I recently covered a press release from Dr. Ben Santer where it was claimed that:
In order to separate human-caused global warming from the “noise” of purely natural climate fluctuations, temperature records must be at least 17 years long, according to climate scientists.
Bob Tisdale decided to run the numbers on Ar4 models:
17-Year And 30-Year Trends In Sea Surface Temperature Anomalies: The Differences Between Observed And IPCC AR4 Climate Models
By Bob Tisdale
We’ve illustrated and discussed in a number of recent posts how poorly the hindcasts and projections of the coupled climate models used in the Intergovernmental Panel on Climate Change’s 4th Assessment Report (IPCC AR4) compared to instrument-based observations. And this post is yet another way to illustrate that fact. We’ll plot the 17-year and 30-year trends in global and hemispheric Sea Surface Temperature anomalies from January 1900 to August 2011 (the updates of HADISST data used in this post by the Hadley Centre can lag by a few months) and compare them to the model mean of the Hindcasts and Projections of the coupled climate models used in the IPCC AR4. As one would expect, the model mean show little to no multidecadal variability, which is commonly known. Refer to the June 4, 2007 post at Nature’s Climate Feedback: Predictions of climate, written by Kevin Trenberth. But there is evidence that the recent flattening of Global Sea Surface Temperature anomalies and the resulting divergence of them from model projections is a result of multidecadal variations in Sea Surface Temperatures.
WHY 17-YEAR AND 30-YEAR TRENDS?
A recent paper by Santer et al (2011) Separating Signal and Noise in Atmospheric Temperature Change: The Importance of Timescale, state at the conclusion of their abstract that, “Our results show that temperature records of at least 17 years in length are required for identifying human effects on global-mean tropospheric temperature.” Sea surface temperature data is not as noisy as Lower Troposphere temperature anomalies, so we’ll assume that 17 years would be appropriate timescale to present sea surface temperature trends on global and hemispheric bases as well. And 30 years: Wikipedia defines Climate “as the weather averaged over a long period. The standard averaging period is 30 years, but other periods may be used depending on the purpose.”
But we’re using monthly data so the trends are actually for 204- and 360-month periods.
ABOUT THE GRAPHS IN THIS POST
This post does NOT present graphs of sea surface temperature anomalies, with the exception of Figures 2 and 3, which are provided as references. The graphs in this post present 17-year and 30-year linear trends of Sea Surface Temperature anomalies in Deg C per Decade on a monthly basis, and they cover the period of January 1900 to August 2011 for the observation-based Sea Surface data and the period of January 1900 to December 2099 for the model mean hindcasts and projections. Figure 1 is a sample graph of the 360-month (30-year) trends for the observations, and it includes descriptions of a few of the data points. Basically, the first data point represents the linear trend of the Sea Surface Temperature anomalies for the period of January 1900 to December 1929, and the second data point shows the linear trend of the data for the period of February 1900 to January 1930, and so on, until the last data point that covers the most recent 360-month (30-year) period of September 1981 to August 2011.
Figure 1
Note also how the trends vary on a multidecadal basis. The model-mean data do not produce these variations, as you shall see. And you’ll also see why they should, because they are important. Observed trends are dropping, but the model mean trends are not.
I’ve provided the following two comparisons of the “raw” Sea Surface Temperature anomalies and the 360-month (Figure 2) and 204-month (Figure 3) trends as references.
Figure 2
HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH
Figure 3
COMPARISONS OF SEA SURFACE TEMPERATURE ANOMALY TRENDS OF CLIMATE MODEL OUTPUTS AND INSTRUMENT-BASED OBSERVATIONS
In each of the following graphs, I’ve included the following notes. The first one reads,
The Models Do Not Produce Multidecadal Variations In Sea Surface Temperature Anomalies Comparable To Those Observed, Because They Are Not Initialized To Do So. This, As It Should Be, Is Also Evident In Trends.
And since those notes in red are the same for Figure 4 through 9, you’ll probably elect to overlook them. The other note on each of the graphs describes the difference between the observed trends for the most recent period and the trends hindcast and projected by the models. And they are significant, so don’t overlook those notes.
There’s no reason for me to repeat what’s discussed in the notes on the graphs, so I’ll present the comparisons of the 360-month and 204-month trends first for Global Sea Surface Temperature anomalies, then for the Northern Hemisphere data, and finally for the Southern Hemisphere Sea Surface Temperature anomaly data. Some of you may find the results surprising.
GLOBAL SEA SURFACE TEMPERATURE COMPARISONS
Figure 4
HHHHHHHHHHHHHHHHHHHHHHHHHHH
Figure 5
NORTHERN HEMISPHERE SEA SURFACE TEMPERATURE COMPARISONS
Figure 6
HHHHHHHHHHHHHHHHHHHHHHHHHHH
Figure 7
SOUTHERN HEMISPHERE SEA SURFACE TEMPERATURE COMPARISONS
Figure 8
HHHHHHHHHHHHHHHHHHHHHHHHHHH
Figure 9
CLOSING
Table 1 shows the observed Global and Hemispheric Sea Surface Temperature anomaly trends, 204-Month (17-Year) and 360-Month (30-Year), for period ending August 2011. Also illustrated are the trends for the Sea Surface Temperature anomalies as hindcast and projected by the model mean of the coupled climate models employed in the IPCC AR4.
Table 1
Comparing the 204-month and 360-month hindcast and projected Sea Surface Temperature anomaly trends of the coupled climate models used in the IPCC AR4 to the trends of the observed Sea Surface Temperature anomalies is yet another way to show the models have no shown no skill at replicating and projecting past and present variations in Sea Surface Temperature on multidecadal bases. Why should we believe they have any value as a means of projecting future climate?
SOURCE
Both the HADISST Sea Surface Temperature data and the IPCC AR4 Hindcast/Projection (TOS) data used in this post are available through the KNMI Climate Explorer. The HADISST data is found at the Monthly observations webpage, and the model data is found at the Monthly CMIP3+ scenario runswebpage.
I noticed no change between the age of 0 & 16 or between 17 & 32 or between 33 &48 or since I was 48. Seems to work.
Just shows a sine wave pattern and with the last one no higher than the previous, also shows no influence from other than natural cycle. If there was extra warming from humans the sine wave would have shown higher trend. The evidence has been in for ages now that the trend doesn’t show any AGW influence and like on Bob’s previous posts the ENSO over one of these sine waves, controls global ocean temperatures and there atmospheric temperatures generally.
Great stuff Bob.
I don’t think anyone has shown the climate model projections for ocean temperatures before and now we know how far off the projections are compared to the actual results.
Your analysis shows that global warming is not significant for the oceans (Santer and his 20 co-authors certainly did not know this or they wouldn’t have written that paper). The lower troposphere will soon meet Santer’s 17 year timeline for significance as well. You’ve also shown the ocean heat content actuals versus climate models before and they are also far off.
Land temperatures?, well they are four times more variable than the other measures and we should wait to see how they respond to the current La Nina. We have to start thinking there is Land amplification signal like there is a polar amplification signal (at least that is what the data shows).
“Global warming” must be renamed to “Insignificant-and-far-less-than-predicted-warming.” That is what I am going to call it from now on.
The graphs in this post present 17-year and 30-year linear trends of Sea Surface Temperature anomalies in Deg C per Decade on a monthly basis, and they cover the period of January 1900 to August 2011 for the observation-based Sea Surface data and the period of January 1900 to December 2099 for the model mean hindcasts and projections.
If I have understood it correctly, than the matter of this work is twice; first to give scientific arguments for the global SST profile in the past, and second to make hindcasts and projections for this century.
I have a problem by using linear trends in physics if there is no physical mechanism and/or no physical relation as basis discussed (like the trend of a linear increasing velocity from a constant acceleration of a mass).
That what I can see in the SST profile for the last century from the hadsst data is a lot of frequencies of different power, and because it is not out of the question, that there are frequencies lower than 1/100 y^-1 there is a need to know the temperature profile for one millennium back or two. This can help to identify low frequencies of about 1/1000 years^-1.
This is significant because it is well known that ~500 yeaes ago the climate was cold but ~1000 years ago, the climate was warm as today. And this is an important point, because it concerns the climate level in this century.
There is indeed an astronomic function known with a main frequency of 2/1827 y^-1 or 1/913.5^y-1 from the tide profile of the plutinos couple of Quaoar and Pluto. Solar spring tides are related to warm times, and nip tides are related to cold times. But this relation seems not to be only a special geometry function of that synodic couple, mainly all neighbour couples in the solar system from Mercury to Quaoar are involved in that music. Summing up all these (empirical weighted) tide functions, one gets a function (GHI x) to check the hindcast. In the end it is no different work to sum up the tide functions for this coming hole century.
http://volker-doormann.org/images/s17_0a.gif
http://volker-doormann.org/images/s17_1000.gif
http://volker-doormann.org/images/s17_1500.gif
http://volker-doormann.org/images/s17_1800.gif
http://volker-doormann.org/images/s17_1900.gif
http://volker-doormann.org/images/s17_1960.gif
http://volker-doormann.org/images/s17_2000.gif
http://volker-doormann.org/images/s17_1900b.gif
http://volker-doormann.org/images/s17_2000a.gif
I think if some twilight climate authorities argue with the term ‘trend’ as a quotient of temperature divided by an arbitrary earth time interval, this function must not taken into the scientific discussion about the causes of the climate profile with its frequencies from month to millennia.
In the other thread about the ethics of J. Hansen I have given a hindcast to his trivial triple function (CO2 + volcanos + sun):
http://volker-doormann.org/images/hansen_verification1.jpg
Sometimes in science it needs more than Occams knife.
V.
What KnR said.
It is now so obvious as to be beyond question.
Oh, and can someone please bring Al Gore up to date about extreme weather event attribution? Thanks in advance.
cui bono says:
November 20, 2011 at 2:01 am
/////////////////////////////////////////////////////////
If 1998 is seen as an outlier (due to a strong El Nino) there has been little warming (even on the basis of the Team’s adjusted/harmonised data sets) since 1995 and that is why in 2010 Phil Jones conceded that there had been no ‘statistically significant’ warming for the past 15 years (ie., since 1995).
On that basis, there is an argument that the 17 years is up at the end of 2012.
2010 was quite a warm year (another reasonably strong El Nino) and that caused Phil Jones in 2011 to suggest that there was now ‘statistically significant’ warming since 1995. Whilst this is a matter of some conjecture, since the 2011 data is not yet in, it is likely that 2011 will not be regarded as particularly warm and will tend to depress the warming since 1995 such that it is likely that once the 2011 data is in, Phil Jones would be forced to accept that between 1995 and 2011 there is no ‘statistically significant’ warming. That being the case, it makes 2012 a very interesting year.
Many are of the view that 2012 will be influenced by La Nina conditions. If so, 2012 is unlikely to be particularly warm. One can therefore foresee a significant likelihood that when the temperature for the period 1995 to 2012 is considered, it will consist of 17 years worth of data showing no ‘statistically significant’ warming trend. If it pans out like that, I wonder what Santer (and/or other members of the Team) will say.
Bob Tisdale writes, in his conclusion: “…the models have [ ] shown no skill at replicating and projecting past and present variations in Sea Surface Temperature on multidecadal bases. Why should we believe they have any value as a means of projecting future climate?”
Forgive me my deep feeling of smugness: as I read Bob’s article and studied the graphs I came to exactly the same conclusion before I ever got to the summary. The hind-casting is just so grossly out – apart from anything else!
WUWT, you are a great educator. Thanks.
Thanks Bob for again going the extra step to keep us informed and doing all these facts checks . Your contributions are much appreciated and they have been invaluable in this climate debate
steven mosher says:
November 20, 2011 at 12:46 am
1. A noisy observation dataset
2. decadal oscillations in the system
3. A signal that is small relative to these.. in the PAST
====================================================================
That nails it !!
Bob Tisdale says: November 20, 2011 at 1:56 am
“There are signifcant differences between the HADISST dataset I used in this post and HADSST2 that you’ve presented, one of which is an upward bias in the HADSST2 data in 1998 that happened when they spliced two different source datasets together.”
Indeed so, and also with HADSST3. I wrote a post here which compares them all. Except for the period 1930-1960, when there is some large oscillation, the comparison of the three with the models (graph here) shows that the model tracks the path of the observations better. This underlines Tamino’s point that you can’t expect a model mean to show the variability of climate; it is much smoother than any of its constituent model instances. There’s also a gadget in the post which illustrates the underlying determination of the trend variation.
Thank you very much Bob!
Excellent article!
No warming since the global warming alarm was sounded: False alarm.
steven mosher says:
November 20, 2011 at 12:46 am
“No. It implies three things.
1. A noisy observation dataset
2. decadal oscillations in the system
3. A signal that is small relative to these.. in the PAST”
I find it disingenious that you added the threatening “…in the PAST” to your point 3; but omitted the necessary caveat at 2.:
2. decadal oscillations in the system … that the models cannot simulate.
Because that would spell out too clearly that the models are crap, wouldn’t it, Steven.
I see Tamino has done his usual “drive-by”.
What he doesn’t understand is that he is saying the model runs which have lots of variability (almost multi-decadal) and the runs that produce almost-no-warming are the most accurate runs.
He’s right. The climate models which have 65% less warming than the average and a multidecadal natural signal are the only accurate models.
Instead of trying to defend all the climate models by saying that a few of the no-warming ones are accurate enough, they should just get rid the inaccurate high warming models.
Remember when they used to tell up how this warming was the worst in 6000 years… Then the worst in 1000… Then that it was the rate that was abnormal, rather than the peak… Then that the last 30 years was where the real problems are… then that the 30 year trend are what is important, not the last 14 years.. and so on.
Santer is really playing a losing game here because his new argument requires a dramatic warming to occur in the next 16 years when all signs point to nothing of the sort in the cards. His faith is being tested.
Kinda amusing how Mr. Foster didn’t notice the large deviation at the beginning, and the large deviation at the end of his last graph.
If his graph isn’t a poster child of how poor the models are doing, both hindcast and future cast, then I need new glasses.
He has only confirmed Mr. Tisdale’s point.
How come all the IPCC hindcast projection graphs show a reduction after 2050, but their predictions of warming in the atmosphere continue?
“David says:
November 20, 2011 at 7:48 am
How come all the IPCC hindcast projection graphs show a reduction after 2050, but their predictions of warming in the atmosphere continue?”
They are trying to find the “Hot Spot”.
Tamino’s post at
http://tamino.wordpress.com/2011/11/20/tisdale-fumbles-pielke-cheers/
is really funny. He tries to debunk Bob with Mosher’s multi-model argument. Look at his last figure – the single model runs he shows vs. the “multi model mean”. How can a multi model mean have a higher value than all the single runs? Only in climate science.
Also, he admits defeat:
“There are definitely problems with the models. For one thing, they don’t reproduce the rapid warming of sea surface temperature from 1915 to 1945 as strongly as the observed data indicate. But overall they’re not bad, and the amount of natural variability they show is realistic.”
The problem is, Tamino, that you can’t cry wolf all day and at the same time admit that your forecasting technology is still in its infancy. We KNOW it’s in its infancy; we’ve been saying that for years, it’s nice that you agree. Would you CAGW fellows now shut up for the next fifty years and stop demanding control of the global economy. Oh, and expect a slight reduction in funding, your toys are just not that important.
“cui bono says:
November 20, 2011 at 2:01 am
Can anyone confirm the moment when global warming stopped and add 17 years?”
It depends on the data set you use. On HadCrut3:
http://www.cru.uea.ac.uk/cru/data/temperature/hadcrut3gl.txt
The peak was February, 1998. However the interesting thing is that if you plot from February 1998 to the present and take the slope, it is positive. But if you plot a bit further back to May 1997, you get a negative slope. So to answer your question, see what happens with the HadCrut3 slope in May, 2014.
AndyG55 says:
November 19, 2011 at 5:05 pm
If 17 years is necessary to see a trend, HOW did the Global Warming Scare start in the late 1970′s?
Because anyone that studies climate knew at that time based on previous cycles that the climate was going to switch from a cold cycle to a warm cycle (happens every 30 years), and by predicting that things were going to get warm they stood to make themselves famous. They just needed to tie it all to pollution so they could get the environmental movement on board.
I do not understand the use of the word initialized regarding the failure to reproduce multidecadal variations in the models. It seems more likely to me that the models do not have an understanding of the causes of the multidecadal variations. If the models have a decent understanding of the multidecadal variations, is it simply that the users do not know how to initialize their own models?
“John F. Hultquist says:
November 19, 2011 at 8:33 pm
So I wonder, if natural fluctuations can swamp the signal, then natural fluctuations ought to be able to push the system beyond some presumed “tipping point” without any help from humans.”
That is why life is extinct on earth. The logic of AGW is that we must keep CO2 within a narrow range, otherwise as the earth warms due to CO2, more CO2 will be naturally released, leading to run-away warming.
So, since we now know natural variability can itself push temperatures outside that narrow range, natural variability will lead to run-away warming without any human influence leading to extinction of life on planet earth. Therefore none of us can be here, because life is already extinct from run-away global warming due to natural causes.
“richard verney says:
November 20, 2011 at 5:07 am
(ie., since 1995).
On that basis, there is an argument that the 17 years is up at the end of 2012.”
If 1995 is included, then the 17 years is up in six weeks at the end of 2011, is it not? But now we need to know exactly what Dr. Santer meant. Was it no statistically significant warming at the 95% certainty level? Or does the slope have to be 0 or less? Or can the slope be something like 0.006/year or less? And does it require both UAH and RSS to meet his number?
Nick Stokes says:
November 20, 2011 at 5:54 am
“…shows that the model tracks the path of the observations better.”
Can you, or anyone in climate science, explicate the phrase “tracks the path of the observations” in a way that meets at least some rudimentary standard of scientific reasoning?
Does it mean “reproduces the path?” If so, please explain “reproduces the path better.” Any reproduction that is not exact is a failure, right?
Does it mean “looks more like the path?” If so, then you are just eyeballing something and calling it science.
Does it mean “scores higher on our metric than any competitor?” If so, please set forth the “metric” you are using.
Does it mean “here is my bluff?” I think it means this.
Nick Stokes says: “Indeed so, and also with HADSST3. I wrote a post here which compares them all. Except for the period 1930-1960, when there is some large oscillation, the comparison of the three with the models (graph here) shows that the model tracks the path of the observations better.”
Do you really think your graph..
…shows the model mean track the multidecadal variations in the trends of any of those datasets?
Nick Stokes says: “This underlines Tamino’s point that you can’t expect a model mean to show the variability of climate; it is much smoother than any of its constituent model instances.”
Tamino’s post was a distraction from the point of my post. Nothing more, nothing less. I replied to it here:
http://bobtisdale.wordpress.com/2011/11/20/tamino-misses-the-point-and-attempts-to-distract-his-readers/