Watch Anthony Watts discuss his latest report at NOON CDT
A new bombshell report found that 96% of the National Oceanic and Atmospheric Administration (NOAA) temperature stations produce corrupted data due to purposeful placement in man-made hot spots. These stations fail to meet the published standards deemed “acceptable” by NOAA itself.This nationwide study follows up widespread corruption and heat biases found at NOAA stations in 2009, and the heat-bias distortion problem is even worse now.This report eviscerates the media’s narrative that the Earth is seeing runaway global warming.
Has the Earth warmed over the last 20 years? Yes. Has it warmed to the degree the media claims? No.
And most importantly, is it catastrophic? Absolutely not.
Anthony Watts, director of the study, takes the reins for today’s Climate Change Roundtable to provide definitive evidence that the temperature record is flawed. You don’t want to miss this. We go live with the findings at 12pm CT today.
Any chance of a transcript in due course?
I suppose it shouldn’t be surprising to hear people in cities complaining it feels hotter, because it IS hotter in cites from the Urban Heat Island (UHI). Since a healthy majority of people in the world now live in very large cities, it is easier to convince them the whole planet is warming more, when their urban/city surroundings are actually a whole lot warmer than the rural countryside. So it is true for the people in the cities that their world is warming. But the city urban interface is only a tiny fraction of the percentage of the global land surface.
Undoubtedly a lot of the temperature record has suffered a similar fate over many years of official temperature measurements. And then the adjustments. Not only corrupted data collection from poorly cited and urban placed measuring, but then someone fiddling with the data to say what they think it should have been, so it shouldn’t be surprising that we get ‘warming’ in the data.
Getting an honest temperature measurement has to be the goal of every meteorologist and especially for real scientists that then use that data to advise politicians on national and global policy to now try and ‘change the weather’. All this IS the crux of the matter.
It was hot during the summer in large cities in 1960m 1980, 2000 and 2020.
It is always hot in the summer in big cities.
But people live the anyway.
Warming from UHO is most likely to have much effect on an inner city weather station from let’s say 1970 to 2020. But a rural station in 1970, that is now surrounded by a small town in 2020, may have a large UHI effect
Of course 70% of the world is oceans, where UHI does not apply.
Economic growth in the vicinity of any weather station could cause some UHI.
Even formerly rural weather stations.
Perhaps what’s needed is a graphic of how original recorded temp measurements are “treated” to arrive at the derived published numbers that bureaucracies and media serve up to the public –
And all the time we’re being warned about the evils & dangers of “misinformation” and “disinformation”.
You missed out “fiddling with the data”.
That wasn’t all that ex IPCC Chairman Rajendra Pachauri fiddled with.
“Perhaps what’s needed is a graphic of how original recorded temp measurements are “treated” to arrive at the derived published numbers”
I don’t have a graphic. But I do have a computer program, described here. You can do it all yourself; it isn’t hard. You can skip the homogenizing and still get much the same results.
So why do it at all Nick if the results are much the same?
Just work- making with taxpayers’ $$$$s.
“So why do it at all”
OK, don’t. But do something.
Do what?
The world needs more CO2 not less. The world needs more fossil fuels for humanity to thrive.
What are your answers? Poverty for more? Do you want to keep Africa down too?
Shame on you.
Much the same 😉
The outraged, irony-free, paid-for headline: It’s ‘worse than we thought’’
When put under the faintest of light, in fact fizzles out to:
“You can skip the homogenizing and still get much the same results.”
I’d want to get paid a lot in exchange for my honour.
Mr. Watts is doing surprisingly little to address the CRN data considering the fact that it undermines the implications he’s trying to draw in his new report.
It isn’t. It’s the adjustments in the past when CRN didn’t exist which is the main problem.
Except Mr. Watts is trying to point out siting issues in the present day station network. If those problems exist in the present day and they are not adequately addressed by the adjustments then we should see a substantial difference between ClimDiv and USCRN, but we don’t. Attempting to shift the goalposts is not an acceptable response, here.
You believe “averaging” removes all error.
HAHAHAHAHAHAHAHAH
I believe that a larger sample size has a narrower uncertainty in the mean than a smaller sample size. Is that what you were struggling to articulate?
“Is that what you were struggling to articulate?”
No. The minions here struggle to avoid this truth.
Word-salad blob also thinks averaging removes all problems, not a surprise.
The mean of what, exactly?
What is the size of your sample?
Whatever it is you are trying to estimate the mean of. This is a general statistical principle.
Where is the population with a normal distribution from which you are taking random samples?
The Central Limit Theorum shows that this is the case regardless of the distribution of the population.
The CLT cannot be applied to adjustments, as they are a systematic error
The fact that the climate community think they can is astonishing.
Adjustments are used to remove systematic biases. The remaining biases are random, which is addressed by adequate sampling.
The adjustment is a introduced bias, an estimated bias.
It’s a systematic error for the purposes of calculation of uncertainty.
There isn’t another field of science where not treating as a bias would be acceptable.
Just climate sophistry.
The adjustment is not an introduced bias, it is the removal of a bias. We identify a systematic bias in the data and then remove it. The uncertainty is reduced by the removal of identified systematic biases.
What a liar.
Yes it is. Science 101. Any adjustments to raw data is treated as a systematic error for the purpose of calculation of uncertainties.
Your argument could be extended to any data set, where you decide the data is wrong, use any figure you want to change it to fit, and claim super precision.
Pseudo science method
The removal of systematic bias is not the addition of systematic bias. It boggles the mind that I needed to write those words.
That assumes you know what the systematic bias is to start with. Guess work.
It truly boggles the mind that you think they can create perfect data sets by adjusting old data with unknown factors from behind a computer screen.
Nobody thinks you can create perfect datasets using adjustments, but you can certainly improve the data by removing identified biases.
“but you can certainly improve the data by removing identified biases.”
Or just as easily make the numbers less accurate.
I guess you never thought of that?
Adjustments change actual raw data into a judgement of what the data would have been if measured correctly in the first place. Data no longer exist after adjustments,
Human judgement has taken its place, and human judgement (for adjustments and infilling) is subject to bias.
You clearly never studied science in school.
You ran away from answering the question, what a non-surprise.
The truth you are NOT sampling a population and stats 101 stuff doesn’t apply.
You are abusing the math.
Not with systematic bias they don’t.
Larger sample sizes have narrower uncertainty: That’s only true in the case where there’s no systematic error. If most of the stations are compromised, then there .. is .. systematic error.
Its the change in site conditions which is the problem. Using a 20 year period with stations that haven’t been subject to these changes during their lifetime does not mean that the problem didn’t exist in the past, and does not validate the algorithm in the past. Classic sophistry.
There’s a lot of bots voting at the moment. All Mann supporters I suspect.
Yep, data mannipulators.
Its sad that they have to rely on bots voting.
Even loydo has positive votes.
The same siting problems existed in the past.
Unless the laws of physics have changed.
Then the adjustments that are addressing the siting issues today are also accurately addressing historic siting issues.
I suspect the CRN network has forced them to constrain their adjustments to so they match for the last 20 years, but it’s just a selection of parameters, tuned If you like. Outside that 20 years it almost certainly diverges just the climate models do
Your suspicion seems based on nothing but speculation. You don’t have any evidence of the things you’re claiming. That isn’t rational or scientific.
Your comments are not rational or scientific either.
We have evidence that 96% of weather station siting is deficient, That was the subject of the article.
Did you read the article?
Weather statin siting is external evidence of
scientific integrity. The evidence clearly shows
that siting does not matter to NOAA, has not been improved since 2009, and no organization that cares about accuracy would allow so many poorly sited weather stations. Such an organization can not be trusted.
That logic escapes you because you are not very bright — you are a government defender — if the government is in charge of something, you merely assume the work is first class You are a trained parrot of climate alarmism.
the siting is deficient according to Watts and other volunteers, but any siting deficiencies do not appear to have any adverse impacts on the station network. The ClimDiv network is in perfect agreement with the reference network.
To paraphrase – ‘The citing of weather stations is incorrect according to our own rules, but the data they produce matches this other data set. Ergo, we can ignore our own rules.’ That simply means that your other data set is now suspect, not that the station cites are OK.
It’s perfectly rational. How do you think they come up with these adjustments?. It’s not like they have a time machine to go back in time and verify them.
You don’t appear to trust NOAA
But you do trust their USCRN numbers?
That is a contradiction
I have the same trust for the temperature modellers as I do for climate modellers. All the adjustments are modelled.
How do you know the magnitude of ALL biases, including the ones that drift from measurement to measurement?
Are you psychic?
You do not know the magnitude of all biases – you remove the biases you can identify.
The truth is, you don’t know ANY biases.
You REDUCE bias by using properly sited weather stations whose data is compiled by people with no bias.
Not improperly sited weather stations whose data is compiled by people who all believe in a coming global warming crisis, have predicted rapid global warming for many decades, and want the temperature numbers they report to the public to verify the climate predictions they have been making for decades.
If NOAA had integrity, they would never predict medium or long-term climate. Humans have never demonstrated any ability to correctly predict medium or lomg-term climate or weather trends. Any weather prediction beyond a week or two is likely to be wrong. NOAA’s Atlantic Hurricane forecasts are usually wrong. I consider them to be a medium-term prediction.
But you don’t know the magnitude of all the bias, they are all estimates, thumbs on scales estimates.
More AlanJ baloney
There’s no evidence to declare adjustments are offsetting siting issues today. or were ever offsetting siting issues in the past. You just issued an uninformed data-free personal opinion.
There very much is: the adjustments bring the station network in line with a pristine reference series.
COMPLETE LOGICAL FALLACY NONSENSE
“If those problems exist in the present day and they are not adequately addressed by the adjustments then we should see a substantial difference between ClimDiv and USCRN, but we don’t. “
You are claiming that if A is similar to B (since 2005),
then both A and B must be accurate.
That is the logic of a dingbat,
not a scientist/
Anthony has claimed that B is “perfect.” It is a carefully sited and maintained reference network. If siting issues in A were not being adequately addressed by adjustments we should see a difference between A and the “perfect” reference network.
if B is perfect and A agrees with B then A is not plagued by substantive issues. None of you have even attempted a rebuttal to this simple truth.
Why should anyone trust the mannipulations of a dyed-in-the-wool adjustor as you?
Do you think Anthony Watts is a dyed in the wool adjuster? Why do you think he cited the CRN as perfect?
About 3 cents then
And don’t forget to use TheWayBackMachine. ( archive.org/web/web.php )
It’s worth looking up the word “Kriging” on Wikipedia here. Statistics is a dark world to most.
These are all methods of producing ABSTRACT figures with little connection to reality.
IMO: generally; as most people fear mathematics, they tend to treat these figures as somehow being ACTUAL, when they are not; hence false concepts and Memes can and do get created.
Stats are meaningless without context, misapplied stats is doubly so.
YOU MISSED ONE
Bottom line: the full network is perfectly consistent with the US Climate Reference Network (described as Anthony in this interview as being a “perfect” network). There is no reason to think siting issues are not being addressed by the adjustments applied by NOAA.
Adjustments to recorded data are data tampering. The correct way to handle suspect data is to widen the uncertainty limits. If that makes it unfit for purpose, then it is unfit for purpose. Tough luck, get a new record of data that has better methodology which makes it fit for purpose.
You nailed it. But you’ll never convince the climate alarmists like Alan that their GUESSES at what the temperatures should be are not as valid as the actual measurements themselves.
The adjusted data are completely consistent with the, as Anthony says, perfect USCRN data.
It would be wonderful if we could travel back in time to set up the CRN before the turn of the 20th century, but we can’t, so at least we can sleep easily knowing that the historic data agrees with the perfect reference network.
Nonsense. Adjusted data points that agree with different observations are still adjusted data points. Coincidence is not measurable as there is no standard of coincidence to compare to. The tolerance tree of traceablity between the two does not exist and is lost. You can’t point to a method that has no tolerance traceability and claim it’s valid. It is no different than saying “trust me”.
Nobody is saying the ClimDiv data aren’t adjusted, we’re saying the adjustments bring it perfectly into line with the perfect reference data, so it would appear that the adjustments are performing exactly as they’re supposed to perform.
And there is no need for trust – the raw data are available, the methodology is meticulously described in the literature. If you don’t trust the NOAA to perform the adjustments as they claim then go do them yourself. Produce your own surface temp analysis, describe your methods, and post it for all of us to see. Nick has done exactly this and (totally unsurprisingly for those of us not of a conspiracy-theorist mindset), the results are exactly in line with what the NOAA is saying.
And I’m saying that it does not, and that you are a liar.
You see the image above comparing ClimDiv and USCRN and you see that they are in close agreement. There is little else to say on the matter. The adjustments bring the network into agreement with the “perfect” (Anthony’s word) reference network.
Case closed.
Again, if the “adjustments” have no effect, then why bother with them?
And if you understood much of anything about uncertainty, you might understand that all your adjustments INCREASE measurement uncertainty.
The adjustments do have an effect – for CONUS they bring the network into perfect agreement with the pristine reference network. But whether the impacts are big or small (at the global level they are small) it is important to address systematic biases in the data.
Are you really this blind?
“pristine reference network”
Assuming USCRN is properly sited,
which NOAA claims, does not prove
USCRN data presented to the public is accurate.
If NOAA can not be trusted, then ALL their data
can not be trusted. You don’t understand that,
because you are not very bright.
Our host and chief surface station auditor Mr. Watts claims the USCRN is pristinely sited. You’ll need to take that argument up with him I’m afraid.
I do take note of your gradual pivot into wild conspiracy theory though, that is not the sign of someone holding a well defended position.
There is very little else to say — you have said very little so far. You have arbitrarily declared USCRN to be perfect, and then claim that proves all similar temperature datasets are near-perfect.
That is the failed logic of an anti-science dingbat.
“the perfect reference data,”
Merely declared to be perfect does not make a dataset perfect. Your lack of logic falls apart faster than a cheap suitcase.
Epicycles were proven “correct” by Tycho Brahe. Any ‘adjusted’ data is suspect by its very nature.
Or we could just embrace global warming and cooling and defund climate science.
I Anthony ever said “perfect USCRN data”,
then he was stating a personal opinion,
not a proven fact.
And I really doubt he did say this, it was probably more along the lines of “less bad”.
Lol you haven’t even watched the video that is the subject of this post? Maybe you should do that before commenting too widely.
“And I really doubt he did say this”
Here he is on the previous thread:
‘ “By contrast, NOAA operates a state-of-the-art surface temperature network called the U.S. Climate Reference Network,” Watts said. “It is free of localized heat biases by design, but the data it produces is never mentioned in monthly or yearly climate reports published by NOAA for public consumption.’
“Adjustments to recorded data are data tampering. “
In fact the adjustments in this period make very little difference anyway. I did a calculation here which matches the NOAA plot shown. But it also does the calculation using the data with adjustment (GHCN_a) and without (GHCN_u). They are very similar, and both agree with USCRN.
A simple request Nick:
Let’s see that same plot that you posted using the raw datasets from both USCRN and ClimDIV, and how about at the same time we go to a 7-day running average instead of using a camouflaging 12-month running mean.
Why a seven day running mean? That would by the minimum time span to smooth out UHI effects that would vary with urban heat-release activity variation over the five normal workdays versus Saturday/Sunday weekends.
The plot is based on monthly averages, which is what you get in ClimDiv/GHCN. It uses raw USCRN (there isn’t any other kind) and GHCN_u is the unadjusted GHCN, which is pretty much the ClimDiv set.
Here is the plot without 12-month smoothing:
An infilled number could be called “raw: but it is not raw data. It is a raw guess made by people who predict rapid dangerous global warming, and would like to see rapid, dangerous global warming in their numbers, confirming what they have predicted for many decades.
The data is the actual temperature measurements. They sample the US very well. If you don’t allow any deduction about regions outside, you can never make use of temperature measurements at all.
“They sample the US very well. ”
96%poorly sited is ” well ” ???
With the NOAA USCRN as a check and balance, one would assume that NOAA’s other datasets would be similar — they would not want to publicly contradict themselves. But that is a risky assumption.
Of course the first step in data analysis, which Nick Strokes could never even imagine, is to judge the credibility of the people who compile the data.
Do their actions demonstrate a lack of bias and a strong concern for accurate measurements?
For NOAA and NASA-GISS, my answer is “no”
Poor siting of US weather stations
Claiming pre-1900 N.H. numbers are “global”
Claiming false margins of error of +/- 0.1 degree C.
when there is lots of infilling of numbers
that can never be verified.
Hiding the amount of infilling in their numbers,
especially in the 1880 to 1920 period.
Making predictions of rapid dangerous global warming
for decades … that never happens
That adds up to organizations who can’t be trusted.
,
Therefore, no logical reason exists to trust
USCRN data.
Since NOAA can’t be trusted, then USCRN can’t be trusted, and it USCRN can not be a check and balance for other NOAA temperature data compilations.
All could be wrong — similar, but wrong !
Because the organization can’t be trusted.
I’m confident most people here would agree with this argument, except Nick Strokes, who defends government bureaucrat temperature data las if he is in charge of compiling it, or at least on the NOAA payroll.
Yet when asked questions about the data which require numerical answers, Strokes responds with generic word salads. So it seems that people who blindly defend NOAA and NASA-GISS data can’t be trusted either!
The coverage is excellent. The “poorly sited” is a matter of WUWT opinion.
There were weather station photographs in 2009 and I’m confident there are new photographs this year too.
Do you now claim all the weather station photographs are changed or faked?
You are making yourself into the laughingstock of this website by blindly defending all b government bureaucrats, even though unable to answer basic questions about temperature data quality.
You defend government numbers like they were your own and act like you are on the NOAA payroll.
Nick, so you agree there has been no warning in the continental US over the last 15 years or so?
I gave in the previous thread the observed trends 2005-2020:
USCRN 3.10 °C/century
ClimDiv 1.96 °C/century
These are actually quite high trends, but very uncertain because of the short period. FWIW, USCRN gives the higher trend, even though the point values of the two data averages are very close. Again that reflects the uncertainty of the short interval.
Check it out, “trends per century” sans ANY uncertainty limits.
Infilling is the biggest “adjustment” of all.
How is that accounted for in your chart?
It’s not.
Without infilling, you have just the temperatures inside a few thousand boxes. The only way you can relate this sample to the region they are sampling, is by infilling. ClimDiv and I may do it with different method, but we get the same results.
Correct. That’s all you have. To infill you have to make assumptions and the number of infilled values exceeds the number of datapoints.
Infilling does not create raw data.
It is a guess of what data would have been if correctly measured in the first place and can never be verified.
The infilled numbers are whatever government bureaucrats want them to be. If the bureaucrats are biased pro- warming, their infilled number may reflect their bias.
(1) More important is knowing how many surface grids have infilled data in a typical month.
(2) And the average number of data points that are infilled in an average month.
You come here implying you are an expert on temperature data collection.
If you are an expert, you will know the answer to my questions (1) and (2) on exactly how much infilling is done.
I’m interested in the right answer.
Not just any “answer”.
(3) And I’d like to know how any margin of error for the global average temperature can be calculated if there are infilled numbers that are never verified for accuracy.
It is my judgement that you do not know the answers, and therefore you are no expert on temperature data.
Without knowing those answers, it is puzzling why you try so hard to defend the surface global average temperature calculations.
How many other nations have a weather station network with quality siting, like USCRN, that since 2005 can be a check and balance for the USHCN numbers
Of course that statement assumes USCRN is correct,
which could be completely wrong
“(1) More important is knowing how many surface grids have infilled data in a typical month.”
No, that is totally unimportant. Grids are an artificial construct. If you make small cells, many will be empty. I repeat, all you have is point readings. Attributing the grid temperature to that of the measures within is already infilling. Every point that you haven’t actually measured is infilled. And the point of proper sampling is that you can do that.
So the answer to (2) is that they are all infilled.
On (3), you can verify, and people do. The justification for infilling is that you can accurately interpolate. So you can take half the readings, say, and see if you can interpolate the other half. If you can, then interpolating from the full set of readings is likely to be effective.
The classic paper on sampling uncertainty, the main uncertainty in global temperature, is Brohan et al 2006
You have provided a word salad response to that completely ignored the required numerical answers to my two questions.
Therefore, it is obvious you do not know the answers.
If you have no data about the amount of infilling, then you can not be an expert on temperature data.
You come here implying you ARE an expert on temperature data.
We can all see that implication is false.
In fact, Nick Strokes, you are a trained parrot defending all government bureaucrat climate data, without knowing enough about the data quality to provethat you know what you are talking about
You, Nick Strokes, are a science fraud.
Richard Greene
Bingham Farms, Michigan, USA
My own climate science and energy blog
has had almost 329,000 page views.
Honest global warming chart Blog (elonionbloggle.blogspot.com)
Nothing you write will ever appear there,
because you can not be trusted.
My job as editor is to determine which authors
know what they are talking about.
Anthony Watts qualifies.
You would not pass the audition.
Here’s a bonus question on UHI
for you to guess, or ignore,
Mr. fake temperature data expert.
As of two years ago when I wrote a report
on NASA-GISS UHI adjustments for my blog,
I discovered the effect those adjustments
would have on a climate trend over
a 100-year period.
Why don’t you tell us what the net NASA-GISS UHI adjustment was a few years ago, or is now, and how much it would affect a 100-year temperature trend?
I’m looking for a numerical answer in degrees C. or degrees F., not your usual generic word salad,
Mr. “data expert”.
I agree with your assessment here, Mr. Greene.
To this point, the individual station series are not being infilled for missing data points when anomalies are being used – this is one great benefit of using anomalies.
The “infilling” being done is simply that every square mm of earth’s surface is not blanketed by weather stations, and so we necessarily use the places where we do have measurements to represent all the spaces in between. That this is valid for temperature anomalies is common sense from our every day experience – if it is a hotter than normal day at my house it is almost definitely a hotter than normal day a few miles from me, too.
This approach is common in many geospatial applications – for instance I don’t need to measure elevation at an infinite number of dimensionless points on a hillside to get a very robust estimate of the topography – I can measure a finite number of points and infill between them all. There are limits (two points is probably not enough), but those limits can be systematically assessed and we can perform adequate sampling.
Yes, another content-free word salad.
But any adjustments are technically a systematic error, and there should be treated as such in the calculation of uncertainty.
This means don’t use CLT.
“If it is a hotter than normal day at my house it is almost definitely a hotter than normal day a few miles from me, too.”
What if your house has a black roof and no trees around it, while a few miles away the house has a white roof and is surrounded by lots of trees?
What if there are no weather stations at all in your neighborhood (grid) ?
And would you claim a +/- 0.1 degree C. margin of error when there is infilling that can never be verified for accuracy?
And could your guesses ever be checked to see if you have bias — perhaps believing in a coming climate crisis and WANTING to see the fast rate of global warming, that you have predicted for many decades … a belief required to get your job as a government bureaucrat “scientist”?
Your “analysis” completely misses step one of any data analysis: Can the people collecting the data be trusted?
Why?
“we get the same results”
Are you on the NOAA payroll, or NASA-GISS?
Getting the same results does not prove both results are accurate. That is basic logic that escapes you.
Nick that’s just over 20 years using stations that are kept in the same conditions ie with fairly constant UHI effects.
It’s the stations in the past which which weren’t as closely monitored which are the problem. It’s the change in UHI over the lifetime of a station which is the problem.
You’re being disingenuous yet again.
Looks like someone is offended, a reply explaining why would be nice, but they’re probably a bot.
He knows what he is doing.
The claim here is that inspections, this year, invalidate the measurements. And they clearly don’t; the measurements are good. You can’t evaluate problems of last century by looking at the current state of stations.
More sophistry.
Then all future adjustments to the data record are completely unnecessary and therefore fraudulent. I say data adjustments to the record will continue to be made, just as they have been to the entire 21st century record.
Logical fallacy: Appeal to Authority
Your inability to comprehend the written word does not manifest fallacy.
There is a reason to think siting issues are not being addressed properly, when rural stations are adjusted upwards to bring them in line with nearby and not-so-nearby urban or airport stations. There is another reason, when past readings are adjusted over and over and over again.
And the fact that the adjusted stations are in line with the reference network strongly suggests that siting issues are being addressed adequately. It isn’t clear to me why this is not addressed in Anthony’s live discussion today or in his report.
Logical Fallacy: Begging the Question or Circular Reasoning
Please refer to https://www.logicalfallacies.org/.
See also Appeal to Authority in that reference.
Logic fail for AlanJ (AS USUAL)
Merely assuming USCRN data are accurate
does not prove they are accurate.
Proof that USCRN weather stations are properly sited.
does not prove that USCRN averages presented to the
general public are accurate.
Correct. In Climate Science, most especially in the temperature domain (as opposed to Sea Level Rise or Ocean Acidification), there is a lack of a control group, or rather, the control group (i.e. undisturbed stations) is eliminated by absorption. Instead of trying to explain the difference between the undisturbed stations and the ones that are influenced by human activity in theoretical terms, the difference is presumed to be an artifice that needs correction. I think that may be because theory states that CO2 is well mixed, so it’s difficult to explain why the undisturbed stations aren’t showing much warming. They tend to falsify theory.
“Logical fallacy: Appeal to Authority”
No, it is an appeal to evidence. And it is evidence that no-one here seems to want to deal with. The ClimDiv analysis, based on these supposedly corrupted stations, agrees very well with the USCRN record, which is not adjusted at all.
So a no-gain exercise.
Why do it at all?
If the measurements are affected by non-climate cause, you have to adjust for it. There is no choice. Even though the adjustments are likely to be both and up and down, and so cancel, as they do.
You crack me up dude.
Aw Nick, after all our back and forth here on WUWT, I get that you’re totally invested in the taxpayer funded temperatures craps shoots.
But I’m a disciple of human enterprise, productivity and achievement.
So I guess we’ll always see things differently, because I DO believe in choices.
(the 1st choice option always being – call bullshit and move on to more important matters)
Of course there is a choice. The choice is to toss the data or widen the uncertainty of the data and live with it. But that defeats the true objective, which is to use the data whether fit for purpose or not.
“whether fit for purpose or not”
But the data clearly is fit for purpose. It gives results in very close agreement with USCRN.
There is no logical reason to assume USCRN is accurate.
Because NOAA says so, and Nick Strokes says so,
is not good enough.
OK, here is the monthly comparison of these two different sets of stations. You can see glimpses of blue peeping out behind the red. Pretty remarkable agreement for two supposedly inaccurate datasets.
So, no hockey stick panic?
Why then the hundreds of trillions of dollars (that don’t exist) to go Nut Zero?
And still no uncertainty limits provided.
Nope, Nitpick Nick don’t do them, his graphs are platinum.
WUWT shows the same graph (without ClimDiv) on the front page (right bar) in exactly the same style.
Agreement does not prove datasets are accurate
You are not conning school children here, Strokes
Te people who visit this blog are educated and they try to be people who consider data quality and the integrity of the people compiling the data.
They are not trained parrots of climate alarmism, and blind defenders of all government bureaucrats, like you are
“You are not conning school children here”
OK try conning people into believing that two erroneous independent data sources could just coincidentally come up with this kind of agreement:
Look Ma, no hockey stick!
Didn’t I read from you, on this thread
“The US has lots of weather stations and one very good USCRN network. It is most of the rest of the world that does not have such great weather station coverage. Or a network properly sited like USCRN.”
So Richard Greene says so is not good enough either?
Two datasets are similar
Therefore, both must be accurate
That is nonsense non-logic
As expected from you, Strokes.
The US is under 2% of the total surface area of the planet Earth. Strokes. Even less for the 48 contiguous states.
What about the other 98%+ ?
The US is the subject of this station survey.
The US is claimed to have the best weather station network. with exceptional coverage. If 96% of US weather stations are not properly sited, in the “best” country, that suggests similar bad news
for the rest of the world.
“Close enough for Government work”
It’s seems more important to have a lot of stations than to have properly sited stations, just so NOAA can say we have a lot of weather stations. As if that proves data accuracy.
From https://data.noaa.gov/dataset/dataset/noaas-climate-divisional-database-nclimdiv/resource/477062bf-4e49-4e80-80d9-12d8137360cb:
You’re comparing a dataset that is more than 100 years long to one that is on the order of a decade long.
Yes, but this survey is of the current state of ClimDiv stations. That should, if it has any effect, show up in the last decade results. But in that decade, the ClimDiv average matched the USCRN average very well.
That may be the Texas Sharpshooter fallacy. Independently established measurement standards are not falsifiable by looking at the data collected in violation of those standards. You are, in effect, suggesting that the clearly established measurement standards are invalid ex post facto. Steve McIntyre spent years arguing against this practice. If you don’t observe your own standards, then you have defeated the purpose of having standards.
This comes back to Feynman’s principle, often quoted here. You have a theory that some supposed standards breaches should have wrecked the results. But the observation is that the results matched USCRN, which everyone seems to accept as correct. That is reality. Your theory was wrong.
Logical fallacy: Burden of Proof
The breach of standards invalidates the data. Period. You don’t know what the data would have been if the measurement standards had been observed, because that data does not exist. YOU are the one asserting that this non-existent data would have been the same as the data obtained by breaching the standards. A real scientist would go back, correct the errors and repeat the experiment.
Instead, you excuse not following the standards by inverting the burden of proof.
The proof of the pudding is, as they say, in the eating. The supposedly defective stations produced a result that agreed very well with USCRN.
“The breach of standards invalidates the data. Period. “
That’s your theory. And of course the supposed breach is as evaluated by WUWT. But standards are only precautions you take in advance to try to get things to come out right, in your judgement. What actually eventuates is a matter of observation. And again, despite all the fussing about some asphalt here or some rocks there, the result agreed with USCRN.
“USCRN, which everyone seems to accept as correct.”
Consensus science again?
Must be true because most people believe it is true?
Exactly.
Maybe both are wrong since 2005?
That would never occur to a biased person like you.
Again, here is the plot. If both are wrong, the agreement would be a very remarkable coincidence
Oh look, still no hockey stick.
Where is the “climate emergency”?
An still no uncertainty limits.
WRONG AGAIN, AS USUAL, STROKER !
” the agreement would be a very remarkable coincidence”
WRONG
It would be remarkable if NOAA presented two datasets that did not agree. Ammunition for climate realists.
It is NOT a coincidence that both datasets are similar because NOAA would never present two datasets that are different. We will give them credit for knowing doing that would be a public relations disaster.
” the ClimDiv average matched the USCRN average very well.”
So what?
Noth datasets could both be wrong.
We give NOAA partial credit for
presenting similar temperature datasets,
and not contradicting themselves.
That does not prove all post-2005 data
are accurate, or that NOAA can be trusted.
Irrelevant
The US has lots of weather stations and one very good USCRN network. It is most of the rest of the world that does not have such great weather station coverage. Or a network properly sited like USCRN.
The climate scaremongering would continue regardless of data accuracy for current and past temperatures.
But the sorry state of the siting of many US weather stations does reflect on the integrity of the people who compile the data. If they do not care about the siting of their weather stations, then why would we trust the quality of their average temperature numbers?
“Irrelevant”
Certainly not. This post is about the supposed defects of the weather stations. The fact that they give the same average as USCRN is not irrelevant.
“If they do not care about the siting of their weather stations, then why would we trust the quality of their average temperature numbers?”
If you don’t trust them, do the calculation yourself. The data is there.
Nick Strokes defender of all government bureaucrats.
I complain about the quality of data and the amount of infilling, and your “brilliant” response is to tell me to do the calculations myself. You are quite a comedian.
Siting issues are external very visible evidence that accurate measurements don’t matter.
“I complain about the quality of data and the amount of infilling”
No, you complained of a lack of integrity, and that you do not trust them. It’s just a calculation. You don’t have to take it on trust from anyone. You can do it yourself.
Since NOAA publishes USCRN data since 1975 and USHRN data too, it would be very alarming if there was more than a small mismatch. But USCRN does not verify USHCN data from 1880 to 2005.
Even worse is that NASA-GISS claims to publish a global average temperature from 1880, when there were too few land weather station in the Southern Hemisphere before 1920, and not enough of them before 1950.
In addition, sea surface temperatures were mainly measured in Northern Hemisphere sea lanes with buckets and thermometers, with far too few S,H. measurements.
Not that long ago, pre-1900 average temperature data were described as “Northen Hemisphere.
Based on the surface coverage outside the US and Europe pre-1900, the phrase “rough estimate of the Northern Hemisphere average temperature”
would have been more accurate.
Now we have NASA-GISS “global” averages since 1880, and HadCRUT claims an even less believable “global” average since 1850. The “global” claim is science fraud.
Even worse is the claimed tiny margin of error.
+/- 0.1 degree C. the last time I looked.
Whatever margin of error is claimed is science
fraud — no margin of error can be calculated
with so much unverified data infilling.
They are not really global averages before 1900
— they are rough Northern Hemisphere averages.
so official claims that they are “global” averages
are science fraud,
Locations of land-based weather stations in the old days:
Honest global warming chart Blog: Poor distribution of land-based weather stations (elonionbloggle.blogspot.com)
“Since NOAA publishes USCRN data since 1975 and USHRN data too”
No, they publish USCRN data from 2005. What is USHRN? USHCN? That became obsolete in 2014. They publish ClimDiv to date.
You ran away from the tiny error bars, no surprise.
Well, he says that no error bars can be calculated. It’s hard to make progress from that. He gives no reason why the stated bars are wrong. But again, the proof is observation. Two different datasets give almost identical results. That is a good indication that the uncertainty is small.
Bullshite. You don’t have the slightest clue about measurement uncertainty, but you puff yourself up as the expert on everything.
“Two different datasets give almost identical results. That is a good indication that the uncertainty is small.”
Similarity of two datasets is not
an indicator or uncertainty.
You are either misinformed, or not very bright.
The Stroker is an expert on sidestepping questions
that require numerical or complicated answers.
Absolutely correct, he’s about as genuine as a three-dollar bill.
The current name is irrelevant
They arer two NOAA datasets, nitpicker Strokes.
The fact that they are similar IS EXPECTED.
What organisation that produced two datasets
available to the public, would allow them
to contradict each other?
“The fact that they give the same average as USCRN is not irrelevant.”
That fact is irrelevant,
and does not prove
that either dataset
is accurate.
You keep repeating the same nonsense.
This is bullshit because averaging throws away and hides variation. None of the Nick Stokes Graphs(TM) show standard deviations, much less real measurement uncertainty.
In fact the Nick Stokes Graphs (TM) are just variants of the graph that AW proudly displays on the right bar of the WUWT front page, with the same information:
The only difference is that he has masked out the ClimDiv result (see legend top right).
We don’t trust NOAA
Therefore, we don’t trust USCRN
We cite a huge percentage of poorly sited
NOAA weather stations as visible evidence of poor science.
We don’t trust you either.
You are a trained parrot defending all government bureaucrats.
Poorly sited weather stations?
No problem, says Mr. Strokes.
The government “adjusts” data to fix that.
Just thrust them
When asked for details about the temperature data
that he thinks he is an expert on
(a legend in his own mind)
Strokes has no numerical answers.
Just generic word salads.
Another clown show.
Adjustment master
After reading the article, I was curious as to the why the local high school weather station which outputs to weatherundergound always reads higher than my own weather station. The weather station at the high school is located 20 feet above the front entrance attached to the building. The concrete sidewalk directly below radiates the absorbed sunlight directly to that weather station for many hours after the sun sets.
If it sits on top of the building like our local high school then it also gets all the radiated heat from the building roof. This really wreaks havoc during the winter when the heated building pumps up their temperature measurements.
“You don’t want to miss this.”
Awesome, now I can expect this report to be mentioned in the mainstream media?
uh, probably not
“…96% of the NOAA temperature stations produce corrupted data due to purposeful placement in man-made hot spots”
Puerile lies (ably demonstrated on this site already) written by a shill on behalf of fossil fuel interests. Who would touch it with a barge pole. But he could have at least made it 97% for irony’s sake.
97%
BINGO!
We have a winner folks.
It’s Loydo, the lady in the polkadot onesie
Yet you come here to make a fool of yourself pushing fallacies that doesn’t address anything.
It is clear you can’t address the post because it is above your ability to do so.
“Yet you come here to make a fool of yourself”
And Loydo does a good job, if that is his goal
Are you sharing a brain with Simon?
No, but obviously we’re both occupying your’s. Hopefully we’re occupying Watts’ too.
Lol….not at all. Only when I need a laugh. You and Simon are not very bright. Good for a laugh for sure.
Next step is to replace “97% of scientists” in the media with “96% of weather stations”. Then it’s game over.
Nyuk, nyuk.
You can own the 97% proudly 😉
Good stuff, great work Anthony!
Skewing science for political purposes is nothing new. Past eugenics studies of cranial volume per race showed the same presupposed bias.
This is a big so what. So what if temps increased over recent times. Natural variabilities of the sun’s output is the likely culprit. Better warmer than colder. Trying to imply that CO2 is the culprit with no empirical evidence is a fools errand. Show us the proof or STFU. And btw, the Earth’s atmosphere is not a greenhouse and CO2 can’t retain heat like the fraudsters claim it can.
You have to have a bogeyman. How else you gonna get funding?
Settled science does not bring in money. Do schools have departments of gravity 🤓
The sun’s output has barely changed
The Earth’s atmosphere is a greenhouse
CO2 is not claimed to retain heat, it is claimed
to impede Earth’s ability to cool itself
by some unknown amount
that appears to be completely harmless.
On what planet did you study science?
Wow Richard, you are certainly full of yourself. There is a 2006 report out of NASA that claims the sun’s output has dramatically increased over the last century. Google it yourself. I don’t claim to be a solar physicist. Do you?
Define the amount described as “dramatically increased”:
Those aren’t my words. That is what the 2006 study from NASA says. Google it.
The only 2006 study I found did not say what you claimed:
FROM THE ABSTRACT:
“Holocene climate proxies suggest substantial correlations between tropical meteorology and solar variations, but these have thus far not been explained…. Using a coupled ocean-atmosphere-composition model …”
The key words are “climate proxies”, “tropics” and
“models”. This may not be the study you were referring to, but it is obvious what the NASA-GOSS position is today — declining solar energy for over the past 50 years.
Abstract is here:
Pubs.GISS: Shindell et al. 2006: Solar and anthropogenic forcing of tropical hydrology (nasa.gov)
It was from the full list of 2006 studies here:
Pubs.GISS: 2006 Publications (nasa.gov)
NASA-GISS shows lower solar energy since the 1950s.
July 2020 data
They present solar irradiance since 1880, but I can’t imagine any accuracy before the satellite age in the 1970s.
Graphic: Temperature vs Solar Activity – Climate Change: Vital Signs of the Planet (nasa.gov)
Correct—the only solar irradiance measurements that go back that far were ground, using thermopile instruments. The absolute cavity radiometer was not invented until the mid-60s (by a guy from JPL). A model would have to be used to convert the ground irradiance into top-of-the-atmosphere irradiance.
The article was from http://www.space.com. The study was conducted by a team of international scientists in 2006 examining meteorites. The study claims a significant increase in the sun’s output from 1880 through the end of the 20th century. Although it does say that the peak was reached in 1950 but the solar output remained at this higher threshold through the end of the 20th century.
More data – compare daily Tmax and Tmin raw from 2 prime sites a mile apart. Geoff S
http://www.geoffstuff.com/melbversusop.xlsx
Ooops, did not travel well from an early Excel version.
Graph looks fine if you download rather than just view in browser.
Or, easy to plot our own.
Geoff S
Yes, of course they are different. Those are the old Melbourne site (for which WUWT clamored for replacement) and the new one. They have run them in parallel for 18 months to check for the difference, before switching to the new one. The well-established difference of 0.8°C for Tmax will now become an adjustment. Do you think they should not make that adjustment?
Nick,
That difference between two stations a mile apart might average 0.8 deg C but that is 0.8 +/- 0.5.
How do you merge the 2 records while accounting for this uncertainty?
If you keep them as separate stations, fewer headaches.
If you merge them, you fail to recognise that the older one was rejected because of a rate of change that had become too fast from spurious effects like wind change from buildings, UHI etc. When, in the period of fast change, should you start the new record? Ooops, too late, the fast change started before the new station was set up, so the combined record had a splice at a time that was wrong because it introduced spurious warming in the record and in the trend FAIL.
Geoff S
Geoff you are spot on. Merging data, dropping data or making up data….Uhg 🙁
Geoff,
“0.8 +/- 0.5.”
Where did you get that? I get the standard error of the mean as 0.034.
So that means the true mean can never go outside that range. Of course it can.
No, it means what it says.
HAHAHAHAHAAHAHAH
Another clown who believes averaging reduces uncertainty.
Hint—your cherished “standard error” is NOT uncertainty.
Looks like the Bureau of Adjustors has an organized campaign to push the minus signs instead of answering questions.
milli-Kelvin uncertainties — HAHAHAHAHAHAHAH
Nick,
In an email from BOM as I interpret it.
Geoff S
Geoff,
I think that is the sd of individual days. The uncertainty of the mean is much less. That is why they average 500 days of data, rather than settling for just one.
Nick,
Are you trying to say that these numbers are IID?
Geoff S
I don’t think there is much autocorrelation.
Nick,
I repeat.
Are you trying to say that these numbers are IID?
Geoff S
They look iid to me. How do you propose to evaluate the uncertainty of the mean?
What is the obsession with moving stations, merging stations, adjusting data? A weather station only measures the micro-climate at that station at that point in time. Currently where I live there is 3c difference between here and about 400m down the hill. If the wind changes it could go to zero difference or even flip the other way. The point is they are 2 completely different micro-climate and no amount statistical analysis is going to be able to merge there records into a single coherent temperature record. You have 2 separate weather stations measuring different climates, they may as well be a 1000 miles apart. Comparing them for 18months is meaningless relative to the past as you have no idea of the new stations micro-climate in the past.
The big picture presented as simply as I can:
96% of current weather stations are sited improperly
We have had 50+ years of always wrong predictions
of a coming climate crisis, that never shows up
Imagine 100% of current weather stations sited perfectly for the past 50 years
We would still have had 50+ years of always wrong predictions
of a coming climate crisis, that never shows up
Predictions of climate doom would still exist.
The predictions are not related to any past climate trends.
Global warming since 1975 was mild and harmless.
The predictions are for future global warming that is rapid and dangerous.
The predictions have been wrong for 50 years.
But they will continue because people believe them.
When people fear the future climate, they demand that their governments do something. That demand is the goal of climate scaremongering. Climate change propaganda has achieved the intended goal of significantly increased government control over the private sector.
“Climate change” is not about science
Always wrong wild guesses
of a coming climate crisis
are not science.
This explains why government bureaucrat scientists could not care less that 96% of weather stations are improperly sited. They were hired to predict a coming climate crisis, unrelated to past global warming, so they do what they are paid for.
Just like scientists hired by cigarette companies
were paid to claim cigarettes were safe.
Governments get the science they pay for.
96% of current weather stations are sited improperly, not 6% !
(Edited the 6% into 96%) SUNMOD
Richard
Even if it was 6% my reaction would have been the same. Yet another example of why the establishment narrative is absurd. Whether it is SLR or historical temperatures or droughts or floods or oscillations or heatwaves or previous warm periods or the LIA, etc etc, there is research contradicting the IPCC. They would have had more credibility had they spent decades asking questions rather than trying to brainwash the public into believing they had all the answers.
Glad you caught that.
For a minute there I wondered if you infilled with Nick and AlanJ’s comments above!
Great stuff. Another example of not questioning the observations so long as they say what you want.
GOOD to bring this issue to the fore (forefront) again!
Deal: an agreement entered into by two or more parties for their mutual benefit, especially in a business or political context. As in green new deal where the people have been excluded from any decision making. Bad data is intended to support the green new deal.
I’ve not watched the video however here are two examples of contaminated data.
https://www.scotsman.com/news/weather/record-breaking-temperature-scotland-invalid-because-parked-car-584154
https://www.heraldsun.com.au/subscribe/news/1/?sourceCode=HSWEB_MRE170_a_GGL&dest=https%3A%2F%2Fwww.heraldsun.com.au%2Fblogs%2Fandrew-bolt%2Fcar-heats-weather-station-creating-warming-record-for-scotland%2Fnews-story%2F77cbd1ba1e1ab6e76a4568dfee0d1b7c&memtype=anonymous&mode=premium&v21=dynamic-cold-control-noscore&V21spcbehaviour=append
Has anybody else noticed the ‘collection’ of upvotes Nick Stokes [a controversy] has accrued the last number of hours? Seems like a brigade of simpaticos have worked to ‘do him a favor’ and give him “a more than fair election” …
Yeah … that same ‘brigade’ is handing out downvotes from what I have observed too … “a totally fair election” – right?
You are correct, the Nick Stokes groupies have organized a “resistance”.
One that is devoid of technical content.
I think it is just that the readership is filtered by attention span.
“devoid of technical content”
Votes are always devoid of technical content. Mine are mostly negative.
I noticed both the upvotes for Nick and downvotes for those who disagree with him.
It looks like they are doing a search for his mane then voting that thread.
And AlanJ too.
You can add +1 downvote to every Stroker comment for me.
I do not believe in voting on comments.
That creates bias for the reader, to skip the comment
or assume it is wrong.
It could be right,
but is just not a popular comment.
Interpret not a simple observation on my part as an advocacy towards activism; It is usually assumed on a site such as WUWT that all persons participating are adults, however, sometimes we find out a little differently, hence, my observation …
All this corrupted temperature measurements and not a peep in the Western msm.
Let’s just keep on sending $ trillions to China for sun mirrors and wind mills.
But let’s never forget that predictions of environmental doom
have been 100% wrong since the 1960s with no msm mention
of the wrong predictions, or no change to publishing scary predictions.
And the climate scaremongering would ave been the same
if US weather station data were perfect or not.
But the msm did mention Iraqi weapons of mass destruction
And how Trump colluded with Russians to win in 2016.
And that Covid vaccines are perfectly safe and very effective.
And that the 2020 election was the most honest election in history.
Those were big msm headlines.
How many msm readers would care
about weather station siting when the world
is facing a disaster from climate change
and our children may have to move
to another planet to survive global warming?
I’ve lived in Michigan since the 1970s
and have enjoyed the mild global warming.
But what would I know?
I’m not a climate scientist with a supercomputer
and a climate computer game. Just an ordinary
person who has lived in the same home
since 1987, who enjoys our milder winters
with less snow shoveling.
They don’t. Berkley Earth already proved it.
And data from outside the US, like very reliable UK data, shows the same trends as the USA
Then by all means, “turn yourself in to the Carrousel”.
Hasn’t the ‘color’ of the crystal in your left palm changed color? (Reference to movie “Logan’s Run”)
Griff,
I have posted several sets of data that show the Australian cities I studied do not comply with that Berkeley suggestion.
You should read them.
Compare raw with adjusted here, for example.
http://www.geoffstuff.com/eightheatwave2022.xlsx
Geoff S
UK / England temperature Record set at Coningsby Lincolnshire 19 July 2022 but hold on Hell isn’t that right next to the runway ? 53.094, -0.171 https://www.google.com/maps/place/53%C2%B005'38.4%22N+0%C2%B010'15.6%22W/@53.093954,-0.171183,709m/data=!3m1!1e3!4m5!3m4!1s0x0:0x8d1212c7c1e7f2b0!8m2!3d53.094!4d-0.171
The same for the welsh temperature record The Welsh High Temperature record broken as well. Blimey , do you believe in coincidences that station is next to the runway as well!!! 53.175, -2.985
https://www.google.com/maps/place/53%C2%B010'30.0%22N+2%C2%B059'06.0%22W/@53.17487,-2.9881228,706m/data=!3m1!1e3!4m5!3m4!1s0x0:0x539c380fe918659d!8m2!3d53.175!4d-2.985
I once saw a tweet by Chelsea Clinton castigating us all for our Co2 footprint. I did point out to her that her dad flew on Epstein’s Lolita Express so many times that he is the only man with a Carbon dick print
Anthony is avoiding the truth that the others know is true. It is not incompetence, it is deliberate The left set the agenda and they have an infinite capacity to lie.
Data as published in our local newspaper for Friday July 29 showed the highest temperature in the US to be 100 in Waco, TX while showing our local temperature to be a record at 112 in the Tri-Cities of WA. There is no statement of the data source.