Dingo in the Chook House. Beware!
Geoff Sherrington
Where is the claimed Australian Global Warming?
Where is the fingerprint of Australian Urban Heat Island effect?
Why is there only one Monopolies Commission?
Here is a puzzling picture. (click to enlarge)
Background.
Australia’s Bureau of Meteorology, BOM, released a big data product a decade ago. It showed the historic, daily, observed temperatures as maxima and minima in ⁰C for about 1,650 weather stations.
This is the foundational, raw data on which much climate research is based, for Australian temperatures.
My graph uses a sub-set of these foundational temperatures. It is for the 6 State capital cities, plus Darwin and Alice Springs to get more coverage of the country.
Melbourne 1855 is the oldest start date for the data. Perth, 1897 is the most recent start date. All data end in December 2006, because the finish date for the BOM compilers was April 2007.
As is customary (though I do not like this method) the image above shows linear least squares fits of regression lines, or simply trend lines. Missing data are not infilled. Outliers are not removed. Some stations have only one station number, meaning no large changes to the station location, while others have up to 3 moves large enough to earn new BOM station numbers. See the columns labelled ‘station number’ in the Excel file below.
I have downloaded and I acknowledge with thanks, a year 2010 BOM graph of the average annual and decadal temperatures of Australia, with my calculated trend line of 0.96 ⁰C per century equivalent added in blue.
These linear trends in temperatures over time are my main metric here. Trend is used because it should not change greatly with geography. The chosen units of trend are ⁰C per century to allow easy comparisons.
The BOM asserted in 2022 that “Australia’s climate has warmed by an average of 1.47 ± 0.24 °C since national records began in 1910.” That 112 years corrected to 100 years gives a linear trend of 1.31 ⁰C per century equivalent. Source:
http://www.bom.gov.au/state-of-the-climate/
Here are the calculated trends for these 8 cities graphed above, plus all Australia, in units of ⁰C per century but for the century (almost) of 1910 to 2006 incl. The data from 2006 to 2021 indicate higher temperatures than earlier that strongly affect the trend.
| Average Trend | ||
|---|---|---|
| Brisbane | 1887 to 2006 | 0.14 |
| Sydney | 1859 to 2006 | 1.06 |
| Melbourne | 1856 to 2006 | 0.94 |
| Hobart | 1883 to 2006 | 0.66 |
| Adelaide | 1888 to 2006 | 0.26 |
| Perth | 1897 to 2006 | 1.27 |
| Darwin | 1885 to 2006 | -0.6 |
| Alice Springs | 1879 70 2006 | 0.27 |
| All Australia | 1910 to 2006 | 0.97 |
| All Australia | 1910 to 2022 | 1.31 |
Populations today are roughly:
- Sydney 5.297m
- Melbourne 5.031m
- Brisbane 2.628m
- Perth 2.224m
- Adelaide 1.418m
- Hobart 253k
- Darwin 150k
- Alice Springs 30k
- All of Australia 26m
These 8 cities house 70% of Australia’s population.
The original, digital raw data for these 8 cities is in this Excel spreadsheet. It is big. Do give it time to load.
https://www.geoffstuff.com/capital 8 trends long.xlsx
The present station status.
Sydney, Melbourne and Perth have warming trends similar to the all-Australian average. How can this be? Since 5 others show little warming, there must be stations elsewhere in Australia that show even higher warming trends. How can such miscellaneous trends arise from mechanisms that are said to be global, as from a well-mixed gas CO2?
Melbourne. Recently, the old station 86071 was closed in 2015 with a press release about its relocation. “The location at Melbourne Olympic Park represents an improvement in the quality of meteorological observations available and meets international observing standards set out by the World Meteorological Organisation.” The new site 86228, some 2.6 km distant, showed on average to be 0.53 ⁰C cooler than the old one, averaged over a 583-day overlap period starting in June 2013. This difference could, by some ways of analysis, halve the Melbourne warming trend shown in the first graph.
Sydney. The old station was closed in 2022. BOM reported “The Bureau of Meteorology’s official weather observation station for Sydney is moving to a new location at Sydney Observatory Hill. The relocation is due to development plans for the original site”. Colleague Dr Bill Johnston in 2017-8 wrote extensively about Sydney in a report within the following link. He noted that “Site changes have a permanent impact on upper-range extremes and as step-changes in their ratio are significant and result from rainfall and site-changes acting together, it cannot be claimed that upper range temperatures have increased due to changes in the climate.”
Perth. The station in Perth city, 9034, begins the data. Data since May, 1992 are from the airport site. Colleague Bob Fernley-Jones (dec’d) reported on Perth adjustments in 2015.
In Darwin, there could be a site difference between the early Post Office station 14016 and the post-1942 airport station 14015. A neutral adjustor might subtract 0.28 ⁰C from all of the first station temperatures, to produce a zero trend outcome to replace the cooling trend.
Adelaide, Brisbane, Alice Springs and Hobart show little change in linear trend over the many decades to 2006.
What does it all mean?
There is no evidence that these simple trends can cause concerns about Australia heating up dangerously.
Alarming trends are absent from the simple, raw data, presented here Therefore, horrifying trends alleged by others can arise only from manipulating, altering or ignoring the foundational data.
Those seeking to use UHI to explain trends should be aware of a prime difficulty. In Australia, the historic data are not suited to such an exercise. They were collected with diligence and curated by the BOM, for which they are thanked. But they were collected for purposes like understanding farming conditions, airport conditions for pilots, local weather etc. They lack the accuracy required for elegant exercises, such as estimating UHI from the subtraction (Turban-Trural) between nearby stations.
The graphs above have RMS error estimates shown, to support that contention.
I have done years of work on these topics, so I have a large collection of analyses as well as this city set of 8 analysed in many ways, even Mahalanobis Distance methodology and bootstrap forests.
I have presented only simple findings here, but I am also well versed about complex findings. Subsequent articles will deal with analysis of 37 “urban” stations and 45 “pristine” stations selected for UHI detection.
The final lesson?
As with all scientific work, researchers are not permitted to rely on convoluted interpretations when the simple, primary data do not tell the required story.
First explain the simple, primary data.
Geoff Sherrington
Scientist, retired, Melbourne.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
All the trend lines look fairly flat, so if one is attributing temperature rise to CO2 levels, it is a very weak correlation.
fitting straight trend lines is a mistake because it assumes temperature is a function of Time
Like climate models.
True, the last 45 years the global temps should be indicated as 3 long pauses, interrupted by El Nino steps.
That would show there is absolutely ZERO warming effect from CO2.
Finally, moosh is getting there.!
Steven,
Yes, that argument has some merit. There are times when the dependent variable is related by physical mechanisms to the independent variable. Then you are correct.
There are also times when the independent variable can be viewed a proxy for the physical value. Then, there can be the appearance of lack of a physical link, but the plotted graph can still be valid.
Plotting these routine temperatures over time is, in a sense, using time as a proxy. Instead of time from the calendar as the independent variable, you can optionally use, for example, season of the year from a celestial calendar. Or distance between Earth and Sun from astronomy data. You might end up with similar results about how temperature changes. Geoff S
Graphing temperature against time and then fitting an OLS trend line to it seems to be the consensus approach.
What do you suggest as an alternative for the X axis? CO2 concentration, or the log of CO2 concentration might prove instructive.
Yes. One may even have to do a multivariate regression on a local and regional basis to begin searching for the correct combination.
You’d expect the rate of warming to relate to [CO2] (forcing flux). So time still comes into it. You should be integrating that.
You can’t integrate something that has no functional relationship. f(x) = degrees / time, where x = time. How about f(x) = degrees/ CO2. Neither of these appear as a linear function by looking at your graph. There is either other factors or no relation at all.
“You’d expect the rate of warming to relate to [CO2] (forcing flux)”
NO ! YOU WOULDN’T
That is a non-scientific mantra fantasy !
The usual (H1) hypothesis is that doubling the CO2 concentration will increase temperature by around 1.6 degrees C
The implied null hypothesis (H0) is that doubling the atmospheric CO2 concentration will have no effect on temperature.
Some of the alternative hypotheses are:
H2 doubling CO2 concentration will increase temperature by 3 degrees C
H3 doubling CO2 concentration will increase temperature by 4.5 degrees C
There is observational evidence to support H1, but the uncertainty may be too high to definitively reject H0. That was the essence of the recent Scafetta – Schmidt dispute, though that was more along the lines of H1 vs H2
“There is observational evidence to support H1″
No there isn’t. !
All observational evidence supports H0
Check out the Lewis and Curry papers, amongst others.
Correlation papers.
Not proof.
Observational evidence, as stated.
We could of course graph the “temperature adjustments” against CO2.
You said below that there is an increase in CO2 levels from 1950, so it’s not linear by year. There are effectively 2 legs.
Actually, I agree with Mosher on this one. It is why the temperature/time correlation will never show causation. When I was doing revenue and expense budgeting, I NEVER relied on a simple trend line vs time to gauge what might happen. The first question to ask is what the piece parts are that result in the change. If you have multiple streams of revenue, you can’t just assume they will all continue as they were. Same with expenses.
If someone had wanted to prove that CO2 was the cause of heating, doing a regression of some kind with CO2 vs temperature would have shown it a long, long time ago.
Plot the temps against the CO2 record, then.
Exactly! A trend of temp vs time may hint at a connection but that is all. U.S. postal rates vs time gives matching trend.
CO2 vs temp will show if there is a direct and casual link. Why is that so hard for climate science to understand?
Post-1957 CO2 data is available from https://gml.noaa.gov/ccgg/trends/ and https://gml.noaa.gov/webdata/ccgg/trends/co2/co2_mm_mlo.txt. NOAA also has CH4, N2O and SF6.
Law Dome ice core data is available at https://agupubs.onlinelibrary.wiley.com/action/downloadSupplement?doi=10.1029%2F2006GL026152&file=grl21511-sup-0002-ts01.txt
The CO2 concentration and average temperature can be cross-correlated to year to allow charting Tavg vs CO2 concentration.
I’ve only played with the Mauna Loa CO2 data (1959 on) so far, but it’s interesting.
The Tavg vs log CO2 is a noisy flat line or possibly negatively sloped up to about 330ppm, then noisy linear growth from there on.
Why the inflection?
It will be interesting to try with the Law Dome ice core data to extend it back to 1850, if I can find that.
What about the ice cores that showed CO2 lagging temps by, what, 700 years? What happened to that data? Ignored? Sneered at?
Couldn’t find it for easy download 🙁
Found it – https://www.ncei.noaa.gov/access/paleo-search/study/9959
Now to tack in the post-1850 CO2 and temperature figures.
My subconscious worked that out overnight.
I was being lazy, using Libre Office’s charting capability with CO2 as the X axis. Unfortunately, that gives equally spaced data points with the values as labels, so is no better than a time series 🙁
I wonder if there is still some log-linear graph paper hidden in the back of the stationery cupboard…
Playing about with this in Veusz, which seems quite a nice package.
Using a log X axis doesn’t make a lot of difference compared to linear over the CO2 range (320 – 420).
What is interesting is that the curve is more or less sigmoid: flat to around 330, linear increase to 400, then flat again. There are also big jumps in the transition zones.
The next step is to try some curve fitting…
Playing with new toys can be fun 🙂
Adding in the Law Dome CO2 figures from 1850:
There are quite a few gaps – seems to be pretty much bienniel
There are quite a few duplicates, so I arbitrarily selected from those.
The OLS regression fits aren’t bad, but the Land and Combined (Land + Ocean) have narrow peaks centred on about 310 ppm.
All 3 charts are flat or declining from 280 to 300-305, and flat again over 405 or so.
Temperature vs CO2 is overall a much better linear fit than temperature vs year. That appears to have 2 legs; flat until 1950, then rising.
Not that anybody is still following this, but for ln(CO2), I get the following regression trends:
Ocean -13.2 + 2.29 ln(CO2)
Land -29.0 + 5.03 ln(CO2)
Combined -18.06 + 3.14 ln(CO2)
All 3 have the spike at 5.75 (310 ppm)
For 560 ppm CO2 (double 280), that gives Tavg anomaly increases from 280 ppm of:
Ocean 1.6
Land 3.5
Combined 2.2
For 500 ppm, we get:
Ocean 1.3
Land 2.9
Combined 1.8
Assuming, of course, there isn’t an error in the calculations.
The Land and Ocean data series both exhibit the same temperature spike amplitude at 310 ppm.
That has to be some form of artefact, at least in the Ocean series.
The 310 ppm spike seems to be late 1940s.
The Ocean series has a higher spike around 1880, so 280 ppm or thereabouts.
I don’t believe that Melbourne ”trend” for a minute. No way is it hotter now than it was when I was a kid. My late father used to tell me it was ”much hotter” in the fifties than it is today. I have not noticed the slightest difference in average winter low measurement from my thermometer in the 20 years I have been here in a more rural area.
but, like Adelaide the amount of ugly hirise monstrosities blocking breezes and reflecting/holding heat is a hell of a lot more as are the populations in cars etc
and golly gosh after decades of ripping up street trees they finally worked out how important they were…cant fix stupid!
Well, Tie Me Kangaroo Down, Sport. Good report by Geoff, and another interesting glimpse at the UHI effects, which are in favor with the alarmists. I’m going to sing the song the rest of the day, and will only stop when my wife makes threats.
An original verse from that song has been ‘canceled’ now Ron.
See if you can find it, so you can sing the song in its entirety.
Mr., I think you might be referring to the term “Abo”, slang for Aboriginal. Think Jason Day, great PGA golfer from Australia, and part Aboriginal, and with a beautiful family.
Jason is part Philippine from his Mom’s side.
Ralph Harris got cancelled so his songs and art etc all got canned as well
idiot society we now live in
Rolf Harris. 🙂
When she bans that one I suggest to switching to “Waltzing Matilda”, it’s been an ear worm for me for a week since I stumbled across a video of an Ausi army unit (circa WWII) marching to it.
Funny, at international rugby (union) matches, the AU team sings the national anthem (“Advance Australia Fair” (gag)), but the fans around the arena sing “Waltzing Matilda”.
now go look for the pigin english version its even more catchy
beats the crap anthem we now have, whatever version;-)
here you go
and then this
Saw him play at a different venue, absolutely fantastic.
“Tie Me Kangaroo Down”
Got a laugh out of that- been ages since I’ve heard that song. 🙂
rule #1 Skeptics Always reduce the size of a dataset.
rule #2 they always ignore Structural uncertainty, also known as systematics
This is the foundational, raw data on which much climate research is based, for Australian temperatures.
My graph uses a sub-set of these foundational temperatures. It is for the 6 State capital cities, plus Darwin and Alice Springs to get more coverage of the country.
What an utterly meaningless comment from moosh !
Geoff Sherrington is quite explicit that these are the raw data, and outliers have not been removed. Why did you lie in claiming the dataset had been reduced?
Steven Mosher is NO skeptic. See him wait the word of NOAA…..
Steven,
Did you miss the words about a further 45 pristine and 37 urban sites about to come? Or the hundreds of sites studied?
You often have to reduce your data base. For these temperatures, there are some 1650 weather stations with data, but some lack required quality, have too much missing data, many extreme outliers, are too short to compare with other sites, etc. As it is, one of the “prime” sites, Sydney Observatory, has a month of daily data in July 1914, identical with another date, July 1915, for both Tmax and Tmin, cut and paste style. Some culling can result in improvements, rather than being a sceptical slash and burn.
I am unaware if I am classed as a sceptic and I do not know of your “rules”.
Geoff S
Utterly pathetic and meaningless comment, which only makes YOU look really stupid and perverse. Well done Stephen.
The lived experience is not reflected in what we are being told. I live on the Sunshine Coast in southeast Queensland. The winter of 2022 was the coldest, longest, wettest winter I’ve ever experienced. And today I would say I can’t remember an October this cool so late in the month. We are being conned.
Geoff,
You don’t need a special BOM product. GHCN unadjusted has the same numbers, and they are current to 2022. I have used data to 2018 here.
Here is a plot from 1850. I have omitted Perth, because there was a complicated site change. The legend shows the trends for that period.
The trends are not large, but no-one expected much warming before 1950. Here is the plot since 1950, where the trends are much larger:
eg Melbourne 2.75 C/Cen, Stdney 2.11 C/Cen. About in line with global.
The data is unadjusted, but sometimes adjustments are clearly needed. Adelaide moved from city center to seaside airport around 1950. Darwin moved around 1941. In each case there is a clear artificial drop.
So the atmosphere over Australia is cooling. Interesting…..
Mike
Not according to UAH satellite data. The trend of warming in the atmosphere over Australia is in line with global trends for land surfaces.
Australia is +0.17C per decade versus +0.19C globally. You can probably cherry-pick your start dates, but you’d only be fooling yourself in a Monckton-ish fashion.
What planet do we need to move to?
Whatever one Mike’s on.
He’s on earth. It appears earth is doomed due to heat, what do we do?
Grow up and face the problem, instead of denying it?
what problem?
See the first part of my reply above. Grow up.
I did see it. What problem?
That’s the problem. Some people get up every morning and feel doomed while the ones that don’t aren’t empathetic and woke enough about that.
“Grow up.”
Says the child with the mind of a 5-year-old. So funny !
There is no “problem.” Warmer is better. It’s better for humans, and better for biodiversity.
Are you that clueless to not understand that, or is someone paying you to post this nonsense.
You are the one acting like a petulant little child.
Scared petrified of basically ZERO warming.
“Grow up and face reality, instead of denying it?”
There, fixed it for you.
Pretty sure FN’s mind is not on Earth, but on some fictitious fantasy planet of its own imagination.
I don’t understand “The Final Clout Head.”
He or she keeps banging on here about a problem, but he/she won’t tell me what it is! I think it’s got something to do with the current and perhaps even the predicted future temperature of the planet.
Uranus of course.
bill
”You can probably cherry-pick your start dates, but you’d only be fooling yourself in a Monckton-ish fashion.”
I’m cherry picking Nick’s Brisbane regional which shows a decrease since 1950 (not UAHs 1979) So, unless they have been removing buildings and roads for the last 70 years from that area, it shows a slight cooling since 1950. All others show a warming over the same time period. What does that information mean to you?
“So, unless they have been removing buildings and roads for the last 70 years”
Actually they moved the site from the centrl Botanic Gardens to the airport near the sea.
They didn’t “move the site” they stopped recording at that site and started a new record at another.
Since 2016 El Nino.. COOLING quite fast.
Australia is cooler now than it wasaround 1880-1900
THERE IS NO PROBLEM !!
Except in your silly chicken-little mind.
Now…. How much warmer must it have been only 1000 years ago for trees to grow where now there are glaciers.
Still waiting for your answer.
bnice,
I am trying to provide original data that allow tests of statements like yours that Australia is now cooler than the 1890-1900 era.
I have not found evidence in these numbers to support your contention.
What numbers did you use?
I do not claim that numbers tell the whole story. Newspaper reports have to have credence when they report deaths from heatwaves, special trains to shift people cooler, numerous birds dropping dead in flight etc. But that is hard to translate to a cooling Australia. Geoff S
You can get the data from the NOAA web pages, but the data is weirdly either not suited for the purpose or inconsistently formatted.
The GHCNm monthly data sets come only in one large gzipped tar file, and contain TAVG data only.
The GHCNd daily summaries are single-station CSV files, but the formatting gets sketchy.
For example: the format is the familiar comma-separated, surrounded by quotes — but it is not consistent. Here’s a typical record including the headers. I’ve broken them out into separate lines to make viewing easier:
“STATION”,
“DATE”,
“LATITUDE”,
“LONGITUDE”,
“ELEVATION”,
“NAME”,
“PRCP”,
“PRCP_ATTRIBUTES”,
“SNWD”,
“SNWD_ATTRIBUTES”,
“TMAX”,
“TMAX_ATTRIBUTES”,
“TMIN”,
“TMIN_ATTRIBUTES”,
“TAVG”,
“TAVG_ATTRIBUTES”
“FRE00104044”,
“2021-09-03”,
“49.18”,
“0.4558”,
“67.0”,
“CAEN CARPIQUET, FR”,
” 0″,
“,,S”,
,,,, <- ??? Where is the TMAX data?
” 103″,
“,,S”,
” 178″,
“H,,S”
All of the attributes are supposed to be a “,,,” format, but they are not. The busted formatting makes it impossible to consistently extract the values. The files are full of this, too.
This was from straight download from the site, but you get the same problem if you use NOAA’s interactive tool to select data.
Who would you report this to?
The last three months have had a slightly “higher” anomaly due to El Nino and lack of rain (but not the highest anomalies)
They just happened to be in WINTER.
The winter WEATHER has been absolutely GORGEOUS 🙂
There is a non-positive right back to August 2012.
You are the one making an abject FOOL of yourself.
Repeating nonsense over and over does not make it any less nonsensical. Whatever the veracity of his methodology, Monckton does not ‘cherry-pick’ the start-date.
” So the atmosphere over Australia is cooling. Interesting….”
“Not according to UAH satellite data.”
We have only had satellite data since 1979.
Longer term, Australia was just as warm in the Early Twentieth Century as it is today. See the Tmax chart of Australia below.
Thanks, Nick,
But you are adding complexity to my simple analysis by using different observation periods with more complications like recent instrument changes. You are also sowing doubt with subjective, personal beliefs like nobody expecting much warming before 1950. There is no way that you can prove that, so it remains childish, anti-science fairy tale. There are people I know who have no idea what to expect with temperatures like these, which is a more supportable stand.
The question I ask might boil down to this: given that these foundational figures show, with hefty error, that Australian warming seems credible at about half a degree from the late 1800s to 2006, why is there alarm? That half a degree might well sit inside the noise envelope that has never been measured with competence and publicised. My next article addresses that.
Are we staring at a nothing burger?
Geoff S
Imagine that!
Adding ‘complexity’ to an analysis that stops its data nearly 20-years ago. Since your analysis period stopped in 2006, Australia has experienced 8 of its 10 warmest years in the UAH satellite data (since 1979).
2023 is looking likely to make that 9 out of 10.
An “average temperature” of the Australian continent at any time is arrant nonsense.
It literally is the same as that bloke sitting with one arse cheek on a stove top and his other arse cheek on a block of ice, and claiming that overall, he’s quite comfortable.
2023 for UAH Australia is currently ii SEVENTH place on a year-to-date basis.
Stop your petulant little tantrum !
I am always puzzled by the fact that raw, unfudged temperature series show only slight warming due to UHI, yet the “average” temperature of the AustrAlian Continent shows strong warming.
Graeme,
I do not yet have confidence that any past estimate of UHI is credible.
The present work is to either provide an estimate or to show why it is so hard, even impossible using historic data.
There is promise in some satellite based work. More on that later.
Geoff S
“I am always puzzled by the fact that raw, unfudged temperature series show only slight warming due to UHI….”
Not true
https://www.carbonbrief.org/explainer-how-data-adjustments-affect-global-temperature-records/
Again, Simon gets sucked in by one manically adjusted fabrication against another manically adjusted fabrication.
Hilarious. ! 😀
Only in Climate “Science” are “adjustments” to historical records considered legitimate. In any other discipline this would be outright fraud.
You go right on making a fool of yourself, you petulant little baby brain.
“nobody expecting much warming before 1950. There is no way that you can prove that, so it remains childish, anti-science fairy tale”
Simple. AGW says that warming will follow GHG concentration. Here, from the WUWT reference page, is the increase of CO2.
ppm CO2 had risen 20 by 1950, now about 140 ppm.
Not a realistic temperature series… One of the AGW-cult FAKES !
Many places around the world show the 1940s similar to around 2000.
A lot of the warming since around 1960 is due to urban and airport expansion.
On top of that, the fabrication routines make the whole thing totally unreliable.
Nick,
I have just shown some original graphical trend data for Australian temperatures that are different to the generic shape that you correlate with global CO2.
Who is responsible for the inference that the Australian data are compatible with the global data, when I have just shown a sample of it that it is not.
Australian data have a significant influence on global estimates for several reasons. It is not permitted to ignore this primary Australian data and much more like it using excuses such as Australia being only a few % of the global surface area, or whatever.
I have seen no proof that the Australian trends are erroneous, that they ought to be like the global trends, because, like, ” the vibes”
Nor have I seen proof that Australian trends cannot act in a pattern of their own. Have you?
Geoff S
What are the reasons, exactly?
Read the article again maybe you will find it.
It is simple really. They are part of the globe. Australia will influence the global temperature. Your question is simply ridiculous.
Because they are the large part of the actual measurement of historic temperatures in the Southern Hemisphere.
Much of the rest is made up, as Phil Jones wrote in Climategate.
But you know this, so why do you argue?
Geoff S
What did Phil Jones say “in Climategate,” specifically? What is the context? I’m asking you questions, not arguing with you.
Indeed. It isn’t true.
He isn’t going to confine himself to address YOUR presentation details itself which is why his replies are exits……
Geoff,
“Who is responsible for the inference that the Australian data are compatible with the global data, when I have just shown a sample of it that it is not.”
Well, yes, who are they? Can you quote them?
There is always regional variation. Here is a plot of a screenshot from here of a sphere mapping of trends from 1950 to 2020. Australia is warming, although not as fast as land masses in the NH
To complete the picture, here is the far side of the earth:
Again.. total fabrications.. based on.. what ???
If you are basing them on GISS, they are just meaningless pretty pictures.
Mr. Stokes: Your graph is scientific proof of what people expected before 1950??!! Why don’t you simply admit you made an opinion-based statement? You are a legend at picking out such nuggets in others. Instead, you confirm my view that there is no lie you can’t graph.
The AGW issue has been clear since Arrhenius (1896). If you put a whole lot of CO2 in the air, you’ll get warming. But you have to put it there first. At least 80% of the CO2 rise happened after 1950.
WRONG !!
Arrhenius NEVER showed that scientifically…
It was a conjecture, based on the radiative properties he has just discovered inside a glass jar.
It remains an UNPROVEN CONJECTURE.
Like I said, all the Climate Change Alarmists have to “prove” their case that CO2 needs regulation is a bogus, bastardized Hockey Stick chart. Without this distortion of reality, they would have nothing to talk about.
We would never hear from Nick again.
As per mosh’s point about time series being sub-optimal, do you have the data to hand to present similar graphs using the log of CO2 concentration as the X axis?
OC,
I could do that. Maybe for part 2.
That would be good, Nick. Thanks.
Could you do separate graphs with the CO2 axes as linear and log? A linear approximation may be reasonable for the concentration range thus far.
OC,
I’m a bit more reluctant, because I think it is the wrong thing. As said above,the connection should be between [CO2] and flux, which then goes to time rate of change of T.
Yes, surface air temperature is rather indirect, but the flux is even harder to measure than enthalpy or temperatures.
Taking one step at a time often adds a little clarity.
Every one of those sites is strongly affected by urban infill, and expansion.
They are totally unfit for use for climate purposes.
They cannot give a realistic portrayal of real Australian temperatures.
Sydney obs hill.. in the middle of several harbour bridge 3 lane highways
Melbourne, Was at the junction of several busy roads… the absolute worst site imaginable…. move to a place that now channels hot northerly winds straight to the site.
Brisbane: unsure.
Hobart .. a complete joke.. surround on 3 sides by building in a small courtyard in the densest part of Hobart.
Airport.. airport.. ..
Alice springs. Shows temperatures in 1990-2010 being no warmer than many around the 1910-1930 period.
Here is a googoo maps of the Hobart site..
Seriously !!!!!
Note, Building at bottom, angled up to left, has all glass windows facing the site.
So northern sun hits that glass, and reflects….. guess where to. 😉
The site is absolutely ideal for Stokes, Rusty Nail et al since it corrupts the remperature series in the direction they desire.
Iirc, the same afternoon that the new Melbourne site was fired up, it was recording 2C lower at the same time than the old site just a few blocks away.
Both sites are a joke.
The terrible concept of high density housing.
Mr Austin told news.com.au the large housing estates like the one at Marsden Park – with their tiny backyards, black rooves, and concreted everything – were not just unbearably hot, but had created dangerous urban heat islands that can reach temperatures up to 50C.
“Western Sydney is naturally hotter than the rest of Sydney, about two-degrees warmer than it is in the east,” Mr Austin said.
“But now we’re seeing temperatures sitting up to 10 and 12 degrees hotter.
Urban heat islands develop in areas with an abundance of hard, sealed surfaces like concrete, metal, asphalt, bricks, and lack of greenery and vegetation – like these housing estates.
The issue, Mr Austin says, is that while nature can absorb the heat from the sun, a “non-permeable surface like concrete and roofing … spits it back out”.
https://7news.com.au/news/nsw/urban-planners-dire-warning-for-western-sydney-residents–c-11485350
“Every one of those sites is strongly affected by urban infill, and expansion.”
The sites and data were chosen by Geoff, not me. I just showed what the trends really were.
No, Nick, I just showed what some of the foundational trends really are. If yours are different, you have to explain how and why and by how much. Geoff S
Geoff,
The only difference is that I brought the data more up to date, and showed the relevant period for AGW, which is post-1950, when CO2 emissions were getting large, as were the warming trends.
I was jut commenting on the never-ending situation here where someone posts some data and draws some conclusion; I post the same data with different analysis, and, the cry goes up, why did you choose that horrible data? Nobody thinks of asking that of the original poster.
WRONG. There is no scientific evidence that CO2 causes warming.
It was warming from urban expansion and agenda adjustments that caused most of the slight but highly beneficial warming.
There is absolutely ZERO way of determining any CO2 warming from the data you are using.
You are showing just what a low-level con-artist you are.
Yet you KNEW they were unsuitable and said nothing.
Why is that ?
You really are nothing of a scientist, are you.
This research clearly indicates recent changes in Southern Hemisphere surface temperature cannot be attributed to anthropogenic causes.
Jones, J., Gille, S., Goosse, H. et al. Assessing recent trends in high-latitude Southern Hemisphere surface climate. Nature Clim Change 6, 917–926 (2016). https://doi.org/10.1038/nclimate3103
Calling Mosh – Did Nick just reduce the data set here? Does that make him a skeptic? Just curious.
Nick,
The GHCN numbers are derived from the foundational data that I posted. Did you have a reason for mentioning secondary data when primary data are the subject of the article?
Geoff S
Geoff,
The GHCN numbers are exctly the same as BoM. They are just a more convenient source for me, and go to present.
All of the sites used are manifestly unfit for climate use.
No way of knowing just how much is from urban development and other non-climate sources.
Dear Nick,
There is actually no climate-related trend in temperature data for Sydney Observatory, none, zilch, zero, which is why my attached report is important.
As years have past and I have continued to collect information relating to the Observatory (including a visit when I took pictures inside the old office), the attached report is need of an update. I am scheduling this for early next year.
Since the Sydney study was completed, I also developed a range of protocols, including techniques for separating (de-confounding) the effects of the weather and effects due to site changes on the overall trajectory of naive least-squares trend.
I am sure Nick, you understand that for a trend-line to adequately describe the data generation process, data themselves must be free of non-climate impacts. Most of the time if practitioners examined least square trend residuals it would be obvious that the line of best fit poorly reflects the data.
Bomwatch protocols are physically-based, objective, replicable and applicable to temperature datasets from across Australia and Tasmania. I’m surprised no one has looked at data for other continents using a similar step-wise approach. I have gone to great lengths to describe how the protocols work and what is involved. See for example here: https://www.bomwatch.com.au/bureau-of-meteorology/are-australias-automatic-weather-stations-any-good/
I am currently using BomWatch protocols to evaluate the homogenisation of Australian temperature datasets, focusing initially on the Pilbara region of north Western Australia (see https://www.bomwatch.com.au/data-homogenisation/).
Gradually I will dismantle the ‘work’ commenced by Neville Nicholls and his proteges, including the Bureau’s current data-masseur Blair Trewin.
All the best,
Dr Bill Johnston
http://www.bomwatch.com.au
The data should not be adjusted when a station moves, the data should stop for that site at that time, and a new record at the new station started.
What is wrong with you that you can’t see that?
As so often, all I can say is take it up with the original poster (Geoff). He nominated the stations and the data.
I believe I’ve seen you express the same opinion. Do you think a station’s data should be adjusted after a move, or that station’s record ended and a new one begun?
It is not just a move. Even a new technology device in an existing screen should qualify as a stop/start record event. Replacing a screen with a smaller should qualify.
Do scientists in other fields get to go back and CHANGE measured data when new technology causes a change?
Name one that isn’t climate science.
The Urban Heat Island is exactly that: An Island
An island being = somewhere notably different from everywhere surrounding it.
The urban heat effect, which defines The Island, depends upon ‘an island’ of extreme dryness (the city) existing in an ‘ocean of wetness’ (the surrounding countryside)
It is the contrast of wet vs dry that defines the island.
Furthermore and why (elsewhere around the world) some cities have a greater UHIE than others, the size of the UHI (Delta T) is a measure of the moisture contrast – city vs rural.
e.g. Phoenix or Las Vegas would have a much larger UHIE if it they were surrounded by forest than if surrounded by desert. as they are.
For Australia, everywhere is bone-dry all the time – there is no ‘ocean of wetness’ for an urban island to exist in.
Thus and because everywhere in Australia is dry, you cannot construct a island – the cities are as equally dry as the surrounding coutryside so there can not be an urban heat island effect of measurable amount.
Maybe yes briefly after a rainfall event – because the city will drain faster than the surrounding countryside.
Australian temperatures will be rising because the ‘ocean’ of countryside was not a perfect bone-dry desert 100+ years ago
It is strenuous efforts by folks trying to grow arable crops, folks grazing sheep, folks building/operating and failing to maintain incendiary devices called ‘transmission lines’ doing their damnest to wring every last drop of moisture out of the place.
The rising temps we do see are the fruits of their labours.
Obviously you have never been here.
Peta,
My next article shows how hard it is to find a meaningful and useful difference between rural and urban temperatures. Current beliefs in UHI are based on reasonable hypotheses like concentrations of people helping make hotter spots than in remote countrysides.
But, the measurement reality is tough on beliefs.
Geoff S
It rains along the coasts.
https://mag.ncep.noaa.gov/model-guidance-model-area.php?group=Model%20Guidance&model=gfs&area=south-pac&ps=area
It also often has long periods of protracted drought … we might be entering such a period now.
Bone dry apart from the coastal fringe and it too often has long periods of protracted drought.
Mid Hunter Valley is looking pretty “parched” at the moment.
Maybe next weekend.. they say good chance of rain.
Free rain tomorrow!
b
Urban Heat Island effect is real, but is confused by/ indistinguishable from site changes, both over many years and seasonally. Very few sites in Australia are “pristine”, certainly not in the past, Even very small towns have a detectable difference in temperature from sites not far away. However, if the site has not changed (including urban development) over the years trends should be comparable- but finding such sites might be a head ache.
After my survey of sites in 2019 I do not believe the current site network is at all useful for climate research. Nearly half of those surveyed did not comply with the Bureau’s own siting guidelines.
“In Australia it is apparently quite OK to have thermometers beside houses, in bitumen carparks, in a vegetable garden surrounded by a corrugated iron fence, beside incinerators, behind 6 metre prison walls, beside piles of human excrement, in the middle of a dirt road, on the roof of a wharf shed, beside a multi-lane highway, shaded by trees, or in screens that are covered in spider webs, invaded by mud wasps, or used by cattle as a back-scratcher. The area around the screens can be dusty bare dirt, overgrown with grass and weeds, or sprayed out to bare ground.”
Even the supposedly top quality Acorn network is less than 40% compliant.
Hi Ken,
Thanks for dropping in. 🙂
Have perused you site many time.
GREAT WORK exposing the absolute GARBAGE that are many of Australia’s temperature sites.
Hi Ken,
We continue to independently find similar outcomes from detailed data analysis.
I have tried to select many “pristine” sites from hundreds of candidates, but have had to narrow down to a maximum of 45. Some of these will further drop out, but the residue might have worthwhile information content. All is good. Geoff S
Geoff, the major work by Spencer and Christy of UAH indicates 50% (52% in the study) of the ‘official’ Global Warming reported for the USA was actually due to UHI. Even going from 5,000 pop. to 10,000 shows an easily definable UHI increase. It will be interesting to to see the UHI results you obtain.
I suggest, in addition to the long term stations for the main study, that you do supplementary studies of towns that have records long enough to evaluate the UHI affect of doubling (or more) of pop. of small towns.
Gary,
Stay tuned. Almost ready with that next part. Has a large amount of measurement data to let you do your own analysis. Geoff S
There is no UHI in Sydney Observatory data, just a small increase following to the opening of the Western Distributor. The Bridge (Bradfield Highway) is the busiest road in Australia, but its T-signature maxed-out decades ago.
What seems to be NOT understood is that as hot air rises UHI is dissipated vertically – as turbulence, not horizontally. The Horizontal vector acts as the in-feed to the turbulent chimney that carries it all away, possibly to be measured up there in the air by UAH.
Cheers,
Dr Bill Johnston
http://www.bomwatch.com.au
Depends on where the temperature site is wrt to the urban area and prevailing winds.
If the site is close-by down-wind from the urban heat bulge, it will experience urban warming effects.
Dear bnice2000,
Yes/no, but it cannot be generalized that a town of 300 people has less impact on a weather station than one of 10,000 people. If you have such a hypothesis, it would be necessary to compare changes in temperature at a fixed site verses changes in the population of the town, taking account of all other factors. To my knowledge this has not been done. Placing a thermometer on the roof of a car and driving across town does not measure UHI relative to to fixed point. If you want UHI try flying over Sydney in a relatively light aircraft, say a Dash8.
Maximum temperature at Sydney airport, which is an elevated site, stepped-up when they opened another lane in the outbound (east-most) Southern Cross Drive tunnel, which goes under the taxiway/runway. That road, which services the South Coast, Southern Highlands and Canberra is said to be the second busiest urban road in Australia.
While there are many attributable site impacts (including moving the weather station to its current position, installing a 60-litre Stevenson screen and building a vertical wind-profiler array and aircraft parking bays on the western side of Southern Cross Drive) there is no urban warming (or other warming) signal in the data.
Analysing data cannot be done by tossing it against the wall using Excel, to see what sticks.
All the best,
Dr Bill Johnston
http://www.bomwatch.com.au
“but it cannot be generalized that a town of 300 people has less impact on a weather station than one of 10,000 people”
Precisely.
There is no way of telling what the exact effect of urban expansion/infilling and other local changes have had on any particular surface temperature site.
That is what makes them totally unsuitable for finding warming or cooling trends on a wider basis, certainly not globally..
Totally unsuitable of determining changes in global climate over time..
Try running your data for Nobbys Head (Newcastle). It has a long data series and hasn’t been influenced by any Urban Heat Island (based on it’s maritime location). I think you’ll find the trend is notably lower
Actually, Nobby’s may be slightly affected by the urban and industrial expansion… especially when winds with a western aspect (anywhere from NW-SW).
You certainly notice the warmth in those westerlies when you are on Nobby’s Beach
I have, and due to site changes it is actually a very difficult dataset.
Although I have not formally reported on Nobbys data, when site changes are taken into account, there is no residual trend that could be attributed to the climate.
Cheers,
Dr Bill Johnston
http://www.bomwatch.com.au
“The BOM asserted in 2022 that ‘Australia’s climate has warmed by an average of 1.47 ± 0.24 °C since national records began in 1910.'”
I note that the BoM has been making the same claim for more than two years and continues to do so (http://www.bom.gov.au/climate/enso/)
However, there seems to have been an update to this relentless claim over the past couple of months, albeit still ignoring three consecutive La Ninas that have substantially reduced the 1.47C since the assertion was first made.
The BoM claim now reads …
“Australia’s climate has warmed by an average of 1.48 ± 0.23 °C since national records began in 1910.”
It’s remarkable how Australia’s ACORN mean temperature anomaly was 0.56C in 2021 and 0.50C in 2022, yet that’s resulted in Australia’s climate warming by an additional 0.01C to 1.48C since 1910.
I’d say around 1.5C of that warming is from urban site infection.
And of course, the period from 1880-1910 was probably similar or a bit warmer than now, according to the thermometers that existed back then.
I’d say they just made it up!
Cheers,
b
Bom is parlous
local station edenhope airport( yes tarmac very few planes cropdusters air ambulance and the odd rich farmer;-)
if the barometer works the temp/rain bits go missing
in 16yrs I reckon that ive seen barometric numbers maybe 7 times?
Im under half a km at most from the setup rain totals bear no resemblance majority of the time
the last 5 days? we DID have high/low temps showing for yesterday and days prior and now most are missing this happens frequently
The Tmax chart for Australia shows Australia was just as warm in the Early Twentieth Century as it is today.
The Australian Hockey Stick chart in the article shows a different, “hotter and hotter” profile.
Which one repesents the real Austrailian temperature profile?
manipulating, altering or ignoringtorturing the foundational dataIt’s the dark rooves wots causing the global boiling-
Climate change: Dark roofing banned on Sydney’s urban fringe (smh.com.au)
Trust your Gummint and their experts and planners and not the deniers here spreading misinformation and disinformation. What more evidence do the deniers want that the doomsday clock is Tik Toking with the boiling?
Western Sydney housing estates the ‘hottest place on earth’ | news.com.au — Australia’s leading news site
Face facts doomsters and watermelons. White rooves matter-
Colours | COLORBOND® steel
Red alert with all those dark solar panels too-
Solar panel adoption stirs fear of Chinese cyber attacks (msn.com)
We know what you’re up to Commies-
Black Solar Panels Look Nice But Are Less Efficient (solarquotes.com.au)