Image Credit: Compo et al., 2013
From the Huffington Post:
The thermometers got it right. The Earth is warming, another study is reporting.
Climate scientists recognize that changes in weather observation stations’ immediate surroundings — such as neighboring trees being replaced by heat-absorbing concrete — can eventually throw data from such stations into question.
But now, a new study directed by a researcher at the National Oceanic and Atmospheric Administration that recreates climate history without the use of land-based observation systems shows the same thing that thermometers have been reporting.
“This shows that global warming over land is real,” said Gilbert Compo, a scientist at NOAA’s Cooperative Institute for Research in Environmental Sciences at the University of Colorado.
“It is not an artifact of the observing system,” said Compo, lead author of the study, which he presented to the European Geophysical Union on Tuesday in Vienna. “It is happening.”
Compo and his colleagues used an alternate method to review the planet’s temperature history from 1871 through 2010.
They deployed what is called 20th Century Reanalysis (20CR), a physically based, state-of-the-art data assimilation system using barometric pressure records, ocean surface temperatures and other factors independent of land-based readings that can be skewed by changes in their surroundings.
Compo’s team came to a conclusion that supports land-based instruments’ reporting that, since 1952, the Earth has shown a 1.18 degree Celsius increase in air temperature over land.
Compo, in an email, stated that the actual number the 20CR analysis showed for warming since 1952 was 0.78 degrees Celsius, which he termed “statistically indistinguishable” from 1.18 degrees Celsius.
Here’s the associated study; Independent Confirmation of Global Land Warming without the Use of Station Temperatures and abstract.
I’ll leave it to WUWT’s readers to do their own distinguishing…
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
“Compo, in an email, stated that the actual number the 20CR analysis showed for warming since 1952 was 0.78 degrees Celsius, which he termed “statistically indistinguishable” from 1.18 degrees Celsius.”
That would also mean that there result is “statistically indistinguishable” from 0.38 degrees Celsius and so would support equally the idea that there has been significantly less warming that shown in the thermometer record.
The Compo et al reconstruction flattens after the 1997/98 El Nino. The divergence in recent years was more evident when we used typical GISS and UKMO base years. Compo et al shows a significantly lower trend since 1976 and an exaggerated spike in 1943 associated with the multiyear El Niño then. They also show a later start to the rise during the early warming period.
We discussed it here:
http://bobtisdale.wordpress.com/2013/04/08/a-preliminary-look-at-compo-et-al-2013/
With the WUWT cross post here:
http://wattsupwiththat.com/2013/04/08/a-preliminary-look-at-compo-et-al-2013/
I don’t think anyone has disputed that the temperatures have warmed since the 1970’s. I don’t think anyone has disputed that the temperatures have warmed during the 20th century since the end of the Little Ice Age at the end of the 19th century.
What is in dispute is if the warming is of any unusual duration or magnitude outside of previous periods of natural warming. Pay careful attention to this statement:
Translation: Campo and his colleagues used an alternate method to review the planet’s temperature history from the end of the Little Ice Age through the present time.
I would expect to see warming. The study does not prove anything except we have been in recover from the Little Ice Age. Why don’t they take their study back to 1000AD so we can put the current rise into context with the cooling going into the LIA?
Bob Tisdale did a preliminary review of this paper here at WUWT a few days back …
http://wattsupwiththat.com/2013/04/08/a-preliminary-look-at-compo-et-al-2013/
As did Skeptical Science:
http://www.skepticalscience.com/surface-temp-record-accuracy-confirmed-compo.html
I liked these comments at the end of the SkS page:
Abstract: ” Any artifacts in the observed decadal and centennial variations associated with these issues could have important consequences for scientific understanding and climate policy. … This independent dataset reproduces both annual variations and centennial trends in the temperature datasets, demonstrating the robustness of previous conclusions regarding global warming.”
“Robust” is one of the nice political spin words that is conveniently vague enough that any claim made using it is non falsifiable. It has no place in science , thought it is frequently used.
As soon a someone says a result is “robust” I get suspicous, it usually means it isn’t but they’d like you to think it is.
This paper does not demonstrate the “robustness” of the IPCC’s land warming any more that it denomstrates the robustness of doubts raised about the accuracy of the station record in Fall et al and Watt et al.
I’d assume he’d find a change in his salary statistically indistinguishable if he’d only get paid two thirds of what he’d been told to get as well?
Man, there’s those really precise figures again. Gotta be good science if you tack on enough siginificant digits.
OTOH, based on the coverage & precision of the proxies & instrumental records, I’m sure that 1.18, 0.78, 0.0, -4.2, 8, & 0.01 are all indistiguishable.
http://bobtisdale.files.wordpress.com/2013/04/figure-1-compofig1.png
The plot from the paper that Bob reproduced in his article also shows that this study going in the same sense as my contention that the Met Office bias “corrections” are removing a significant amount of the climatic variation from the early (pre WWII) part of the record.
judithcurry.com/2012/03/15/on-the-adjustments-to-the-hadsst3-data-set-2
“The principal effect of these adjustments is to selectively remove the majority of the long term variation from the earlier 2/3 of the data record and to disrupt circa 10-11y patterns clearly visible in the data. These changes are fundamentally altering the character of the original data.”
Since that article was only looking at SST , it seems that _despite_ much of the early variation having already been removed from the SST data they still found greater variation in land temps when using it as a base for this study.
Maybe some ought to reassess the SST record with reference to barmometric data as well.
University of Colorado?
Is this the same unit that has redefined “sea level” from being “the level of the sea measured against the land” to ” the level of the sea measured against what we say WOULD have been the land if certain things WE claim have occurred had not occurred”?
When the time comes for judicial accountability, these will be amongst the first and most severely judged.
Darn, they must have done one heck of a manipulation job to match GISS and HadCrud temperature manipulations.
“Compo” is also a hilarious character from the long running British sitcom, Last of the Summer Wine!
http://uk.search.yahoo.com/search?p=compo+&ei=UTF-8&fr=chr-greentree_ff&type=937811&ilc=12
“Global warming is real”. What a childish meme. In response to the impression that it is somehow “not real”? Of course it’s real. Right now, it’s NOT HAPPENING. But it’s still “real”. But the purpose of this study, is an effort to make it “real”, I guess. Ow, it hurts to read.
“NOAA meteorologist Jeffrey Whitaker, a co-author of the study, explained why some land-based historical readings have come not to be trusted.”
If some data in a database cannot be trusted and nothing is done about it, doesn’t it make the entire database useless?
Does NOAA acknowledge UHI now?
Question from a non-scientist: who decides where the 0 on the left-hand scale goes? Most folks will assume that 0 is a baseline to something, but without knowing what it isn’t helpful to me. Is it the normal BEFORE something happened and temperatures plumetted and we are now above the historical norm? And how many standard deviations are we talking about here in terms of temperature variation around this 0 line?
Crosspatch: Temperatures don’t warm. The air warms or the water warms and temperatures get higher or lower. 🙂
The quality of this reanalysis will depend on the quality of SST data. Which is known to be much better since 2003 because of ARGO. So why aren’t they worried that they have a “divergence problem” for the period with the best data?
Comp et al ” Nevertheless, the time variations of TL 2m in the 20CR are very similar to
tthose previously reported in the station-based datasets [Brohan et al., 2006; Hansen et al., 2010;
rVose et al., 2012; Jones et al., 2012], both over the 1901 to 2010 period and the more rapidly-
A warming 1952 to 2010 period (Figure 1 and Table S2). We have focused on the TL2m anomalies”
More spin masquerading a science.
Of course the 1952 to 2010 period is “warming more rapidly” than the longer 1901 to 2010 period which includes a 30y period of global cooling. So what?
This is yet more of the insidious process of injecting an idea biasing the reader without actually making a factually incorrect statement.
The implied message here is ” recent period is warming more rapidly … as is expected due to CAGW”. What is actually says is that if you take two warming periods plus a cooling period in a data set that shows a strong periodic variation and compare it to just one period that is ONLY a rising segment , the latter will be steeper.
SO what ?
This is neither surprising nor informative, In fact presented like without saying why they consider fitting a linear trend to an periodically varying dataset has ANY validity at all is downright misleading.
Again, they are not stupid. They are trained scientists. If they are smart enough to be doing Kalman filters etc they damn well know that this trend fitting jive has no validity.and is downright misleading.
This is another case that would appear to fall within the NSF definition of scientific fraud and malpractice.
If the actual figure [0.78 deg] is indistinguishable from their estimate [1.18 deg], why not simply use the actual figure? I am tired of mathematical and semantic tricks; give us data!
The Huffington Post is as unbiased and scientifically accurate as the BBC. Land temperatures will be more variable than sea surface temperatures and not just because of the difference in planetary area. Land has far more variables in heat retention, thermal conductivity, rock type, plant cover, height above sea level, amongst many others that will affect surface air temperatures. Thermally the oceans are all composed of the same thing with known thermal properties.
I am carrying out an experiment now using two digital thermometers, with 0.1C accuracy, to measure temperature in my garden. They are 30yards apart and can differ in readings by up to 4C. whilst my experiment may not be the most stringent it does demonstrate the difficulty of measuring a simple thing like temperature.
Climate is not a static thing and we had some (unexceptional) warming during the 20th century. The wavy diverged upwards in a wavt sort of way, and no one has the foggiest (unless you are bigot) which way it will wave over the next century. (except it is at a high point so down seems a good guess).
Send for Nora Batty!! She’d soon sort Compo out.
(sorry – couldn’t resist)
Compo et al: ” the 1901 to 2010 period and the more rapidly warming 1952 to 2010 period”
Here is a plot showing similar “more rapid warming” , a well documented feature of the cosine function. /sarc
http://climategrog.wordpress.com/?attachment_id=209
The actual is estimated to be 0.78; the instrument record shows 1.18.
0.4/0.78 = an exaggeration of 51 %. “Statistically indistinguishable”, he must be rounding to the nearest integer.
DaveA says:
The actual is estimated to be 0.78; the instrument record shows 1.18.
0.4/0.78 = an exaggeration of 51 %. “Statistically indistinguishable”, he must be rounding to the nearest integer.
No, Dave , what they mean is that the uncertainty of their method is larger than the difference between their result of 0.78 and the IPCCs 1.18 degrees. Except that they don’t present it like that.
The corollary of that is that it is also “statistically indistinguishable” from a much lesser rate of change like 0.38 degrees.
In truth this study does not tell us anything new , it just hides behind the vagueness of the word “robust” to pretend to say something when in fact that is meaningless.
The study itself is of interest since it uses different data. However the uncertainty does not justify the conclusion that it in any way confirms IPCC is more correct than a much lesser rate of change.
Did they just say that roughly 50% of the temp record is UHI?