Thursday
Jan312013
by 
Bishop Hill
Bishop Hill Japanese cool
Jan 31, 2013 Climate: Surface NASA's Climate 365 website has a post declaring the marvellous agreement between the different temperature records. Here it is:
;)
But, as Tom Nelson notes, that purple line - the Japanese record - sure seems to diverge from the others. What could it mean?
Update on Jan 31, 2013 by
Bishop Hill
Bishop Hill
Anthony reckons he has pinned down the discrepancy.
Reader Comments (33)
Ooo, scary graph!
...hang on, the vertical scale is half a degree C
And the variations in what they call "the earth's temperature" are shown in much smaller increments.
These lines on the graph are certainly the output from something, but they aren't temperature measurements.
Is "Climate 365" really something officially sponsored by NASA ? It's hard to believe. Visiting that website I not only don't see the temperature chart shown in the Bishop's "Japanese Cool" article but I find a good deal of pathetic stuff (e.g. the "How you can save the world" section, which is not the only one which appear to have been written by someone who can neither spell nor write English.)
Funnily I have just posted elsewhere that the apparent need for there to be no deviation between the data sets is a cause for concern, well it is to me.
I real life, if all my usual suppliers presented the same "facts" I would immediately be off in search of confirmation from pastures anew until either confidence was gained or a different set of circumstances was found.
Competition between the data sets is needed, I think the benefits would be significant. It used to happen, Phil Jones was known to champion his CRU data, especially over GISS, critical of how they handled certain areas of the globe. This has now ceased to be the case and HadCRUT4 now shows a closer alignment.
The fact that the “big three” are always in agreement is a greater cause for concern than a source of solace.
Know nothing about the Japanese data set, will have to take a look.
They should always draw on lines showing the gradients of 2 degrees C per century and 6 degrees C per century so we can see just how amazingly accurate the predictions are looking. Or not.
Here is the explanation ... GHCN runs hot, CLIMAT does not.
"The land part of the combined data for the period before 2000 consists of GHCN (Global Historical Climatology Network) information provided by NCDC (the U.S.A.'s National Climatic Data Center), while that for the period after 2001 consists of CLIMAT messages archived at JMA. The oceanic part of the combined data consists of JMA's own long-term sea surface temperature analysis data, known as COBE-SST (see the articles in TCC News No.1 and this report)."
http://ds.data.jma.go.jp/tcc/tcc/products/gwp/temp/explanation.html
Being skeptical of the output of these graphs is beyond silly people. Everytime I hear or read the argument that temperatures of the entire globe can't be measured in centesimals of Kelvins I facepalm. Such ideas fly in the face of basic statistics. I think even Lubos Motl has touched on that point and explained very well, so I recommend you all to search his article about it and leave this silliness behind.
Dear God, how pathetic is that?
NASA is reduced to this? It is a very sad state of affairs.
That chart looks misleading to me, compared with say this one using HadCRUT calculations for a global mean tempature: http://www.climate4you.com/images/HadCRUT3%20GlobalMonthlyTempSince1850%20WithSatellitePeriod.gif
Or this one using HadCRUT4 at Wood for Trees: http://www.woodfortrees.org/plot/hadcrut4gl
In each you will see a modest overlap of the monthly temperature ranges between the current decades and that of the 1940s. The plot you show above makes that look very unlikely. They are clearly not showing monthly means, but over some longer time period. Perhaps a year, perhaps a running mean of some kind. Whatever, it may well have been chosen for effect. The 'reality' (such an artifice as global mean temperature does deserve quotes here) is that the monthly temperatures show some overlap. Not so good for PR that.
Speaking of NASA, I visited the Goddard Space Flight Visitor's Center when I was in DC last week.
PATHETIC. I guess I shouldn't have expected too much with "free" admission, though. ;)
Andrew
Whatever the agreement, it still looks like Pink Noise to me.
John Shade (3:38 PM) -
The plot is of annual average temperatures, not monthly.
Purple on black....Dark Red on Black.
Axes that are dark grey on black.....
Message: White on black.
In other words: "Listen to us, don't look at the data"
I don't know about the rest of you but I agree that the period from 1910 to 1940 looks very similar to the period from 1970 to 2000, and yet only one of these periods is allegedly related to AGW/CC/EW.
The graph would look better with error bands on it, something on the order of +/- 0.2 C.
It is not the Earth's temperature change that "scientists disagree about". It is the cause of the temperature change that they disagree about !
Oh them Japanese - they don't cook fish and they don't cook data.
No amount of statistics can get a meaningful answer from insufficient data. The historical temperature records are largely derived from a relative handful of weather stations daily max / min records which themselves are not enough information to say much about the daily average temperatures. They are then adjusted in an attempt to compensate for station siting, missing records etc etc in a fairly arbitrary manner, then gridded to cover the vast areas of land where there is no data. All this done by people whose knowledge of statistics has been criticised by many.
Sadly this is rife in acedemia; do a bad experiment, get a small number of dubious results and perform all sorts of statistical trickery on them to get the "right" answer.
Unfortunately in this case the general public is continually being sold the idea that "scientists" have been measuring "the average temperature of the earth" for centuries and that it is currently rising in an unprecedented manner. Part of this is graphs like the one shown, looks scary, actually shows a minor fluctuation on the basis of very little actual data.
"Speaking of NASA, I visited the Goddard Space Flight Visitor's Center when I was in DC last week."
You should have visited the Air & Space Museum (part of the Smithsonium complex on the Mall). NASA in it's prime and at its best!
All charts of this nature used for purposes of promoing te IPCC "consensus" should always scale relative to the assumed cliamte sensitivity.
The Y axis should extend to include the 2-4 degrees so that the extent of historical variation can be observed in proper context to the assumed anthropogenic impact expected over the future.
I thought the main area of disagreement with regard to Global Warming was not whether the globe had warmed or not, most believe it has, but rather the causes, and the relative amount of warming attributed to each cause, of said warming. ignoratio elenchi me thinks.
Roger Longstaff,
Damn. I was right there, too.
Andrew
I thank NASA for this for it is the first time I have been able to clearly see the much vaunted 'anthropogenic fingerprint' of GW :(
"I have been able to clearly see the much vaunted 'anthropogenic fingerprint'"
ssat,
I think that's a smudge on your computer sceen from when you were pointing at some lolcats. ZING
Andrew
"I have been able to clearly see the much vaunted 'anthropogenic fingerprint' "
ssat,
I think that's just a smudge from when you pointed at some lolcats. ;)
Andrew
Just looking at the graph, since about 1945 there has been around .1 degree of warming in the 71 years since then.
Can someone remind me what all the fuss is about again please?
Mailman
Mailman (7:46 PM) -
I make it around 0.5 K from the previous peak (1940-44, using a 5-year average to smooth out the year-to-year variation a bit). But I share your qualitative assessment.
One may wish that 'Climate 365' could not really be from NASA, but indeed it is part of their 'news' team output:
http://wattsupwiththat.com/2013/01/31/japans-cool-hand-luke-moment-for-surface-temperature/#comment-1213338
UCAR's synposis of the Japanese SST data is that their adjustments are "relatively primitive" as compared to the Hadley Center.
When was it decided that periodic variations in temperature would be labelled as "anomalous"?
Good guys or bad guys, there is no longer any excuse for posting a graph without providing a link to a simple file containing the data used by the person who created the graph. NASA should, and does, know better.
Now, looking at this graph, I could also be forgiven for thinking that someone at NASA is not as proud as they should be of data coming from satellites launched by NASA.
You may recall that Nerlich published a totally absurd article on CLimategate - see passim comments at CA http://climateaudit.org/2011/04/23/comments-on-mother-jones/ - in which her research was unequal to the challenge of determining which climate blogs were relevant in the story. SHe omitted the major climate blogs and instead studied a few very minor blogs that played no role in the Climategate story.
Steve, you want the next post down. I demand you retract this egregious comment, wrong and irrelevant and way beyond your skill set as a mining engineer. Only climate psychologists are able to unferstand the actions of skilled published climate psychologists. (Do I need to /sarc?)
I just took at quick look at the JMA COBE sea surface temperature data and there's nothing exceptional about it--which is good:
http://bobtisdale.wordpress.com/2013/02/01/a-quick-look-at-the-jma-cobi-kobi-sea-surface-temperature-anomaly-data/
DMC yesterday - you are right, the most serious objection that many AGW sceptics have about cAGW concerns the extent to which last century's warming is due to CO2, and hence how much more warming we're likely to get for doubling of CO2.
That being said, sceptics also have some concerns about the reliability of the estimates we have of the amount of warming that happened in the last century. I think that few sceptics believe that there was NO warming in the 20th century - but you'll find plenty of sceptics who think that there has probably been a bit less than is portrayed in the average historical temperatures shown at the top of this post. This would be due to factors such as an increase in the amount of urban heat island effects over the century, and the inappropriate use of adjustments to historical sea-surface and land temperatures. There's a good post summarizing this kind of discussion at Climateaudit here.