adjective they used?
“So far, 2023 is the second-hottest year on record”
Here’s that ajective “hottest” once again! Not “anomalously warm”, just “hottest”!
I’m guessing you haven’t even bothered to read the article. If you had you wouldn’t be trying to blow smoke up everyone’s behind!
From the linked article that you apparently didn’t read: “The summer of 2023 was Earth’s hottest since global records began in 1880, according to scientists at NASA’s Goddard Institute of Space Studies (GISS) in New York.”
Here is what the LA Times was actually talking about
Yes, I think these area averaged, convenience sampled surface temperatures are statistical junk but the LA Times is correctly reporting the junk.
Six years ago I summarized the junk as a needed research program:
https://www.cfact.org/2017/05/27/a-needed-noaa-temperature-research-program/
There’s NIck again, using data that he MUST KNOW BY NOW IS TOTALLY FAKE. !!
Fake indeed: https://www.cfact.org/2017/05/18/fake-temperatures/
For Nick, the criteria is not accuracy, but rather does it support the narrative.
Reality looks more like this.
With a cooling trend since 2018
Conveniently missing 2023 from your graph.
Note that your very short cooling trend since 2018 has become an equally meaningless warming trend, of around +0.2°C / decade.
Since we are using this El Nino.. Better take it from the last El Nino.
Even a blind monkey can see the cooling trend since the 2016 El Nino spike.
And that the current spike is far lower than 2016
How am I “using” an El Niño? I was using your start date – chosen to start just before an El Niño. The current one is only just starting, and is unlikely to have had much of an impact yet.
So now you have to move back, and really use a massive El Niño to claim another pathetic cherry-picked cooling trend. And miss of the most recent month.
OK let’s drop the El Nino spike off each end
Think about you are saying , moron !!
“And that the current spike is far lower than 2016”
It’s 0.1°C lower. But remember these are anomalies, and the 2016 peak happened in February, whereas in 2023 it’s happening in August. Comparing like with like, this is the warmest August in the NH by 0.45°C.
Here’s the entire UAH NH data. Warming trend of 0.16°C / decade.
Always using those EL Nino steps…,
The only warming there has been.
So sad you are incapable.
Note, basically no warming from 1980-1997
Basically no warming from 2001-2015
Just those small non-CO2 El Nino bumps
UAH is not anyone’s reality. But, FWIW, it was a record in UAH too:
Relying on two months.. meaningless, and certainly nothing to do with CO2.
The claim that you are disputing is that the NH summer was the warmest on record – your own UAH dataset confirms that it was. That is why Nick is using JJA (summer) temperature.
Are you having difficulty keeping your own position straight?
Two months is NOT CLIMATE, muppet.
Now explain how humans caused this two-month spike.
And of course both those two months had anomalies LOWER than the 2016 peak anomaly, and only one of them was above the 1998 peak anomaly.
JJA means June, July and August. That’s 3 (three) months.
And it’s the same three (3) months referred to in the article – northern hemisphere summer.
And yes, 2023 was the warmest NH summer in the UAH record, as Nick pointed out.
Even Spencer and Christie’s report makes that clear.
Current spike is MUCH lower than 2016.
Two months IS NOT CLIMATE.
No-one in the NH experience any significant extra warmth, nearly all the anomaly was north of Siberia.
Stop your childish PANIC !!
But NH summer 2023 was the warmest in the UAH record, as it was in all the other ones. Are you simply incapable of acknowledging this. Is it psychologically too hard for you to accept?
Your heroes Roy and John have let you down, is that it|?
NO, the highest anomaly was in 2016.
The warming in August was in places basically no-one lives.
No-one in the NH experienced any significant warming except for a 2-3 day weather event.
“UAH is not anyone’s reality”
It is certainly FAR more real than the junk maladjusted urban surface data that you “believe” in !
I guess that means the weather balloon data is not anyone’s reality, either.
The UAH and weather balloon data correlate with each other, confirming each others’s readings. If Nick wants to throw away UAH, then he must want to throw away the weather balloon data, too.
And the only pristine-ish surface data, USCRN. (Since it is a close trend match to UAH USA48)
And ClimDiv since 2005. (Since it is manipulated to match USCRN)
But USCRN has been warming faster than ClimDiv since 2005!
Yes, the so-called ‘pristine’ sites have a faster warming trend than those affected by urban heat island effect since 2005, their joint period of record.
Are you saying that the scientists are deliberately lowering the warming trend in the US sites affected by UHI?
That adds a whole other dimension to the silliness on display daily here at WUWT. This is beyond parody!
Yes we know ClimDiv is being adjusted to match USCRN.
You still haven’t the remotest comprehension of “significant difference” have you.
So sad that you are so incredibly INEPT !!
So are the weather balloons saying summer 2023 wasn’t the warmest NH summer in the instrument record? Do you have a data link, for instance? In contradiction to all the surface and lower stratosphere data?
Otherwise, what does this post mean?
The weather balloons say the same thing the UAH satellite says: Which was my point.Nick wants to trash UAH, so he must want to trash Weather Balloon data, too. Get it now?
Do they?
Weather balloons are a very close match to UAH.
You do know that 2 months is NOT CLIMATE, but a weather event.
don’t you.
Come on muppet.
Explain, using real science, how July and August temperature spike was caused by humans.
This will be HILARIOUS watching you squirm around in your own putrid, oozing imagination.
“Reality looks more like this.”
Oh dear, now you will have all the usual suspects yelling at you for not including error bars, or suggesting that it’s possible to tell what UAH is doing when the monthly uncertainties are several degrees.
But if you do want to use UAH, here is what their Northern Hemisphere data look the summer months.
Pretty decisive.
I suspect WUWT will find an excuse to drop UAH from their sidebar in the coming months – given what’s likely coming.
It is climate alarmists who are prone to dismiss data that doesn’t agree with their preconceptions.
Like the fact that it was just as warm in the recent past as it is today.
Like the fact that the United States has been in a temperature downtrend since the 1930’s.
Like the fact that the global temperature record has been so bastardized that it is completely divorced from reality.
And so many more.
I noticed that Monckton has already paused his monthly pause articles. It’s not going to surprise me at all if UAH gets yanked from this site.
Why would it happen?
The climate crew assumes that everyone else is as venal as their heros.
That sounds like wishful thinking to me.
“The UK and US indices agreed that it was a record for the Northern Hemisphere summer average, and that is what the LA Times reported, correctly.”
Anomalies don’t make “hottest eveh!”. Anomalies don’t determine the summer average temperature. Anomalies are excursions from a baseline. What baseline was used in your graph? Did it include 1920-1940?
If they meant anomaly then they should have said ANOMALY, not average temperature. Those are two totally different things and you are trying to equate the two!
It’s lying by omission or by ignorance. Take your pick.
NOAA baseline is 1901-2000
NOAA is massively adjusted from the measured temperatures to get rid of the 1930s/40s peak, and also contains mostly urban and airport data.
It is totally unrepresentative of global anything !
And it is *not* representative of the northern hemisphere either.
NOAA, NASA/GISS, Japanese Met Agency (JMA), UAH_TLT and RSS_TLT all reported NH summer 2023 as the warmest on their respective records. HadCRUT still to report August but June and July both set new monthly records in that set, so very likely them too. Some conspiracy.
Rusty Nail gets it wrong again. No warming in June in Tokyo:
https://wattsupwiththat.com/2022/07/03/june-in-tokyo-hachijojima-island-hasnt-warmed-in-decades/
Didn’t say Tokyo, did I? I said JMA “Northern hemisphere”.
It was the warmest summer IN THE NORTHERN HEMISPHERE (sorry, but you literally have to shout it out at some folks) in the JMA record.
Do you understand now the difference between JMA’s Tokyo record and its Northern Hemisphere record?
You don’t, do you?
Funny that temperatures from single locations show no warming whatever. It’s almost as though “average” temperatures are an artificial unphysical construct.
Gees this “warmest summer” (since the 1940s) has you in a complete chicken-little tizzy, doesn’t it
Despite you knowing that nearly ALL the last 10,000 years were warmer than now, and civilisation developed during that period, and struggled during the couple of colder period.
But understanding basic climate and human history is going to be hard for someone with a mental age of 5, you poor thing.
NOAA GISS et al use the same corrupted data with all the 1930/40s adjusted way down to get rid of the blip.
They are also using surface sites that are heavily tainted by massive amounts of urban warming .
Also a large increase in the % of unreliable airport sites.. (or do you still think jet engine exhaust has a cooling effect ???)
Oh and just in case you are still a confused little muppet..
… Two warm months IS NOT CLIMATE
Awaiting your scientific proof that they were caused by human anything….
Yeah, it’s all a big conspiracy that’s lasted decades without anyone from any of the many groups reporting coming forward to expose it, despite the enormous wealth that would gain them from the fossil fuel companies, etc. Very believable, cough!
If it’s not a conspiracy, why does Gavin Schmidt go to such lengths to prevent publication of a paper by Willie Soon which disproves the GW narrative?
https://wattsupwiththat.com/2023/09/15/are-you-or-have-you-ever-published-the-work-of-a-climate-skeptic/
Yeah?!
So, we all notice that you are still helping fund those fossil fuel companies.
Electricity, computers, communications, travel, food, transport etc etc etc.
Helping them to support the rest of society by continuing the provision of the things EVERYONE relies on.
Very charitable of you to be so incredibly hypocritical.
How quickly you forget the exchanges during ‘Climategate.’
I think he is in denial.
Yeah, and before that it was a Human-caused Global Cooling conspiracy.
That went away when the temperatures starting climbing and the Human-caused Global Cooling advocates could no longer maintain this fiction.
My guess is that at some time the Human-caused Global Warming advocates will reach a point where this fiction can no longer be maintained, either, after the temperatures start to cool.
You are going on and on about a “blip” up in temperatures, but you never mention the underwater eruption of Hunga Tonga. No connection, you think?
Anomaly is *NOT* temperature. How many times does that need to be pointed out before it is understood?
Anomalies *have* to be weighted against what the absolute temperature is. None of the organization you list do that. They weight a +1C change from -30C to -29C the same as a +1C change from 20C to 21C.
The impacts on the actual climate of either change is NIL – i.e. ZERO! If the temperatures are “averages” rather than actual absolute measurements then the average even hides what the variance of the temperature inside the average is – so how can you tell if it is *hotter*? The average can go up because minimum temps go up while it doesn’t get hotter at all!
If things don’t thaw out at -30C then they won’t thaw out at -29C. Evaporation at 20C is the same as at 21C, at least as far as it can be measured using today’s infrastructure.
So how do you relate an ANOMALY to climate?
You didn’t address the main point. Anomalies are *NOT* temperature. If the LAT meant anomalies then why didn’t they say anomalies?
They are temperature anomalies. Differences from long term average temperatures. They’re not just random numbers.
For example, from the UAH August 2023 report, Spencer and Christie explain the relationship between anomalies and the temperatures they are derived from:
40 milli-Kelvin, BFD. You ascribe meaning to this difference?
Rusty is clueless about metrology, amongst other things.
Bingo.
FN will not understand what that means.
Doesn’t realise just how insignificant it is.
Doesn’t realise how insignificant he is.
Don’t know what you mean, sorry.
“Don’t know what you mean”
Yes, we are well aware of that fact. !
Anomalies are *NOT* temperatures. They are temperature differences! How do you relate an ANOMALY to climate? The anomaly calculated from averages tells you ZERO about what actually changed! Minimum temps could be the main contributing factor to an average going up – yet it seems like you and your buddies assume that the anomaly going up means it got HOTTER! I.e. the Earth is going to burn up from higher temps.
Except the ANOMALY can’t tell you that! It’s a SUBJECTIVE OPINION from those with an agenda to push!
What do you not get about this ‘anomaly’ thing, Tom?
Why do you think it is that all the climate science global temperature databases use anomalies? Including the (here) much-lauded UAH?
It’s very simple to use and understand. It’s just a difference from a long-term average temperature.
What is the problem?
“It’s just a difference from a long-term average temperature.”
That long-term average temperature tells you NOTHING about climate. It’s a mess of a distribution – summer temps combined with winter temps, both with different variances, no weighting given to actual local changes in climate (i.e. a change in the average from -30C to -29C is hardly a *change* in climate), no propagation of uncertainty meaning no one knows if the differences in averages are meaningful or not.
The SEM of the global average is *NOT* the uncertainty of that average. The uncertainty, according to the GUM, is the dispersion of values around the mean, not how accurately you calculate the average (which is what the SEM tells you).
What is the problem? The entire mess is based on tradition – using Tmax and Tmin to drive a value that is truly meaningless on a global basis. When is climate science going to join the rest of science in using degree-days and/or enthalpy? Ag science has changed, HVAC engineering has changed – yet climate science remains stuck using the same methodology and protocols used in the 1700’s! If climate science was medicine we’d still be “bleeding” sick people!
The anomaly in 2016 was larger than now….
… what don’t you “get” about this anomaly thing ?
And no, it is a very short-term average, based on a rather cool period after the much warmer 1930s and 40s.
The period around the time of the “global cooling” propaganda scare.
If it is so simple, why don’t you understand it !
Why don’t you understand that the “base period” is the period around the global cooling scare.. a COLD period
Why don’t you understand that 1.5C warming out of the COLDEST PERIOD IN 10,000 YEARS….
… is nothing to go into a total lather of MANIC PANIC about …
… and is in fact a very beneficial thing to happen ?
“How do you relate an ANOMALY to climate?”
OK, so when UAH says their July temperature is 266.06K, what does that tell you about climate?
“OK, so when UAH says their July temperature is 266.06K, what does that tell you about climate?”
It tells me NOTHING! It’s an AVERAGE! What’s the variance of the data that created that average? What’s the shape of the distribution of the temperatures that is associated with that average?
When the uncertainty interval for UAH is at least +/-1C then what do differences in the tenths or hundredths digits tell you? The values that go into those decimal places is UNKNOWN!
Why don’t you ANSWER the questions posed to you? Answering a question with a question is a logical fallacy. It’s not an answer to the question. In essence it’s a form of the argumentative fallacy known as Red Herring.
*ALL* of the analyses of the temperature data bases are built upon a statistical edifice that has no foundation at all. Each temperature is a sample with a sample size of 1. Meaning the SEM is undefined since one data element, the recorded temperature, has no standard deviation. When you combine samples of size 1 together there is no guarantee of what the distribution will be, it could be multi-modal, gaussian, skewed left, skewed right, or of any form. All those samples are nothing but “stated values +/- uncertainty”. Since you can *NOT* just assume all those uncertainty intervals for a random, Gaussian distribution you can’t assume that they all cancel. Therefore the uncertainty of the average value becomes the uncertainty intervals added in quadrature. This gets so large on a global basis that even the value of the units digit is UNKNOWN.
Climate science just ignores all this and makes the unwarranted assumptions that the variances of the temperatures in a month across different locations can be ignored and that everything forms a random, Gaussian distribution.
I have yet to find *any* study or analysis of the “global average temperature” that properly handles the uncertainty of the individual temperatures. They all assume the SEM of the calculated average is the uncertainty of the average. According to the GUM the SEM is *NOT* the uncertainty of the average. The uncertainty of the average is the dispersion of values around the average and can be one of three things: 1. the standard deviation of the distribution values, 2. The propagated uncertainties of the distribution values, or 3. a combination of 1 and 2.
The issue with 1. is that it assumes the same thing is being measured multiple times. That is *NOT* an appropriate assumption for a global temperature database. It’s not even a valid assumption for the mid-range daily value of temperature for a single station on a single day!
bellcurveman likes to claim that “sampling” is the “largest source of uncertainty” in these global air temperature measurements. He never explains exactly what he means and it is never quantitative, always just qualitative hand-waving.
My best guess its a take-off on Stokes’ absurd claim that with enough stations, bias error magically transmogrify into random errors to justify climastrologers ignoring them.
Sampling error is a red herring. Each temperature measurement is a sample with a size of 1! The SEM then equals the standard deviation of the population divided by 1. Except with ONE data point in the population there is *NO* population standard deviation! Which means there is no SEM – voila! = no sampling error.
Jamming all those samples of size 1 together doesn’t make a distribution that is even capable of being analyzed statistically, especially when systematic uncertainty exists in each and every individual sample. Systematic bias in each and every single data point simply doesn’t turn into random errors that cancel. They might PARTIALLY cancel but that just means you add them in quadrature. You can’t just assume they totally cancel, even though that is what climate science does!
With every averaging step the crystal ball gets cloudier and cloudier. Climate science assumes just the opposite, the crystal ball gets clearer and clearer. They are so similar to the fortune tellers at the circus that it is amazing! “I see warming in your future!”
“Except with ONE data point in the population there is *NO* population standard deviation! Which means there is no SEM – voila! = no sampling error.”
Still clueless.
The max temp here today was 83F. What’s the standard deviation for that? What is the SEM? What is the average?
If karlomonte had ever tried to read any of my posts, instead of just responding with cheap insults and pathetic nicknames, he might have figured out by now what I mean by sampling uncertainty. He might also realize I am not talking about global temperatures, but about sampling in general.
It isn’t that difficult. Take the mean of a random sample from a population, and the uncertainty is how close that sample mean is likely to be to the sample mean. At the simplest this will be indicted by taking the standard deviation and dividing by the square root of the sample size. That’s what I mean by sampling uncertainty, as opposed to the uncertainty caused by measurement errors.
In general you would expect sampling uncertainty to dominate in the uncertainty of a mean, as both decrease with the root of N, but the deviation from sampling should be bigger than the standard measurement uncertainties.
If, for example, you had Tim’s 100 thermometers each with a random and independent uncertainty of 0.5°C, the uncertainty of the mean would be 0.05°C. But if the temperatures had a standard deviation of 10°C, than the SEM would be 1°C, and the combined uncertainty would be √(1² + 0.05²) ≈ 1.0°C
In general it makes sense that the measurement uncertainty is less than the deviation in the sample, becasue why would you be using a measuring device that was less accurate than the difference in the things you are measuring?
Of course this assumes that there is no major systematic error in the measurements, and that you understand how to propagate random measurement uncertainties properly.
As to how to estimate the uncertainty in an actual monthly global anomaly record, I couldn’t say. It’s different becasue they are not random samples and require many different processes in determining the average.
“He might also realize I am not talking about global temperatures, but about sampling in general.”
bellman Evasion Rule No. 1: redirect the discussion to a different topic.
The issue being discussed here *is* TEMPERATURE, specifically the global average temperature!
“In general it makes sense that the measurement uncertainty is less than the deviation in the sample, becasue why would you be using a measuring device that was less accurate than the difference in the things you are measuring?”
bellman Evasion Rule No. 3: Assume the reader is stupid. Temperature measurement around the globe is done by using THOUSANDS of measuring stations of unknown accuracy with unknown systematic biases. It is *not* an issue of “a measuring device”.
“Of course this assumes that there is no major systematic error in the measurements”
Which *is* the standard assumption in climate science – and your typical assumption as well! And it not justified at all when combining thousands of measurements made by thousands of field temperature measuring devices of unknown calibration, unknown microclimates, and unknown systematic biases.
“As to how to estimate the uncertainty in an actual monthly global anomaly record, I couldn’t say. “
Then you are just wasting everyone’s bandwidth – a typical troll. Nothing to offer that is on point.
“It’s different becasue they are not random samples and require many different processes in determining the average.”
Every time you do an average and supply only that average there is data loss that is absolutely required in order to judge if the average is meaningful or not.
The variance of the “global average temperature” and its associated anomaly is so wide that the average *is* basically meaningless. The uncertainty cloud is at least in the units digit and most likely in the tens digit. Trying to identify differences in the hundredths digit is no different than a fortune teller at the circus “seeing” something in a cloudy crystal ball.
“As to how to estimate the uncertainty in an actual monthly global anomaly record, I couldn’t say. “
This would require him to do some actual meaningful work — much easier to just hand-wave another load of nonsense and hope no one notices.
“This would require him to do some actual meaningful work”
What work do you think I need to do? I’m not the one claiming all the published uncertainty estimates are completely wrong.
You’re the troll who claims all other sources of uncertainty can be ignored except for your “sampling” nonsense, you do the work.
The Empirical Rule in statistics tells us that a good estimate of the standard deviation is 1/4 the range (1/6 for a more conservative estimate). For a global temperature of over 200 deg F, the SD should be about 50 deg based on annual temperatures. Not a single troll here has even addressed this issue, let alone provided a satisfactory explanation of how one can claim a precision of +/-0.01 deg F when the Empirical Rule suggests a value at least 4 orders of magnitude larger.
The above should be “200 deg F range.” A longer period of time gives a range of about 300 deg F, and dividing by 6 still gives about 50 deg.
This is why it’s useful to look at anomalies rather than temperatures. Global temperatures might vary by tens of degrees, but anomalies don’t.
Why?
The magic of the almighty anomaly cannot cancel non-random error.
It can in bellman’s statistical world.
Anomalies inherit the variances of the components used to calculate the anomaly. If the variances, i.e. the uncertainty, of the components is wide then the uncertainty is high for the anomaly – meaning it is very uncertain as well.
This has been explained to you multiple times but you insist on ignoring it. It’s the only way you can justify claiming the anomalies are accurate to the hundredths of a degree.
Read my comment and the one I was replying to. This has nothing to do with how uncertain you think an individual anomaly is. It’s pointing out that Clyde Spencer’s uncertainty caused by the huge variation in global and seasonal temperatures is greatly reduced by taking anomalies. Simply because there is far less variance in anomalies than in absolute temperatures.
Only because you throw all of them in the rubbish.
Only in bellman-land does the variance decrease when you subtract two random variables.
He will be sure to lie and deny when confronted with how he forces everything into his nice, neat little Gaussian box.
“ It’s pointing out that Clyde Spencer’s uncertainty caused by the huge variation in global and seasonal temperatures is greatly reduced by taking anomalies.”
It is *NOT* reduced by taking anomalies!
Baseline = A +/- u(a)
Annual temp avg = B +/- u(b)
A – B ==> u(a) + u(b)
A + B ==> u(a) + u(b)
Uncertainty grows whether you add or subtract. The anomaly uncertainty is the sum of the uncertainties of each component.
“Simply because there is far less variance in anomalies than in absolute temperatures.”
This is basic statistics! Variances add when you combine random variables, whether you add the variables or subtract the variables.
Variance *is* a measure of the uncertainty of a distribution because it describes the range of possible values. The wider the range of possible values the higher the uncertainty is.
You simply can’t get *anything* right, can you?
His religion would collapse if he acknowledged that variances increase, so they must be swept under the rug.
This is called “honest” science for climastrologers.
“It is *NOT* reduced by taking anomalies!”
Of course it is. Are seriously claiming there is as much variance in anomalies as there is in absolute temperatures. You only have to look at any of the UAH maps to see anomalies are varying across the globe by a few degrees rather than the 10s of degrees for global temperatures.
All you are doing is obessing over the uncertainties of individual values and the very small increase in uncertainty when subtracting a 30 year mean. But, that is not not what Clyde Spencer is talking about. He’s specifically talking about the uncertainty caused by having a wide range of temperatures.
“It is *NOT* reduced by taking anomalies!”
Only if you IGNORE all the variances from ALL the intermediate averaging steps!
You will never learn this lesson.
Oh dear, he still doesn’t understand this, and I’m not sure if it’s worth explaining it again.
PeeWee goes back into projection mode.
Variance and uncertainty are related issues. They both add whether you are adding or subtracting the associated stated values.
The variances/uncertainty of the baseline and the measurement ADD, whether you do an addition or subtraction of the baseline and the measurement.
“Are seriously claiming there is as much variance in anomalies as there is in absolute temperatures.”
YES! That is the plain truth!
What *IS* the variance? The variance is a measure of the variability in your data. The more spread the data the larger your variance is!
The spread in the absolute temps is reflected into the spread of the anomalies.
It’s truly no different than any measurement uncertainty propagation.
I already showed you how this works.
Baseline = A +/- u(A)
Monthly = B +/- u(B)
A – B = anomaly
u(anomaly) = u(A) + u(B)
So the anomaly should be given as “stated value +/- uncertainty” which is Anomaly +/- [u(A) + u(B)]
The *global* anomaly is the average of the individual anomalies. The uncertainty in the individual anomalies are propagated onto the average anomaly. The SEM of the anomaly mean is *NOT* the measurement uncertainty of the anomaly!
“ But, that is not not what Clyde Spencer is talking about. He’s specifically talking about the uncertainty caused by having a wide range of temperatures.”
You *still* haven’t figured it out! A wide range of values, i.e. the temperature values, indicates a high variance which, in turn, indicates a high level of uncertainty in the mean!
And you still don’t get it.
You are talking about the uncertainty in any specific anomaly measurement. I, and Spencer, are talking about the variation of all the values, whether absolute temperature or anomaly.
He is saying that as the standard deviation of temperatures across the globe, and across a year, is large, than the mean of a sample will have a large uncertainty.
I am saying that is true for absolute temperatures, but less true for anomalies. Anomalies flatten out the absolute range of values, and instead show the relative change in temperature compared to a base period. The standard deviation of anomalies across the globe, and across the year, is smaller than that for absolute temperatures.”
“The SEM of the anomaly mean is *NOT* the measurement uncertainty of the anomaly!”
As I explained before and you keep ignoring, the SEM will usually be a lot larger than that for the measurement uncertainties.
“You are talking about the uncertainty in any specific anomaly measurement. I, and Spencer, are talking about the variation of all the values, whether absolute temperature or anomaly.”
In other words both of you are applying the meme: “all measurement uncertainty is random, Gaussian, and cancels. So we can use the SEM as the uncertainty of the average”
“Anomalies flatten out the absolute range of values”
Anomalies are nothing but a way to scale values. The flatter a distribution is the higher its uncertainty is. A distribution that is very spiked, i.e. a small SD, has far less uncertainty than one with a very small rise at the mean, i.e. a “flat” distribution.
As I already showed you for the last three days here in Kansas the anomalies have a much wider spread than the absolute values do. If you are going to compare the two distributions you have to normalize the them to the same scale. Otherwise you could just divide everything by 10^6 and say the variance is small! Which *is* what you are trying to do!
“In other words both of you are applying the meme”
The only person who says that everything is Gaussian is the voice in your head – which I’m beginning to worry might be some sort of parasite.
“Anomalies are nothing but a way to scale values.”
Wrong – but I guess “scale” is another one of those words you don’t understand the meaning of, but like to throw around.
“The flatter a distribution is the higher its uncertainty is.”
Wrong. But make your mind up. Are anomalies just scaling values or are they changing the distribution?
“A distribution that is very spiked, i.e. a small SD, has far less uncertainty than one with a very small rise at the mean, i.e. a “flat” distribution.”
Nonsense on stilts.
“As I already showed you for the last three days here in Kansas the anomalies have a much wider spread than the absolute values do.”
And you were wrong.
But as I’ve told you, you are missing the point of anomalies if you are only looking at one point in time in one place.
“Otherwise you could just divide everything by 10^6 and say the variance is small! Which *is* what you are trying to do!”
Try to think about what you are saying and why it might be wrong.
Every single thing you said here is wrong. Take the “flat” curve. It will have a larger SD than a peaked curve. That means that the mean has a lower probability for one curve – i.e. more uncertainty.
You can’t seem to get *anything* right! It stems from your insistence on arguing that black is white and white is black.
“Take the “flat” curve. It will have a larger SD than a peaked curve. That means that the mean has a lower probability for one curve – i.e. more uncertainty.”
Just not true. Take the probability distribution of a single 6-sided die, and compare it with the sum of 100 6-sided dice. Which has the flatter curve? Which has the larger uncertainty? You keep ignore the absolute value of the standard deviation of the curves.
But you’re wrong in any case. For the uncertainty of a mean the shape of the curve can be expected to be close to a normal distribution regardless of whether it’s based on anomalies or absolute temperatures – thanks to the CLT. The larger the standard deviation of the population the wider the standard deviation of the means. As anomalies have smaller standard deviation the standard deviation of the means will be smaller, the curve is more peaked, hence less uncertainty.
“You can’t seem to get *anything* right!”
From now on I’m just going to assume whenever you use that phrase, it means you have lost the argument.
YMHW
We went through this once before. As usual, your memory is so bad you can’t retain anything.
Temperature is not a six-sided die. Temperature does not have discrete values, it is a continuous distribution.
“But you’re wrong in any case. For the uncertainty of a mean the shape of the curve can be expected to be close to a normal distribution regardless of whether it’s based on anomalies or absolute temperatures – thanks to the CLT.”
Wow! Just WOW! You can’t get anything right. The CLT tells you about the distribution of the SAMPLE MEANS, not the distribution of values around the population mean!
I’ve given you a graph of the difference before, in fact just a few messages ago! I’ll attach it again. See if you can finally figure out what it is showing!
“As anomalies have smaller standard deviation the standard deviation of the means will be smaller, the curve is more peaked, hence less uncertainty.”
I’ve shown you from my own temperature data why this is untrue. You can’t even understand normalization and scaling. If what you are claiming here is true then just divide everything by infinity and the standard deviation will approach zero in the limit.
Where did you learn your math?
“Temperature is not a six-sided die.”
Gorman evasion rule number 1 – distract about some irrelevant difference rather than actually answer the question.
OK.
Take the rectangular probability distribution with range -1 to 1, and compare it with a Gaussian probability distribution with mean 0, and a standard deviation of 200 . Which has the flatter curve? Which has the larger uncertainty?
“Wow! Just WOW! You can’t get anything right.”
If only you would spend more time considering what I’m sayinbg, and what you are saying, rather than this endless pearl clutching.
“The CLT tells you about the distribution of the SAMPLE MEANS, not the distribution of values around the population mean!”
Yes, because you were talking about the probability distribution of the mean, not of the sample. But regardless the answer is the same. There is less deviation in a distribution of anomalies than there is in a distribution of absolute values.
“I’ve given you a graph of the difference before”
Your graph is just showing the difference between a standard deviation of a population and the SEM of a sample of 10. It has nothing to distinction between absolute values and anomalies. But thanks for illustrating my point, that uncertainty decreases with increased sample size.
“I’ve shown you from my own temperature data why this is untrue”
”
And I’ve explained above several times why that is irrelevant and wrong. Show the data over a year not 3 days, and look at the actual standard deviation, not as a meaningless percentage.
As I have it loaded at the moment let me look at CET daily values for 2022. Across the year the standard deviation is 5.4°C, with a mean of 11.1°C.
Converting the daily means to anomalies based on the average from 1981 – 2010 gives a standard deviation of 2.7°C, with a mean of +1.2°C.
The standard deviation of the anomalies was half that of the absolute values – entirely down to removing the seasonal variation. The bigger the seasonal variation, the bigger the result.
Even re-scaling as you for some reason want, results in a sd for the scaled absolute values of 0.17, verses an sd of 0.13 for the scaled anomalies.
“If what you are claiming here is true then just divide everything by infinity and the standard deviation will approach zero in the limit. ”
Well yes – the smaller the values the smaller the standard deviation. But that’s not what is being done with anomalies. They are not the result of re-scaling the absolute values. If the data was stationary the standard deviation would be identical for both absolute temperatures and anomalies. You are just subtracting the mean from all values. The reason the standard deviation of anomalies is smaller is not because you have divided the values by anything, it’s becasue you are subtracting the non-stationary means – that is seasonally adjusting the values.
“bellman Evasion Rule No. 1: redirect the discussion to a different topic”
In Gormanland, clarifying what you said is “evasion”.
“The issue being discussed here *is* TEMPERATURE, specifically the global average temperature!”
The issue was about sampling. I’ve s[pent well over two years trying to explain to your group how sampling works, and why it does not mean that uncertainty increases with sampling size. I’ve also made it clear that actual global anomaly uncertainties are not simply obtained by dividing the standard deviation of all the readings by the square root of the sample size.
“bellman Evasion Rule No. 3: Assume the reader is stupid. ”
When people keep asking what the uncertainty of a sample is, I have to make some assumptions.
“Temperature measurement around the globe is done by using THOUSANDS of measuring stations of unknown accuracy with unknown systematic biases. It is *not* an issue of “a measuring device”. ”
And it’s reasonable to assume these devices are capable of detecting changes of a few degrees.
“The variance of the “global average temperature” and its associated anomaly is so wide that the average *is* basically meaningless.”
And Gorman evasion rule number 1, change the subject.
We are talking about the uncertainty of the mean, not how meaningful it it is.
But obviously it is meaningful or people here wouldn’t be so desperate to claim it’s so uncertain.
“The uncertainty cloud is at least in the units digit and most likely in the tens digit.”
And yet Tim will still be certain that global temperatures were warmer in the 30s than today, despite there being an uncertainty of tens of degrees.
“In Gormanland, clarifying what you said is “evasion”.”
Malarky. You are trying to change the subject from the uncertainty of the “global average” to hypothetical sampling error – AN EVASION.
“The issue was about sampling.”
No, the *issue* was the uncertainty of the global average temperature.
“The issue was about sampling. I’ve s[pent well over two years trying to explain to your group how sampling works,”
But you *NEVER* relate it to the uncertainty of the mean, not the SEM, the MEASUREMENT uncertainty of the mean.
“When people keep asking what the uncertainty of a sample is, I have to make some assumptions.”
NO ONE is asking what the SEM is. They are asking what the MEASURMENT uncertainty of the mean is!
“And it’s reasonable to assume these devices are capable of detecting changes of a few degrees.”
It is?
You’ve been given graphs of temperatures from NE Kansas any number of times and asked why there is such variance in the temperatures. You *NEVER* address that issue. You just continue to make claims that all measurement uncertainty is random, Gaussian, and cancels.
See the attached image. Why the spread of temperature over such a small distance? 7F between Manhattan and Hiawatha, 6F between Topeka and Hiawatha. 5F between Manhattan and Clay Center.
Assuming that the temperature measurement devices are at least minimally accurate then why the variance in the temperatures? What *is* the variance of the data? It’s almost 4F. The average is about 84 with a standard deviation of about 2. That means there is a significant uncertainty in that mean!
Expand that to the GLOBE! That standard deviation won’t go down, it will only get larger as the variance of the data set grows.
“And Gorman evasion rule number 1, change the subject.”
Nope. The ISSIE is the measurement uncertainty of the global average temperature. You have admitted you have nothing to offer on that issue. So why do you keep posting crap hoping to distract people from the issue!
“And yet Tim will still be certain that global temperatures were warmer in the 30s than today, despite there being an uncertainty of tens of degrees.”
My parents and grandparents LIVED through that time. I have direct historical context on it. Have *YOU* ever cracked a history book on that time? Have *you* read the Grapes of Wrath?
Another load of bellcurvewhinerman hand-waved nonsense.
You still don’t have Clue One about uncertainty. To toss all other sources in the trash you first have to analyze them all, then justify which ones can be neglected.
But this would require actual work, using tools that are not in the trendologists’ toolbox.
“Another load of bellcurvewhinerman hand-waved nonsense.”
And on and on it goes. karlo complains I won’t explain what I mean by sampling uncertainty. I explain. He pretends to not understand the explanation and then falls back on yet more pathetic name calling.
And in a few days time he will be telling anyone who will listen that I never explain what sampling uncertainty means.
Poor whinerman, doesn’t get the respect he thinks he deserves.
That is only true for data that have the property of stationarity. That is, the mean and standard deviation don’t vary with time.
Because for a very long time instruments could not be manufactured that had the desired accuracy and precision. As long as we have to depend on measurements made along time ago in order to establish changes, we will have to live with that problem.
“That is only true for data that have the property of stationarity.”
Not true. I’m talking about a random sample. That’s random with regard tot he population. If you are looking at the average over a fixed period of time, and take genuinely random values over that time period, you still have identical distribution – the distribution being the temperature over that entire time period.
Of course, as I also said, for an actual global average over a time period, you are not actually taking random values. You are usually taking one reading a day. It’s not a random sample but a systematic sample. That should eliminate the uncertainty caused by a random sample.
“Because for a very long time instruments could not be manufactured that had the desired accuracy and precision.”
They were accurate enough to give useful results, or there would have been no point using them.
My point is the issues with uncertainty for 19th let alone 18th century global temperatures is far less how accurate the measurements were, and much more to do with the coverage.
How do you make a “random sample” from a bucket load of air temperature measurements?
You get ONE try at measuring an air temperature and it is gone forever.
All your averaging gods cannot overcome this inconvenient fact.
This right here is hand waving: “should”. You just hope it does.
Bullshit, Pat Frank has demonstrated the historic liquid-in-glass instruments were not up to the tast.
Reality and Bellman will never meet, let along mix.
“How do you make a “random sample” from a bucket load of air temperature measurements?”
I wouldn’t. I’d use the existing data, which as I keep pointing out is not a random sample.
“You get ONE try at measuring an air temperature and it is gone forever.”
Hopefully though you write it down, so it isn’t really gone.
Typical bellcurveman smoke and noise.
You need to stop digging, the hole you are in is deep enough.
If you don’t take a random sample then the mean you calculate is meaningless. There’s no guaranteed way to assume that a non-random sample represents the population at all.
In essence, you are saying that the temp data being used today can’t possibly give an accurate global average temperature since it isn’t a random sample of global temperatures!
His answer about writing down a measurement shows just how disingenuous he truly is. A total avoidance of the real issue.
Then you need to spell out exactly what you think the “real” issue is. All you utter are this gnomic phrases such as “when one measures a temperature it’s gone forever.”
It has no relavance either to a random sample, or to the real uncertainty of a global average anomaly. If you have something relevant to say, say it.
Been said a thousand times, it doesn’t sink into your skull.
You’ve been told another thousand times—I’m not going to waste any time trying to educate you.
Translation: “karlo doesn’t have a clue what he’s talking about, but hopes that if he just throws out clever sounding phrases whilst refusing to answer when called out, people will be fooled into thinking he’s a genius.”
I’m not going to wade through more of your hand-waved trendologist word salads.
“If you don’t take a random sample then the mean you calculate is meaningless.”
You think all means are meaningless, so why do you care?
But again you are just wrong. A monthly average based on taking one daily value every day is going to be more accurate than one based on taking 30 random daily values. The uncertainty on a sample comes from the randomness. That’s because there is always a chance that the random sample doesn’t reflect the population. Taking a more systematic sample reduces that chance. It’s common sense really, so I doubt you will ever get it.
Your definition of “accurate” reveals (again) your “everything is gaussian” bias.
By your logic a single measurement of anything is 100% error-free!
If you actually had any confidence in your claims, you’d just make the case rather than just inventing all these strawmen.
WTF is a “random daily value”, clownpants?
A daily value taken at random. The clue is in the words I use.
Say you want to know the average temperature for a month at one station. You record the temperature for each day, and average them. That gives you a sample covering the entire month, but it isn’t random
If you insist on a truly random sample, you could get that by selecting 30 days at random (with replacement). You might get one day come up several times, and other days not appear at all. The average you get will have come from a random sample, and you can divide the SD by the root of 30 to estimate the SEM. But it won’t be as accurate a mean as just taking each of the 30 days for your sample.
Why in the world would you ever want to apply random sampling to a time series?
Nutty.
You wouldn’t. If you actually tried following the threads rather than butting in with childish jokes, you might have understood that’s what I’ve been saying.
No, that is *NOT* what you have been saying. If you had bee saying that then you would realize that the daily mid-range value does *NOT* capture the time varying signal throughout the day.
It must be so easy to think you are winning these arguments when you just ignore everything I actually say.
What you say nothing but hand waving, without anything quantitative.
What do you want me to say quantitatively about someone just lying about what I’m saying?
More of your non-physical bullshit.
“You record the temperature for each day, and average them. That gives you a sample covering the entire month, but it isn’t random”
That simply doesn’t work for a time varying signal.
“If you insist on a truly random sample, you could get that by selecting 30 days at random (with replacement).”
You are measuring daily temperatures in a month. How do you select 30 days at random? You only have 30 days total!
“The average you get will have come from a random sample, and you can divide the SD by the root of 30 to estimate the SEM. But it won’t be as accurate a mean as just taking each of the 30 days for your sample.”
Thank you Captain Obvious. Calculating the population average using the population values is more precise than using a sample to estimate it! (btw, it doesn’t make the mean more accurate, just more precisely located).
That really isn’t addressing what you originally said about single measurements being more accurate than multiple measurements.
He’s already backpedaled out of this quicksand.
“You think all means are meaningless, so why do you care?”
The mean of a multi-modal distribution such as one you get from combining winter and summer temperatures into a data set and then using them to calculate anomalies *IS* meaningless. The variances of winter and summer temps are different, combining them is like combining the heights of Shetlands with the heights of quarter horses.
The mean of a skewed distribution is meaningless. For a skewed distribution you need the mode and the median along with the mean to make even a little sense out of the mean. Combining Tmax from a sinusoidal time series with Tmin from an exponential decaying time series can’t do anything but give you a skewed distribution. There are published, peer reviewed studies out on the internet that document the biases caused in the temperature record from doing this.
The mean of a random, Gaussian distribution has meaning but that requires multiple measurements of the same thing using the same instrument (with no systematic bias) under the same environmental conditions. That restriction is *NOT* met by the global temperature databases.
“A monthly average based on taking one daily value every day is going to be more accurate than one based on taking 30 random daily values.”
You simply do not understand how to measure a non-stationary, cyclical process at all. A single daily value tells you what about the daily temperature profile? Answer? Literally nothing. From one value you simply cannot identify that the daytime temp is sinusoidal and the nighttime temp is an exponential decay. Without that information that one single, daily value is meaningless. So is the average you calculate from that one, single, daily measurement. At least with 30 random daily values you have the possibility of identifying the actual temperature profile.
“The uncertainty on a sample comes from the randomness. That’s because there is always a chance that the random sample doesn’t reflect the population.”
Then why are you so insistent on using the SEM as the uncertainty of the average?
“Taking a more systematic sample reduces that chance. It’s common sense really, so I doubt you will ever get it.”
That’s simply not true for a non-stationary time series. If I take a systematic sample of a time varying voltage, V, then it will look like a DC voltage whose value is dependent on where the systematic sample falls on the time varying signal. The mention of Nyquist just went right over your head or it went in one ear and out the other while encountering nothing to slow it down.
Yep! He has statistics sampling on the brain and cannot see anything but.
“A single daily value tells you what about the daily temperature profile? Answer? Literally nothing.”
Why do you keep trying to answer your own questions, and then getting the answer completely wrong?
“From one value you simply cannot identify that the daytime temp is sinusoidal and the nighttime temp is an exponential decay.”
Yet somehow you know it is, almost as if you have more information than just a single value.
“At least with 30 random daily values you have the possibility of identifying the actual temperature profile. ”
You really can never see the wood for the trees, can you? The point of taking daily temperatures is not to determine the daily temperature profile, it’s to determine how temperatures change day to day, and to estimate a monthly average.
If you can only take one temperature a day, it’s best to ensure you take it at the same time. That way you are getting a like for like comparison – but as always it won;t tell you everything, it’s just the best you can do with limited resources.
Better is to have an estimate of the average temperature for each day, either obtained from taking several readings at fixed times of the day, or using a max/min thermometer.
Throwing away information and data—the essence of trendology.
“Yet somehow you know it is, almost as if you have more information than just a single value.”
I’ve given you graphs of the temperature profile MULTIPLE times, all based on 5min data I have collected locally since 2012. As usual, your memory is so poor you can’t remember any of them.
“The point of taking daily temperatures is not to determine the daily temperature profile, it’s to determine how temperatures change day to day, and to estimate a monthly average.”
If you don’t know how the temperature changes during the day then you have NO chance of accurately determining how it changes from day to day or over a month. Multiple climates can result in the very same mid-range values – meaning that you really don’t know what is happening to the climate anywhere!
This is why I say that climate science should move to using degree-days instead of the mid-range value.
“If you can only take one temperature a day, it’s best to ensure you take it at the same time.”
Like I’ve said, doing it this way can turn a sinusoid into a DC value. You simply don’t get a good indication of what is going on with the temperature. In addition, you totally miss daily weather impacts which can drastically change the temperature that you get. If on Day 1 there is full sunshine at the time of reading and on Day 2 there is temporary cloud cover at the same time of day you get a biased set of data that does not represent the actual temperature profile.
Again, this is why the degree-day measure if far more appropriate for temperature. The Argument to Tradition means *never* changing how the climate is analyzed – even though other disciplines have made the change. Climate science will remain stuck in the 1700’s way of analyzing temperatures!
“Better is to have an estimate of the average temperature for each day,”
You can’t get a daily average from one reading which is what you propose.
Nor can you get a good “average” from Tmax/Tmin. Different Tmax/Tmin readings can give the same mid-range value from different climates. That’s *not* a good way to try and analyze climate.