The latest data set from The Met Office shows that the January 1851 southern hemisphere temperature anomaly was 1.800 C.
They calculated the southern hemisphere land temperature (within 0.001 degrees) based on a single thermometer in South America. In fact, it was the only thermometer south of Cuba.
We now have one-tree Keith and one-thermometer Phil.
I guess James “this is for my grandchildren” Hansen gave Phil Jones statistics lessons, and what an excellent student Phil is!
Well, that settles it then….. clearly we know the earth has warmed by Rio thermometer.
Tell me this is true or isn’t, I can’t believe it. Is this a madness of CRU or the Met Office?
Someone should teach The Met Office about the proper use of trailing zeroes.
“They calculated the southern hemisphere temperature (within 0.001 degrees) based on a single thermometer in South America.” !!!!!
This should be the final straw for any thinking man. And three decimals of accuracy to boot! Jesus, Joseph, and Mary!
They just have no shame.
Perhaps http://www.niwa.co.nz/ have a thermometer they could offer Phil; rectally, of course
Thru the magic of statistics, you can poll one blind Chinaman in a dementia ward and determine with 99.99% certainty who is going to win the US election.
CRU(and The Met Office) insist that they are scientific.
They are using a definition of “scientific” to which I was previously ignorant.
If you believe their method and I got a computer model to sell you.
Today’s featured Wiki article had me checking the date is not 1st April.
http://en.wikipedia.org/wiki/Voluntary_Human_Extinction_Movement
I still did not believe it.
http://vhemt.org/
Believe it! There is a real group that promotes this.
Science is being trashed by dead beats. Climate change has morphed into the biggest white collar crime on the planet.
Climate is science.
Climate change is bias.
Climate ‘change’ as in ‘change you can believe in’.
Silence! Henchmen at work.
I don’t want to doubt you, in fact I’d prefer to submit your blog to a news aggregator, but you don’t really document your claim here.
Can you demonstrate this new temperature estimate is based on one thermometer?
The data file is from their web site. You will need a program which can read netcdf. Good luck with that on Windows.
http://www.cru.uea.ac.uk/cru/data/temperature/CRUTEM4v.nc
There’s a bunch of tools here:
http://www.esrl.noaa.gov/psd/data/gridded/tools.html
I haven’t checked any of them, but at least one is Java-based so it should work on Windows. Thanks for reference to netCDF; I’ll possibly need it for work in the future.
The fallacy of false precision. Or is that the precision of a false phallus?
This is so blindingly, gobsmackingly pathertic, CRU should be laughed out of the court of public opinion.
That’s like reading the entire earth’s CO2 levels from just one station in Hawaii.
Actually, no, it isn’t. CO2 is a well-mixed in the atmosphere, so measuring it in one place is quite a reasonable thing to do, provided the location is chosen carefully. Even so, that isn’t what we do, because there are many stations all around the world monitoring CO2 levels, and they all agree quite closely with the value measured at Mauna Loa
Nonsense:
http://data.gosat.nies.go.jp/GosatBrowseImage/browseImage/fts_l2_swir_co2_gallery_en.html
Nice link to some maps.
woops………LOL
Look at 2011/12 and 2010/12…you can’t even compare the same month year to year
It does make sense that the Northern Hemisphere runs higher than the Southern Hemisphere.
You seem not to understand the meaning of “well mixed”. Here’s a hint: it does not mean that CO2 emissions instantly diffuse around the entire planet. The physical processes involved take time.
And over time, wherever you measure the CO2 level, you will find that it is increasing at around 2ppmv per year. This really should be utterly uncontroversial, even amongst people who dispute the effect of increasing CO2.
Trivia — The Met Office Hadley Centre for Climate Change is named in honour of George Hadley. So much intrigue surrounds the Met, I decided to do some research into the man the Centre was named after.
This research led to a paper by Anders O. Persson with The Department for Research and Development at The Swedish Meteorological and Hydrological Institute entitled Hadley’s Principle: Understanding and Misunderstanding the Trade Winds.
Some interesting quotes from Persson’s paper –
Stimulated by Ferrel and Mohn opposition to Dove’s and Hadley’s theories grew among theoretical German meteorologists in the 1870’s 57. More articles by Finger, Sprung and Thiesen followed. The debate for and against the “Dove-Hadley Principle” ended with a victory for the theoreticians58. The break with the Hadley-Dove model was not always easy. For many it took some years to realise that the Hadley-Dove model was not a “simplified” or “incomplete” version of Ferrel’s correct model, but completely wrong59.
There were three major points in the criticisms of “Hadley’s Principle” in the German
meteorological community: (1) Hadley’s explanation only works for north-south motion, although the deflection is valid for all directions; (2) The underlying conservation principle should not be one of absolute linear momentum, which only yields
Thanks Stephen for the link (which I see you already had linked and I missed it).
Sadly, you’re absolutely correct again, on Windows I have no idea how to read that.
So you may not find this useful, but well, I read you from time to time, and find your claims interesting and enlightening, but often, not documented or dumbed down enough to truly be able to make sense of them and verify them.
I’d love to submit them to various news aggregators, but I can’t, not with how you write the posts currently.
I guess they are a bit too inside baseball.
You mean Phil didn’t consult Mr Sallinger?
I’m curious about where Mr Goddard’s data comes from. He claims that CRUTEM4 shows a Southern hemisphere temperature anomaly of 1.8C in 1851, but the graph he shows us seems to indicate a best estimate of -0.4C. Even the error bars (which he seems not to have noticed when castigating the alleged precision of sparse data) do not extend as far as 1.8C.
And the CRUTEM4 Southern hemisphere series starts in 1856, not 1851, precisely because of the sparsity of data until then. Only the Northern hemisphere series starts in 1851. Mr Goddard (or any of his readers) could have learned this in the few seconds it takes to reach the Met Office website:
http://www.metoffice.gov.uk/hadobs/crutem4/data/diagnostics/hemispheric/southern/index.html
I’m curious why you don’t see the link to the data right under the map?
Oh, I saw it. But a link to a 19MB NetCDF file doesn’t really explain your treatment of the data it contains. If the published CRUTEM4 Southern hemisphere data analysis starts in 1856 (I gave you a link), why have you taken it upon yourself to extend that analysis to 1851 and then use your own analysis as a stick with which to beat Jones?
Huh? CRUTEM is the land data for HadCRUT. Are you suggesting that HadCRUT shouldn’t start until 1856?
In 1856, CRUTEM had a massive five stations in the southern hemisphere. Two in South America, One in Australia and two in New Zealand. Less than 1% coverage.
So why does your personal data reanalysis start in 1851? And where exactly is the scandal? Short of requiring Jones to go back in time to install more thermometers, what did you want him to do? And why have you completely ignored all the error bars in making the following claim:
“They calculated the southern hemisphere land temperature (within 0.001 degrees) based on a single thermometer in South America.”
That’s wrong in every single respect, isn’t it? There was not “a single thermometer”, and there is no claim of 0.001 degree precision.
In fact, for 1856 (the earliest year in the CRUTEM4 series), the best estimate anomaly is -0.384C, but with upper and lower bounds based on the combined effect of all the uncertainties of 0.180C and -0.913C.
Why are you accusing Jones of saying things that he has not said?
Are you unfamiliar with CRUs published data? They show temperature anomalies within 0.001 degrees.
http://www.cru.uea.ac.uk/cru/data/temperature/hadcrut3vgl.txt
Even so, are you seriously trying to claim that they can calculate a hemispheric anomaly within +/0.50 degrees – with only one land based temperature reading? Have you thought through how ludicrous that idea is?
For some reason there is no reply option to your latest reply, so I’ll have to reply to the previous one again.
“Are you unfamiliar with CRUs published data? They show temperature anomalies within 0.001 degrees.”
No, I’m very familiar with that. It makes perfect sense, but does not falsely imply precision, as you seem to think. Are you perhaps unfamiliar with the meaning of error bounds?
“Even so, are you seriously trying to claim that they can calculate a hemispheric anomaly within +/0.15 degrees – with only one land based temperature reading? Have you thought through how ludicrous that idea is?”
They didn’t. Only your own reanalysis of the raw data has done that. And I have no idea where your 0.15 degrees comes from. For 1856 (the earliest year in the Southern hemisphere series) the uncertainty is +0.564C or – 0.529C.
If you think that is unreasonable, based on the sparsity of available data, then feel free to show it. But just saying it doesn’t make it so.
Temperatures this month in Alaska are 5-10C below normal, and 7C above normal in Illinois. I would put the error bar for one site as being +/-7 C.
The whole HadCRUT trend is 0.7C. Their error is greater than their trend, making their time series worthless.
njp – Are you unfamiliar with reporting an appropriate number of significant digits? You are only allowed to report to a precision as precise as your least accurate data. Error bounds are a completely different issue.
Having had taken statistics classes in the past and regularly read medical studies, I cringe at what “climate scientist” have done with their data on multiple levels.
“njp – Are you unfamiliar with reporting an appropriate number of significant digits? You are only allowed to report to a precision as precise as your least accurate data. Error bounds are a completely different issue.”
I suspect you are confused about what these numbers really mean. The number of digits arises not from the precision of individual measurements (which is of course much lower), but by what happens when you average a number of them together – which is what you need to do in order to derive a global or hemispheric temperature series. The reason why hemispheric averages are given to a precision of three decimal places is to enable seasonal values to be calculated to ±0.01°C. The extra precision implies no greater accuracy than two decimal places.
“Having had taken statistics classes in the past and regularly read medical studies, I cringe at what “climate scientist” have done with their data on multiple levels.”
I suggest that you have blindly accepted other people’s claims of what they have done, rather than what they have actually done.
….so you can only get these numbers through math
So, you’re saying that since the hemispheric averages are (at least) three orders of magnitude too precise we can safely calculate that the seasonal values are also (at least) three orders of magnitude too precise. So the seasonal variation is at best known to ±10°K, which is precisely the point that everyone who understands the numbers is making.
Glad I could clear that up for you.
Er, no. You seem not to understand how mathematics works, let alone how thermometers work. If I asked you to calculate the mean of three values: 1, 3 and 4, what would your answer be, and why?
And how do you arrive at your frankly bizarre claim that temperatures are uncertain to ±10°K? Can we see some working?
3
The methods are all in the text above. If you don’t understand what an order of magnitude is, perhaps you can look it up.
I can see why you people always find yourselves at odds with mainstream science and mathematics. That’s quite an unusual method of calculating an arithmetic mean you’ve invented!
In the 7th grade I learned what the term “significant digits” means and how it is used. Did you actually pass the 7th grade, or was your education sub-standard?
And yet you don’t seem to have learned how to calculate an arithmetic mean…
It is really complicated. If you pick a station in Alaska this month, your number is 10C too low. If you pick a station in Illinois you get 10K too high.
One station in a hemisphere is completely meaningless. Five stations in a hemisphere is completely meaningless. Incredible that anyone would argue otherwise.
I don’t understand why you think thermometers in Alaska have a systematic bias of +10C. Or why you think thermometers in Illinois have a systematic bias of -10C. Or why you flip between Kelvin and Celsius for no apparent reason.
And I don’t understand why you choose to focus on a period right at the start of the CRUTEM4 temperature series, whilst ignoring the entire method used to analyse the raw data, and replacing it with your own, which is to apply no analysis at all.
And I don’t understand why you still haven’t corrected your original claim of a 1.8C anomaly for the Southern hemisphere in 1851, based on a single thermometer, which does not exist in CRUTEM4, but only in your own “analysis” of the data.
Don’t you think you owe Jones an apology?
Yeah, why would Alaska be 10°K (or more) lower than the hemispheric mean? Man, that’s a really tough question there. You might want to lie down and hang your head off the end of the bed so the blood pools up there while you’re thinking about it, cos it’s really gonna be a taxing think, I’m sure.
Why does that have to do with my question?
Agree w/ Stark. The answer for the “mean” is 3. The “median” also happens to be 3. One significant figure. That is what we learned in engineering and statistics classes in undergrad. Points would be docked for reporting too many sig digits.
That said, you can carry over many sig figures throughout your calculations, but when reporting a “final answer” you can only report to a precision of your LEAST accurate data.
It doesn’t work like that. Let’s get back to thermometers, since this is what this is all about. Let us suppose that the only error you have in your thermometer reading is a random uncertainty in the reading. There are of course other sources of error, but we don’t need to consider them for this argument.
Suppose this uncertainty is 0.2C. If you want to calculate a monthly mean temperature, this will be based on at least 2 readings per day, giving 60 or so readings in total. The uncertainty in the monthly mean temperature from this single thermometer is then:
0.2 / sqrt(60) = 0.03C
So now we have a greater precision in the mean temperature from many readings than we had in the individual reading.
The whole reason why generating temperature series is a complicated business, which Mr Goddard chooses to completely ignore when referring back to very old raw data, is that you have to make allowances for all sorts of errors – hence the carefully calculated error bars.
Of course, when the hapless scientists do this, people like Mr Goddard accuse them of fraud.
Can I ask you if you agree with him that CRUTEM4 claims a temperature anomaly of 1.800C in 1851, as he claims? If true, it would somewhat undermine the case for global warming, which Mr Goddard seems not to have noticed in his enthusiasm to attack Jones!
Significant figures, in addition to other aspects of error theory/propagation, are your friend, unless you’re a CAGW proponent. Then all pretense of adherence to science, math, statistics, and logic are like unicorn rainbows blowing in the wind.
Even 0.1° +/- is a chimera. As we observe in the civil engineering/land surveying field, too many folks estimate here, estimate there, and carry their calculations to 12 decimal places and proclaim they’ve arrived at the *answer*.
What a crock.
Of course by using more elaborate statistics means that you can do away with most of the actual weather stations as the averaging process will even out the anomalies.
Believe that and I’ve got a computer model to sell you.
Measurements to this degree of exactness must no doubt rely on the most complex and regular tree rings ever to be grown in the history of the planet…excepting, of course if they disagree with the anticipated outcome.
Silly sodomites.
You seem a bit confused about this temperature series, which is based on the instrumental record, and not on temperature proxies.
And you seem a bit confused about sarcasm .
Well, as per the Faith Of The Hippies, money can generate invention (the more the faster): it therefore stands to reason that The One Thermometer was wrapped in IOU’s totaling billions of adjusted dollars, and therefore was auto-magically correct.
This is how the Egyptians did their stuff, in case any of you were wondering.
Silly humans.