It’s worth going back every so often to see how projections made back in the day are shaping up. As we get to the end of another year, we can update all of the graphs of annual means with another single datapoint. Statistically this isn’t hugely important, but people seem interested, so why not?
For example, here is an update of the graph showing the annual mean anomalies from the IPCC AR4 models plotted against the surface temperature records from the HadCRUT3v and GISTEMP products (it really doesn’t matter which). Everything has been baselined to 1980-1999 (as in the 2007 IPCC report) and the envelope in grey encloses 95% of the model runs. The 2009 number is the Jan-Nov average.
As you can see, now that we have come out of the recent La Niña-induced slump, temperatures are back in the middle of the model estimates. If the current El Niño event continues into the spring, we can expect 2010 to be warmer still. But note, as always, that short term (15 years or less) trends are not usefully predictable as a function of the forcings. It’s worth pointing out as well, that the AR4 model simulations are an ‘ensemble of opportunity’ and vary substantially among themselves with the forcings imposed, the magnitude of the internal variability and of course, the sensitivity. Thus while they do span a large range of possible situations, the average of these simulations is not ‘truth’.
There is a claim doing the rounds that ‘no model’ can explain the recent variations in global mean temperature (George Will made the claim last month for instance). Of course, taken absolutely literally this must be true. No climate model simulation can match the exact timing of the internal variability in the climate years later. But something more is being implied, specifically, that no model produced any realisation of the internal variability that gave short term trends similar to what we’ve seen. And that is simply not true.
We can break it down a little more clearly. The trend in the annual mean HadCRUT3v data from 1998-2009 (assuming the year-to-date is a good estimate of the eventual value) is 0.06+/-0.14 ºC/dec (note this is positive!). If you want a negative (albeit non-significant) trend, then you could pick 2002-2009 in the GISTEMP record which is -0.04+/-0.23 ºC/dec. The range of trends in the model simulations for these two time periods are [-0.08,0.51] and [-0.14, 0.55], and in each case there are multiple model runs that have a lower trend than observed (5 simulations in both cases). Thus ‘a model’ did show a trend consistent with the current ‘pause’. However, that these models showed it, is just coincidence and one shouldn’t assume that these models are better than the others. Had the real world ‘pause’ happened at another time, different models would have had the closest match.
Another figure worth updating is the comparison of the ocean heat content (OHC) changes in the models compared to the latest data from NODC. Unfortunately, I don’t have the post-2003 model output handy, but the comparison between the 3-monthly data (to the end of Sep) and annual data versus the model output is still useful.
Update (May 2012): The graph has been corrected for a scaling error in the model output. Unfortunately, I don’t have a copy of the observational data exactly as it was at the time the original figure was made, and so the corrected version uses only the annual data from a slightly earlier point. The original figure is still available here.
(Note, that I’m not quite sure how this comparison should be baselined. The models are simply the difference from the control, while the observations are ‘as is’ from NOAA). I have linearly extended the ensemble mean model values for the post 2003 period (using a regression from 1993-2002) to get a rough sense of where those runs could have gone.
And finally, let’s revisit the oldest GCM projection of all, Hansen et al (1988). The Scenario B in that paper is running a little high compared with the actual forcings growth (by about 10%), and the old GISS model had a climate sensitivity that was a little higher (4.2ºC for a doubling of CO2) than the current best estimate (~3ºC).
The trends are probably most useful to think about, and for the period 1984 to 2009 (the 1984 date chosen because that is when these projections started), scenario B has a trend of 0.26+/-0.05 ºC/dec (95% uncertainties, no correction for auto-correlation). For the GISTEMP and HadCRUT3 data (assuming that the 2009 estimate is ok), the trends are 0.19+/-0.05 ºC/dec (note that the GISTEMP met-station index has 0.21+/-0.06 ºC/dec). Corrections for auto-correlation would make the uncertainties larger, but as it stands, the difference between the trends is just about significant.
Thus, it seems that the Hansen et al ‘B’ projection is likely running a little warm compared to the real world, but assuming (a little recklessly) that the 26 yr trend scales linearly with the sensitivity and the forcing, we could use this mismatch to estimate a sensitivity for the real world. That would give us 4.2/(0.26*0.9) * 0.19=~ 3.4 ºC. Of course, the error bars are quite large (I estimate about +/-1ºC due to uncertainty in the true underlying trends and the true forcings), but it’s interesting to note that the best estimate sensitivity deduced from this projection, is very close to what we think in any case. For reference, the trends in the AR4 models for the same period have a range 0.21+/-0.16 ºC/dec (95%). Note too, that the Hansen et al projection had very clear skill compared to a null hypothesis of no further warming.
The sharp-eyed among you might notice a couple of differences between the variance in the AR4 models in the first graph, and the Hansen et al model in the last. This is a real feature. The model used in the mid-1980s had a very simple representation of the ocean – it simply allowed the temperatures in the mixed layer to change based on the changing the fluxes at the surface. It did not contain any dynamic ocean variability – no El Niño events, no Atlantic multidecadal variability etc. and thus the variance from year to year was less than one would expect. Models today have dynamic ocean components and more ocean variability of various sorts, and I think that is clearly closer to reality than the 1980s vintage models, but the large variation in simulated variability still implies that there is some way to go.
So to conclude, despite the fact these are relatively crude metrics against which to judge the models, and there is a substantial degree of unforced variability, the matches to observations are still pretty good, and we are getting to the point where a better winnowing of models dependent on their skill may soon be possible. But more on that in the New Year.
Completely: #689
“Really? May I remind you:
“Think of it as being more like a discus with a flat top.””
LOL. Okay, Completely, now think of it as being a lopsided discus. It still has areas of thin ice and thick ice. Peel away some thickness of ice and some of the thin ice disappers and becomes a smaller ice area.
RE Tilo Reber
What is the noise in that data set? Significance is a function of the trend and the error, so measurements that show a strong trend and are damped (er, show less variability), like ocean heat content, may be significant at shorter time periods.
Tilo: “LOL. Okay, Completely, now think of it as being a lopsided discus.”
Nope, not that either.
A discus isn’t 15% wood and 85% air and still counted “discus”.
“Peel away some thickness of ice and some of the thin ice disappers and becomes a smaller ice area.”
But unless it peels away to less than 15% of the original ice area, you can still get “100%” ice coverage and therefore “read” more ice than there is.
See also:
http://www.google.com/search?q=site%3Atamino.wordpress.com+tsonis
Tilo, you’re making this stuff up without any source.
Kids, you can easily check this kind of claim. Just paste his words into Google, using them as search terms, and read the results. You’ll find out what’s actually known.
Here, for example:
http://www.arctic.noaa.gov/reportcard/seaice.html
Read the whole page, it’s not that long, to understand what’s going on.
Excerpt follows:
Recent estimates of Arctic Ocean sea ice thickness from satellite altimetry show a remarkable overall thinning of ~0.6 m in ice thickness between 2004 and 2008 (Figure. S4a). In contrast, the average thickness of the thinner first-year ice in mid-winter (~2 m), did not exhibit a downward trend. ….
The total multiyear ice volume in the winter experienced a net loss of more than 40% in the four years since 2005 while the first year ice cover gained volume due to increased overall coverage of the Arctic Ocean. The declines in total volume and average thickness (black line in Figure S4a) are explained almost entirely by thinning and loss of multiyear sea ice due to melting and ice export. These changes have resulted in seasonal ice becoming the dominant Arctic sea ice type, both in terms of area coverage and of volume….
PS, the above link tells you the observations, which contradict the PR talking points being repeated above:
the thin (mixed and first year) ice is near the pole and the thicker (multiyear) ice is nearer the edge, where it gets pushed during the melt season and stacks up.
See the illustration:
http://www.arctic.noaa.gov/reportcard/figures/s3.jpg
Figure S3. Arctic sea ice distribution in March of 2007, 2008, and 2009. Multiyear ice is in white, mixed ice aqua, first-year ice teal, and ice with melting surface red. Dark blue is for open water and brown for land….
#695 Tilo Reber
I did read your #676 and you were responded to in #685
Your logic does not make any sense. In the context of this argument sea ice is like a block of cheese. In fact it is easy to imagine a reduction of ice thickness and even an expansion of ice extent. I simply don’t understand why your brain seems incapable of understanding this?
El Nino is at full strength?
http://www.esr.org/pdo_index.html
How do you figure?
#701 Tilo Reber
Okay, now think about reality. Or can you prove that Atlas is swimming around in the Arctic and maybe with Poseidon is holding up the Arctic ice sheet and tipping it?
Wow, Tilo has created a new Pseudo-science: Greek mythology mixed with fanciful storytelling…
It’s called ‘Mythcience’
Just wait till the universities start writing new textbooks. Tilo is going to be rich beyond the dreams of Avarice.
Tilo: “Peel away some thickness of ice and some of the thin ice disappers and becomes a smaller ice area.”
Up in 660 I provided you with two links to observations disproving this too-simple idea of yours and here you are continuing to repeat it. Do you have a reading comprehension problem? Do periodic arthritic seizures prevent you from clicking? As a child were you bitten by Ayn Rand? Does telling porkies fire off the pleasure center in your brain? You tell me.
#697 Tilo Reber
Sorry, I sent you PDO to show the PDO phase in contrast to 1997/8
http://www.cpc.noaa.gov/products/analysis_monitoring/enso_advisory/ensodisc.html
El Nino “at full strength” is a relative term. Do you mean in comparison to the 97/98 event, or just for it’s current phase? There is a difference in “strength” you know.
Yes, as I said above (and Tilo more or less laughed at), Tilo is ignoring currents and winds.
The ice isn’t sitting there like a discus. Or frisbee. Or a loaf of bread.
It’s being pushed around, and when thick chunks get floated out through the ummm Fram(?) Straight into warmer waters they just *melt*.
BTW, look how wiggly this years refreeze ice extent line is compared to the 2007-2008 one.
Thinner ice moving apart, closer together, above the 15% coverage threshold, then below, etc? Just a guess but it’s interesting to see how different this year’s refreeze is (oh, yeah, it’s refreezing at a slower rate, too, tied with 2007-2008 at the moment despite the summer minimum being significantly higher).
“recovery” my a** …
not sure why some of you persist in encouraging Tilo’s foolishness, but I think his persistance may have something to do with his view of RC
#712 flxible
I appreciate his foolishness. Just take a look at RC”s site stats. They are getting 6 to 10k visitors per day lately. The more Tilo posts here, the more we can help those thousands of visitors understand just how silly his points are.
C’mon Tilo… post some more… I and others here are ready to correct your silliness for those who actually care about learning about ‘Real Climate’.
Deech56 #702
What is the noise in that data set? Significance is a function of the trend and the error, so measurements that show a strong trend and are damped (er, show less variability), like ocean heat content, may be significant at shorter time periods.
Okay, I buy that in concept. But what if we remove variability for the surface temperature record. For example, by using ENSO corrected data. Then wouldn’t that also allow you to have a shorter period that was significant.
Completely:
“A discus isn’t 15% wood and 85% air and still counted “discus”.”
Okay, I admit, that went right over my head. Help me out.
“But unless it peels away to less than 15% of the original ice area, you can still get “100%” ice coverage and therefore “read” more ice than there is.”
At the perimeter much of the ice that is counted as part of the ice area is already close the 15% range. And much of it is thin ice. Take away a meter of thickness, like you are doing in the thickest part, and pieces of that ice disappear completely. That then removes the area from the map.
Thanks for the link, flxible. In particular, I liked this from Tilo: “The attempt by the scientists at the Hadley Climate Research Unit to illicitly control the debate about global warming by using their influence with publishers and the IPCC will come as no surprise to the thousands of skeptics that have had their legitimate questions censored at Real Climate.” (For those without a scorecard, the Hadley Center and the Climate Research Unit are entirely distinct institutions, aside from both being located in Blighty.) As Ronald Reagan once said, facts are stupid things.
I liked the headline on the next item, too: “More Fraudulent Censorhip at Real Climate.” Earth to Tilo: In the English as she is spoken, “fraudulent” plus “censorship” does not equal “fraudulent censorship.” One of the things I’ve noticed about scientists is that they tend to express themselves with, if anything, an excess of precision, usually to the benefit of clarity (big exception for Mike Hulme in this latter regard). You’re kind of the opposite. I don’t expect you to ever change the essential dishonesty of your views, but if you want to get better at expressing them I suggest you make a study of how RP Jr. goes about it.
Completely: #685
“But it;’s more like a slush puppy poured out in to a large bowl.
Sea ice extends to where there is about 5x more “not ice” than ice.”
Kind of hard to measure ice thickness on a slush puppy, isn’t it? But okay, I’ll take the slush puppy model if it makes you happy. As long as you don’t claim that it’s like a block of cheese.
In a slush puppy some of the ice is still thick and some is still thin. Some chuncks are large and some are small – if you like. Again, this means that removing a certain thickness of ice will mean that some of the surface ice will disappear. At the perimeter where there is already just 15% coverage, that ice will drop below 15% and the area will no longer be counted.
“… and who knows, in time, maybe he’ll learn to cite …”
[Response: nice ref – gavin]
John: #710
“Do you mean in comparison to the 97/98 event, or just for it’s current phase?”
I’m talking about the current phase. Here:
http://www.cpc.noaa.gov/products/analysis_monitoring/lanina/enso_evolution-status-fcsts-web.pdf
See page 5 of 37. The charts on the right look like we may be maxed out in all but Nino 1+2. The next month should tell us for sure.
Re #710: John, just so we’re clear, there’s no particular reason to think that the PDO drives much of anything. Tilo and friends don’t understand what it is, so for them it’s become kind of a climate hope chest to drop all their fears into (sorry, Tilo’s rhetorical excesses are inspiring me). Plus of course “Pacific Decadal Oscillation” sounds important.
Anyway, the PDO involves looking at North Pacific SSTs and seeing if they’re very slightly warmer or cooler than the global signal. We can do the same thing with any region of the ocean and absent anything else going on it wouldn’t mean bupkes, but in the case of the North Pacific it was noticed that certain fish species that people like to eat are rather sensitive to those slight changes. In other words, the PDO is of economic but not (much) climatological importance. I say not much as opposed to no importance since the PDO does indeed describe real differences from the overall temperature trend and so can be linked to climate in its region, but is very much not a significant driver. This recent review is worth a read if you want more detail.
dhogaza:
“Yes, as I said above (and Tilo more or less laughed at), Tilo is ignoring currents and winds.”
No laughing dhogaza. I believe that the ice moves around to some extent. But I still need an explanation for why that means the thin ice wouldn’t melt completely and as a result why thinning wouldn’t also reduce area. Take a look at a chart like this:
http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh
See the green ice. It is around the edges. See the darker red ice, it’s generally closer to the center. As you watch the change through the year the ice on the outside will grow last and disappear first. The stuff around the edges will melt before the stuff in the center. Again, it’s not perfect, but it does follow that pattern. The currents can also change this to an extent. But the pattern is still similar.
To chip in my own two cents on programming languages:
When I am prototyping, I usually use Python + numpy since it’s quick and dirty, is reasonably fast, and the strong and dynamic typing makes it hard for me to write code that will result in a segmentation fault so I don’t have to be as paranoid while coding.
When I write non-prototype scientific code where I really care about performance, I first write the main number-crunching kernels in Fortran because it is both *fast* and *expressive* for brute-force number crunching on multi-dimensional arrays. Then I write the code for the algorithm itself using a higher-level language that has stronger typing and better support for expressing complex control flow, and have it call the Fortran kernels in order to do the heavy lifting.
My personal favorite algorithm language of late is Haskell because not only is it very expressive, but its type system gives you a lot of power to ensure that particular kinds of objects can only be used in very specific ways, which lets you prevent yourself from making whole classes of mistakes. However, in the past I have also used Python as the “glue” language to tie Fortran kernels together, and I would be open to using C++ for this role as well. (Incidentally, I know that many people disagree with me on this, but C++ templates rock since they let you be able to write algorithms in a very general way without losing performance for special fast cases; there have been times when this has *really* helped me a lot, since it prevented me from having to write two versions of the same code just to have both a flexible and a restricted but fast version.)
So in short, my philosophy is that one should be open minded about the possibility of using multiple languages in a project so that each part of it can be written in the language that is optimized for solving that particular problem. There is a strong downside to this approach, though, which is that it assumes that other people are willing and able to learn more than one language in order to work on your project.
John P. Reismann wrote in 707:
John, that is a link to a page on the Pacific Decadal Oscillation. It is somewhat similar to the El Nino Southern Oscillation but not the same.
You can see the difference in patterns here:
The Pacific Decadal Oscillation
http://jisao.washington.edu/pdo/
Additionally, El Nino is quasi-periodic (~2-7 years) whereas PDO is often regarded as closer to pink noise.
As for ENSO, it is strongly positive. Full strength? Hard to say, but certainly not comparable to the super El Nino of 1998.
Please see:
ENSO cycle as indicated by 1st EOF
of surface current and SST anomalies
http://www.esr.org/enso_index.html
ENSO Update
17 December 2009
http://iri.columbia.edu/climate/ENSO/currentinfo/QuickLook.html
Regardless, global temperature typically lags ENSO 4-6 months. If ENSO continues through the spring and the sun continues to become more active…
Disappearing sunspots may signal end to global warming
Examiner.com – Dec 16, 2009
http://www.examiner.com/x-32936-Seminole-County-Environmental-News-Examiner~y2009m12d16-Disappearing-sunspots-may-signal-end-to-global-warming
Oops! I mean:
2009’s Sleepy Sun Finally Woke Up in December
By Alexis Madrigal
December 31, 2009
http://www.wired.com/wiredscience/2009/12/the-year-in-sunspot
… we might have an interesting year.
PS
John, sorry, I see you already made the correction. However, hopefully my above comment will contain some useful or at least interesting information for some.
Dictionary definition of recovery:
re·cov·er·y (r-kv-r)
n. pl. re·cov·er·ies
…
2. A return to a normal condition.
The only possible way that you can characterize 2009 Arctic ice conditions as a “recovery” is “2009 has recovered to the long-term downward trend”, as though it is perfectly normal for Arctic ice to be declining year-on-year.
I’m guessing by your name, Tilo, that English may not be your first language, so you can be forgiven for not understanding what “recovery” means. But now that it has been pointed out to you, I trust you won’t continue to say that Arctice sea ice has “recovered”, will you?
Arctic ice isn’t what it used to be, or so it would seem, what it appears to be…
Please see:
… and:
Barber et al. (2009), Perennial pack ice in the southern Beaufort Sea was not as it appeared in the summer of 2009, Geophys. Res. Lett., 36, L24501
#719 Tilo Reber
Thanks. Let’s see how it plays out. But the effects seem to be less. The 97/8 event was like a good El Nino surfing a wave on the north shore.
#720 Steve Bloom
Right, I don’t know if it’s a driver, but I still kinda look at it as a wave of temperature that can feed something else. I don’t know enough about it to understand what the mechanisms are but it looks to me like El Nino is the surfer and PDO is the wave and the bigger the wave, the more rad the ride.
#723 Timothy Chase
My bad. I tried to correct in #710
#724
An awesome post Timothy! Nice to know my screw ups can inspire something good once in a while :)
Tilo’s arguments confused the heck out of me, since while exhibiting complete ignorance of ice dynamics, they also didn’t seem to help his larger anti-warming argument. So I dug back and found his original weird claim, and it became clear where he went wrong.
Tilo, forget about the ice thickness, please. It seems to confuse you. Instead, consider ice age for a moment. Single year ice forms every year, and most of it will melt again. Some of it will survive and become thicker second year ice. Once, there was a very high proportion of very thick, very old ice in the Arctic. Now, all the ice is young – mostly very young. Young means thin.
Now, due to thermodynamics that are beyond my power to explain, it does not follow that a warm Arctic means thinner than normal single year ice. For ice to form, below freezing is all it takes. In fact, counter-intuitively, warmer temperatures may cause faster and thicker ice formation. In the Arctic, “warmer” just means “not so impossibly mind-numbingly cold”.
The take-home point is that massive melting and low sea ice extent minima year after year reduces the average age of the ice, and consequently reduces the average thickness of the ice.
When it comes to long-term measurements of actual multi-year ice thickness, data is very, very sparse. There have been a few studies, but I wouldn’t place too much weight on them. My point is that we don’t need these studies to make the claim that ice is getting thinner (although the studies agree with this conclusion). We don’t need to worry about definitions of extent, melting mechanisms, or any of the weird similes that have been proposed. Ice, like any natural phenomenon, can be studied in incredible detail, and while it might be fun to hammer out all the minutia, we risk losing sight of the main point: it’s melting.
Claiming that ice extent is “recovering” is as daft as claiming that global temperature is “cooling”. Why do people expect global records every single year?
Re #727: John, that’s entirely reversed as to the PDO. I also think the wave analogy isn’t very good, although it’s a bit less so if applied to El Nino (which would in turn make the PDO and other smaller-scale climate phenomena the riders). The review I linked to describes attempts to establish whether the PDO significantly drives or even significantly affects the adjacent components of the climate system, the short answer being that it mostly doesn’t. That’s in sharp contrast to ENSO, which affects things at great distances and directly influences the global temperature trend. ENSO is as powerful as it is because it’s driven by heat accumulation and loss from the West Pacific warm pool, which is one hell of a driver.
It’s the same error deniers make when reading any graph! The first derivative doesn’t tell us anything about whether a time-series has peaked. It has no predictive powers at all! Following this screwy logic, we would have concluded that it had “peaked” in September-October, too. November would have been a huge shock! It’s the “cooling since 1998” fallacy all over again.
Has it peaked? I have no idea. The forecasts say El Niño will continue throughout spring, but I understand prediction beyond that is very uncertain.
694, Tilo Reber:
“Jason, it’s almost as though you haven’t paid attention to anything that we have talked about.”
Oh, the irony!
“Take Michaels objection to starting with an El Nino year and ending with a La Nina year. Why do you think that we have been talking about ENSO corrected data all of this time?”
And you’ve missed the point — unless you think ENSO is the only thing that affects variability?
The forcings are small relative to the variability — IOW, the signal-to-noise ratio is very poor. Therefore, a lot of data is required to tease out the signal. Removing the effect of ENSO does improve the signal-to-noise ratio and thereby can allow trends to be detected with shorter time periods, but why do you assume something magical about starting in 1998?
Given your refusal to accept direct observational evidence when it comes to Arctic sea ice, I also have to wonder — why are you so willing to accept the mathematics involved in removing that ENSO effect? If somebody was to bother doing it for you (since you have made absolutely no effort to do it yourself) and it showed a positive trend on your favourite data, would you still treat it as gospel or would you throw it into the same bin that you throw everything else that doesn’t agree with your preconceptions?
“Take Michaels point about not making the assertion that global warming has stopped. I’ve never made that statement. My statement is that there hasn’t been any since 1998. Those are two completely different things.”
If you say so; your statement is still obviously wrong.
“Same point. That’s why we want to use ENSO corrected data.”
No, you want to use ENSO corrected data because you believe it will show a flat trend on your favourite data set. (You know, the data set from “the climate scientists involved in the CRU e-mail scandal”, to use your words.) I actually doubt it will, but knock yourself out — just do it already!
“The skeptics statement is that the climate has been essentially flat since 1998. I think that ENSO corrected HadCrut3 and ENSO corrected UAH and ENSO corrected RSS will all show this.”
Why all this thinking? Why not doing? Just eyeballing the curves I have a really had time believing that UAH or RSS will show what you think they will.
“If you want to use the outlier, then be my guest. But even your outlier has less warming than predicted by the IPCC.”
Are you trying for some record in the number of obviously erroneous claims per sentence? You did read the article that these comments are in response to, did you not?
Are your posts an example of those that you say “are allowed because they carry no weight on the actual debate”? Hmm… You might be on to something there. Gavin is obviously censoring all of your “more difficult” posts and just letting the vacuous ones through.
“Now, regarding the element of significance, I don’t accept Tamino’s arguments. Tamino censors everyone with significant disagreements. I don’t consider any of his posts as being resolved issues since he doesn’t give the opposition a chance to speak.”
Ah, yes, I guess I should dismiss the theory of evolution as well if talkorigins.org doesn’t given the Creationists a chance to speak. That’s obviously the best way to determine the accuracy of the science.
“Besides, what happened to your demand for only considering peer reviewed papers as authoritative when it comes to quoting your favorite web site.”
Where did I demand that?
“I also find Tamino’s approach as unconvincing because it is not adequately related to physical processes.”
Just how do you think Tamino created the ENSO-corrected data that you’re so desparate for?
What you seem to be saying is that if Tamino or someone else did what you are clearly unwilling to do yourself and produced an ENSO-corrected HadCrut data set, you would dismiss it immediately if it didn’t show what you wanted it to show on the basis that the approach is “unconvincing because it is not adequately related to physical processes”. But if it did show what you’re hoping for (i.e. a flat trend from the cherry-picked start date of 1998) then it would suddenly be gospel.
Don’t you see the problem with this?
The only reason you are cherry-picking a start date of 1998 and a temperature series from “the climate scientists involved in the CRU e-mail scandal” — you know, the ones who refused to release all their source code and data, and who demonstrably omit areas shown to be warming faster than average — and then use an approach that is “unconvincing because it is not adequately related to physical processes” to filter the effects of ENSO from that data is because you think it will support your claim that “the climate has been essentially flat since 1998”.
If it doesn’t, you have so many built-in excuses for ignoring the result that I have to wonder why you think anybody would bother doing it for you?
Tell me again, without looking at any data — what is your logical reason for favouring a temperature series that omits the Arctic and many other areas with stronger warming than average, produced by people who have refused to release their source code or data? What is your logical reason for thinking that ENSO is the only source of variability in the weather that allows you to arbitrarily choose the start date once it is removed without any regard for signal:noise ratios?
“Apparently, Tamino has now come up with a number that says 15 years is significant.”
I think you read too much into it — what he’s saying is that given that data, less than 15 years is not significant. It’s setting a lower bound on that particular set of data.
JasonB:
“Since the poles aren’t the only places that HadCrut has no data for, and those “missing” places just happen to also be warming more than average, it’s not surprising that there’s a difference in trend.”
“The satellites are not missing those other areas, and their trend is slightly lower, even, than HadCrut3.”
Yes, but the satellites aren’t measuring the same thing — they’re reconstructing the temperature of a signficant body of air rather than just the temperature of the air near the ground, and longer-term, the difference in trend between them is not that great — the 2000s were 0.1932 degrees hotter than the 1990s according to GISS, 0.1551 degrees hotter according to RSS, and 0.1632 degrees hotter according to UAH. Doesn’t seem like a huge disconnect to me. Not only that, but adding in the past decade of temperatures to the UAH data actually increases the trend:
http://www.woodfortrees.org/plot/uah/plot/uah/trend/plot/uah/to:1999/trend
Yes, that’s right — the trend for UAH is higher if the last decade is included; how does that square with the notion that there hasn’t been any warming since 1998? More importantly, the satellite records respond much more strongly to ENSO than the surface temperatures do, and I’ve cherry-picked the dates above to include 1998 in the shorter trend and the recent La Nina-affected years in the longer trend — why you expect removing the effect of ENSO would support your argument is beyond me.
#729 Steve Bloom
Thanks. I’m always trying to find good analogies. I’m going to have to go deeper on the PDO ENSO connections.
Tilo might want to read this:
“If you’re looking for some button-down traditionalist who relies on so-called induction, conventional logic, and verification to arrive at what the scientific community calls ‘proof,’ then I’m afraid you’ve got the wrong guy,” said the intrepid 44-year-old rebel, who last month unveiled a revolutionary new model of atomic structure that contradicted 300 years of precedent. “But if you want your results fast and with some flair, then come with me and I’ll prove that the boiling point of water is actually 547 degrees Fahrenheit.
What my hopelessly pedantic colleagues fail to realize is that their scientific method is just that—their method,” said Hapner, whose self-published 2004 thesis argued that matter exists in four states: solid, liquid, gas, and powder. “After all, would a chemist who closely observes a phenomenon, formulates a hypothesis, predicts a likely outcome, and then tests the hypothesis be capable of proving that photons, far from being subatomic particles, are actually the size of a child’s fist?”
It’s true that I’ve been condemned and ridiculed by the world’s most prominent chemists, as well as by a good number of amateur hobbyists,” Hapner said as he rubbed a balloon on his head to demonstrate a basic principle of hydrodynamics. “But then, wasn’t Einstein ridiculed when he unveiled his theory of relativity, or Copernicus when he posited that the Earth revolved around the sun? True, I have since proved them both wrong, but at least they took risks.”
http://www.theonion.com/content/node/49180
I had written above in 723:
… but I wanted to check this as it had been a while since I had read this — and also wanted to prive a reference.
The figure appears to be 6 months:
However, there is the suggestion that a rise in temperature actually drives ENSO with a lag time of about 5 months:
… but this is refering to a smaller “fluctuation”:
…Where it is the release of the heat to and spread of the heat over the ocean surface which further heats the troposphere:
http://arctic.atmos.uiuc.edu/cryosphere/multiyear.ice.quikscat.mov
2007
“… Multiyear sea ice is older and generally thicker ice – sea ice that has survived at least one melt season (shown in brighter white).”
Not thicker at the middle.
Tilo Reber says:
5 January 2010 at 8:18 PM
dhogaza:
“Yes, as I said above (and Tilo more or less laughed at), Tilo is ignoring currents and winds.”
No laughing dhogaza. I believe that the ice moves around to some extent. But I still need an explanation for why that means the thin ice wouldn’t melt completely and as a result why thinning wouldn’t also reduce area. Take a look at a chart like this:
http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh
See the green ice. It is around the edges. See the darker red ice, it’s generally closer to the center. As you watch the change through the year the ice on the outside will grow last and disappear first. The stuff around the edges will melt before the stuff in the center. Again, it’s not perfect, but it does follow that pattern. The currents can also change this to an extent. But the pattern is still similar.
But why show us 30yr old data, here’s the current data: http://arctic.atmos.uiuc.edu/cryosphere/NEWIMAGES/arctic.seaice.color.000.png
Here’s the moving ice: http://i302.photobucket.com/albums/nn107/Sprintstar400/Drift.png
Steve Bloom wrote in 720:
Actually the Pacific Decadal Oscillation is fairly important in the North Pacific. But during the warm phase of the Pacific Decadal Oscillation the North Pacific is predominantly cold. We call it the “warm phase” of the Pacific Decadal Oscillation because there is a warm sliver of water along the US Pacific Northwest — but what’s in a name?
Please see:
Is Pacific Decadal Oscillation the Smoking Gun?
Saturday, 3 May, 2008
http://www.skepticalscience.com/Is-Pacific-Decadal-Oscillation-the-Smoking-Gun.html
So “skeptics” get fooled twice when they assume that the oscillation must be important since it has the words “Pacific” and “Decadal” in its name — and then they get fooled again when they try to argue that it is the “warm phase” of the oscillation that is driving global warming.
*
Steve Bloom wrote in 729:
As for ENSO driving PDO, you might also want to check:
Matthew Newman et al. (1 Dec 2003) ENSO-Forced Variability of the Pacific Decadal Oscillation, Journal of Climate, Vol. 16, No. 23 (here)
Just curious… would the importance of ENSO contrasted to PDO in climate variability be partly due to water vapor partial pressure being an exponential function of temperature? A difference of one degree Celsius in the tropics will mean a great deal more water vapor than a difference of one degree in the cold North Pacific — although it might be a little hard to see this living in the Pacific Northwest, warmed by the Japanese current.
TR: Take Michaels point about not making the assertion that global warming has stopped. I’ve never made that statement. My statement is that there hasn’t been any since 1998.
BPL: Your statement is wrong. Not a different point of view, not a clever new take on things, not a brilliant contrarian idea. Just flat-out dunce-cap wrong. As has been pointed out to you over and over and over and over and over again on many climate blogs over the past several years.
Stop it with the Big Lie technique, okay?
1870 13.37 University of Illinois Sea Ice Dataset
1871 13.44 million square km
1872 13.40
1873 13.56
1874 13.50
1875 13.46
1876 13.44
1877 13.57
1878 13.47
1879 13.44
1880 13.52
1881 13.79
1882 13.73
1883 13.53
1884 13.60
1885 13.47
1886 13.39
1887 13.25
1888 13.54
1889 13.48
1890 13.50
1891 13.62
1892 13.59
1893 13.48
1894 13.45
1895 13.58
1896 13.47
1897 13.35
1898 13.25
1899 13.42
1900 13.46
1901 13.39
1902 13.47
1903 13.24
1904 13.20
1905 13.21
1906 13.34
1907 13.23
1908 13.30
1909 13.35
1910 13.53
1911 13.62
1912 13.60
1913 13.62
1914 13.47
1915 13.39
1916 13.61
1917 13.82
1918 13.73
1919 13.49
1920 13.37
1921 13.42
1922 13.34
1923 13.20
1924 13.37
1925 13.12
1926 13.43
1927 13.49
1928 13.47
1929 13.49
1930 13.20
1931 13.17
1932 13.36
1933 13.12
1934 13.79
1935 13.67
1936 13.40
1937 13.51
1938 13.24
1939 13.45
1940 12.99
1941 13.60
1942 13.61
1943 13.56
1944 13.53
1945 13.48
1946 13.79
1947 13.60
1948 13.76
1949 13.94
1950 13.78
1951 13.85
1952 13.90
1953 13.30
1954 13.32
1955 13.33
1956 13.47
1957 13.20
1958 13.10
1959 13.46
1960 12.91
1961 13.08
1962 13.16
1963 13.50
1964 13.35
1965 13.39
1966 13.29
1967 13.03
1968 13.28
1969 13.51
1970 13.28
1971 13.17
1972 13.39
1973 12.93
1974 12.88
1975 12.95
1976 13.15
1977 13.00
1978 13.05
1979 12.45
1980 12.54
1981 12.42
1982 12.80
1983 12.61
1984 12.28
1985 12.44
1986 12.45
1987 12.65
1988 12.56
1989 12.45
1990 12.11
1991 12.20
1992 12.45
1993 12.26
1994 12.37
1995 11.89
1996 12.07
1997 12.07
1998 12.41
1999 12.40
2000 12.14
2001 12.34
2002 12.13
2003 12.14
2004 11.90
2005 11.65
2006 11.49
2007 10.33
Going up or going down, Tilo? Or staying the same?
Gavin,
I have taken a little time this morning to refresh myself on the ocean heat content data that you show. Obviously, there are still some unresolved issues remaining with the correction of the XBT warm bias. Consider the Dominguez et al (2008) reconstruction. Near the end of their series in 2001, they showed a heat content of approximately 9X10^22 Joules. This compares to the Levitus (2009) curve which shows only 5.8X10^22 Joules.
[Response: These are anomalies to possibly different baselines. They are not commensurate. – gavin]
Now, if I splice the Argo data since 2005 onto the end of the Dominguez reconstruction, and compare to the GISS model E ensemble mean (extrapolation) from your posted diagram, it is quite evident that the rate of heat uptake in the GISS-ER ensemble mean far exceeds the rate of heat uptake inferred from the observed ocean heat content trend time series since around 1995. If the Dominguez data set is accurate, then the net radiative imbalance implied by GISS-ER over the time period since 1995 is too high. There is also something I don’t understand which is going on between what you show as the GISS-ER ensemble mean, and the GISS-ER runs shown in the Dominguez paper. Although the radiative imbalance looks about the same over the time period since 1995, the heat content in your latest post is shifted downward relative to the data posted in Dominguez, and yours better matches the Levitus data set. Can you describe what the reason for the difference between your latest post and GISS-ER data shown in the Dominguez paper? I suspect that the importance is not the absolute values of OHC, but rather the rate of change over time (TOA radiative imbalance).
Now, if we look at the larger AR-4 spread shown in Dominguez (2008), things look even more curious, when we splice onto the tail of Dominguez’s plot the recent OHC data since 2005 (as shown in Levitus, 2009). In this plot, it becomes quite obvious that over the last 15 year period, the ocean is accumulating heat far below the rate implied by the multi-model ensemble mean. (Incidentally, I estimate that over the last 15 years, the rate of heat uptake by GISS-ER is about the same as the larger AR-4 multi-model ensemble mean.)
Although the match with the Levitus data that you show looks good now, with the net radiative imbalance implied by extrapolating the GISS-ER into the future, a very large delta will quickly develop between the model ensemble mean and observations, unless there are some very large upward step changes in the observed ocean heat content in the next several years. Time will tell.
Steve Bloom:
“Do you have a reading comprehension problem?”
No, but you seem to have a topic comprehension problem. I never claimed that there were not assertions in the literature that ice thinning was happening; so why bother me with links to information that I’m already aware of. My argument is that the thinning claims are difficult to believe without a corresponding shrinkage of sea ice area. Now if you can provide me with a link that can explain the physical process of having the ice thin without the area shrinking then give it to me. Also tell me that is what it is, because I have no interest in reading a lot of irrelevant links.
Timothy Chase:
“Regardless, global temperature typically lags ENSO 4-6 months. If ENSO continues through the spring and the sun continues to become more active…”
Gavin:
“If the current El Niño event continues into the spring, we can expect 2010 to be warmer still.”
I always find it interesting that people who are warning us of temperature rise at the same time seem to be praying for that temperature rise.
[Response: Don’t be stupid. My statement is purely factual (and correct) and my desire not to have significant climate change far outweighs my desire to be proven correct in forecasting it. – gavin]
Anyway Timothy, the lag time is closer to 3-4 months. At least that is what I found from looking at the data. Comparing this table:
http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ensoyears.shtml
with the temperature record. Hansen states 3 months. Tamino is using 4 for his ENSO correction.
If we are in fact peaked out on this El Nino, then between the wind down and the lag we will probably get noticable effects through mid summer. I don’t know if that will be enough to make 2010 special for the AGW people. We will have to see.
Didactylos: #728
“Instead, consider ice age for a moment.”
Why would I consider ice age. If we take one cubic meter of ice that is at a certain temperature and is one year old and if we take another cubic meter of ice that is at the same temperature but is 2 years old, then are you saying that one will take more heat to melt than another. If that is not what you are saying, then I don’t care what the age is. If your only point is that older ice is thicker, then I still don’t care how old it is because the only thing that is still relevant is the thickness.
“Now, due to thermodynamics that are beyond my power to explain, it does not follow that a warm Arctic means thinner than normal single year ice.”
If you are saying that ice formation is independent of temperature, then why doesn’t it extend further south. Especially if all of the Arctic is “so mind numingly cold”. Too bad it’s beyond your power to explain. Because without an explanation I’m not going to believe it.
Didactylos:
“Claiming that ice extent is “recovering” is as daft as claiming that global temperature is “cooling”.”
Let’s be clear on the claim here. I’m claiming that there was a slight recovery in the sea ice area since 2007. If you want to argue that point, then we can. Let’s not pretend that the argument is about something that I didn’t say. It’s very easy to beat your own strawmen. I’m not making claims beyond 2009. What I am saying is that the temperature has flattened. I don’t know if it will continue to be flat, but IF it does, then what I am saying is that what has happened in the last two years may be an indicator of a stabilization starting in the sea ice. And I have said that we will need more time to be able to tell. Charts that show sea ice decline up to 2007 are pointless since I am not questioning them. The only question that I have in that regard is the shortness of the satellite record. I don’t think that we can say from that record what normal variability is.
Barton:
Going up or going down, Tilo? Or staying the same?
That’s a lot of wasted space, Barton; epecially considering that I only said that the ice has recovered slightly since 2007 and since you stop at 2007. See my response to Didactylos to see what I am claiming. I know that everyone here is extremely excited about ice are shrikage up to 2007. But unfortunately you cannot transfer your excitement to the debate, since I am not doubting it.
Gavin,
could you please attend to 541? Thanks.
Here’s a copy for convenience:
What would the output of the GCMs look like if there were no CO2 increase? Any links to a graph? thanks!
[Response: Assuming you mean no forcings of any kind, then the ensemble mean would be flat, but you’d still see excursions of the same magnitude as the grey bands above. – gavin]
No, I mean just without an increase in CO2, all others stay the same. Thanks. (Sorry for the delayed follow-up, forgot I asked…)
[Response: For the period shown in the graphs? You’d see a slight decline over time in the mean. See figure 9.5b in IPCC. – gavin]
JasonB:
“And you’ve missed the point — unless you think ENSO is the only thing that affects variability?”
No, you’ve missed the point. The debate around the slope of the trend since 2008 is mostly centered around ENSO. The 98 El Nino and the 2008 La Nina. So let’s get that removed. There hasn’t been much volcanic activity, so we don’t need to worry about that. Then we will have a trend line that we can talk to. If that trend line is still relatively flat, then we need an explanation about what other elements of variation are suppressing the supposed signal. If it is relatively flat and we cannot identify the natural elements of variation that are responsible, then the IPCC climate sensitivity number is in big trouble.
JasonB:
“but why do you assume something magical about starting in 1998?”
I’m not sure why you have trouble understanding that. I want to know, what is the period of time during which we have not followed IPCC predictions. That is something that you don’t want to know apparently. I could use dates before then, but I’m not arguing the fact that the temperature was rising before then – so why should I. I could use dates after then, but that would not answer the question that I want to ask.
“No, you want to use ENSO corrected data because you believe it will show a flat trend on your favourite data set.”
No, having a favorite data set is your problem and Tamino’s problem. I’m willing to accept one of three ENSO corrected data sets.
“Just how do you think Tamino created the ENSO-corrected data that you’re so desparate for?”
We were talking about his claim for significant time periods, not his ENSO charts. I believe that his ENSO chart for GISS is probably about right. If you read the rest of what I said that should have been obvious.
“Doesn’t seem like a huge disconnect to me.”
It’s not relevant to the discussion. I was talking about the divergence since 1998. My charts back that up. I was not talking about comparisons between the 90s and the 00s.
JasonB:
“What you seem to be saying is that if Tamino or someone else did what you are clearly unwilling to do yourself and produced an ENSO-corrected HadCrut data set, you would dismiss it immediately if it didn’t show what you wanted it to show”
I’m really not interested in your speculation about my motives. If you go back and look you will see that I asked Gavin to update his ENSO corrected HadCrut3 and failing that to give me the algorithm so that I could do it myself. And I offered to share the results, regardless of what they were.
Re #737: Thanks for the additional PDO info, Timothy. This post from Atmoz (found via your Skeptical Science link) is also informative. The upshot:
“(T)he mode of variability known as the PDO has the same spatial and temporal characteristics as the mean global surface temperature anomaly. PDO doesn’t cause global warming, the PDO is global warming. (Insert all the caveats of PCA; statistical relationship not causal, linear, etc.) One reason that the time series of the PDO may look so noisy compared to the time series of the mean surface temperature anomaly is simply because it has less area, and therefore less data.”
This is non-expert handwaving, but I think the answer to your question is that ENSO and the PDO are qualitiatively very different. El Ninos push up global temperatures in the short term by taking a bunch of hot West Pacific warm pool water and distributing it out across the ocean surface where it is relatively quickly lost to space, thus the very real impact on gloabl climate. The PDO does nothing of the sort.
Re #741: As you would know if you actually paid attention to the nice observations, Tilo, older (and thus thicker) ice accumulates in a gyre, the contents of which have been preferentially flushed out through the Fram Strait in recent years.
Phil:
“But why show us 30yr old data, here’s the current data:”
Just lazy I guess. It’s the first thing you get when you go to the page. But thanks for the update.
Tilo, the satellite data are particularly susceptible to ENSO, so it is unlikely that the MEI would adequately compensate for the effect. I would say that in a period where El Nino is important, it is probably important NOT TO USE UAH or RSS.