UPDATE from Girma: “My title should have been ‘How to arrive at IPCC’s climate sensitivity estimate’ instead of the original”
Guest essay by Girma Orssengo, PhD
1) IPCC’s 0.2 deg C/decade warming rate gives a change in temperature of dT = 0.6 deg C in 30 years
IPCC:
“Since IPCC’s first report in 1990, assessed projections have suggested global average temperature increases between about 0.15°C and 0.3°C per decade for 1990 to 2005. This can now be compared with observed values of about 0.2°C per decade, strengthening confidence in near-term projections.”
Source: http://www.ipcc.ch/publications_and_data/ar4/wg1/en/spmsspm-projections-of.html
2) The HadCRUT4 global mean surface temperature dataset shows a warming of 0.6 deg C from 1974 to 2004 as shown in the following graph.
Source: http://www.woodfortrees.org/plot/hadcrut4gl/from:1974/to:2004/trend/plot/hadcrut4gl/from:1974/to:2005/compress:12
3) From the following Mauna Loa data for CO2 concentration in the atmosphere, we have CO2 concentration for 1974 of C1 = 330 ppm and for 2004 of C2=378 ppm
Source: http://www.woodfortrees.org/plot/esrl-co2/compress:12
Using the above data, the climate sensitivity (CS) can be calculated using the following proportionality formula for the period from 1974 to 2004
CS = (ln (2)/ln(C2/C1))*dT = (0.693/ln(378/330))*dT = (0.693/0.136)*dT = 5.1*dT
For change in temperature of dT = 0.6 deg C from 1974 to 2004, the above relation gives
CS = 5.1 * 0.6 = 3.1 deg C, which is IPCC’s estimate of climate sensitivity and requires a warming rate of 0.2 deg C/decade.
IPCC’s warming rate of 0.2 deg C/decade is not the climate signal as it includes the warming rate due to the warming phase of the multidecadal oscillation.
To remove the warming rate due to the multidecadal oscillation of about 60 years cycle, least squares trend of 60 years period from 1945 to 2004 is calculated as shown in the following link:
Source: http://www.woodfortrees.org/plot/hadcrut4gl/from:1945/to:2004/trend/plot/hadcrut4gl/from:1945/to:2005/compress:12
This result gives a long-term warming rate of 0.08 deg C/decade. From this, for the three decades from 1974 to 2004, dT = 0.08* 3 = 0.24 deg C.
Substituting dT=0.24 deg C in the equation for Climate sensitivity for the period from 1974 to 2004 gives
CS = 5.1* dT = 5.1* 0.24 = 1.2 deg C.
IPCC’s climate sensitivity of about 3 deg C is incorrect because it includes the warming rate due to the warming phase of the multidecadal oscillation. The true climate sensitivity is only about 1.2 deg C, which is identical to the climate sensitivity with net zero-feedback, where the positive and negative climate feedbacks cancel each other.
Positive feedback of the climate is not supported by the data.
UPDATE:
To respond to the comments, I have included the following graph
Source: http://www.woodfortrees.org/plot/hadcrut4gl/mean:756/plot/hadcrut4gl/compress:12/from:1870/plot/hadcrut4gl/from:1974/to:2004/trend/plot/esrl-co2/scale:0.005/offset:-1.62/detrend:-0.1/plot/esrl-co2/scale:0.005/offset:-1.35/detrend:-0.1/plot/esrl-co2/scale:0.005/offset:-1.89/detrend:-0.1/plot/hadcrut4gl/mean:756/offset:-0.27/plot/hadcrut4gl/mean:756/offset:0.27/plot/hadcrut3sh/scale:0.00001/offset:2/from:1870/plot/hadcrut4gl/from:1949/to:2005/trend/offset:0.025/plot/hadcrut4gl/from:1949/to:2005/trend/offset:0.01
I have got a better estimate of the warming of the long-term smoothed GMST using least squares trend from 1949 to 2005 as shown in the above graph, which shows the least squares trend coincides with the Secular GMST curve for the period from 1974 to 2005. For this case, the warming rate of the least squares trend for the period from 1949 to 2005 is 0.09 deg C/decade.
This gives dT = 0.09 * 3 = 0.27 deg C, and the improved climate sensitivity estimate is
CS = 5.1*0.27 = 1.4 deg C.
That is an increase in Secular GMST of 1.4 deg C for doubling of CO2 based on the instrumental records.
Pamela Gray says:
May 18, 2013 at 8:11 am
It’s the time zone …
w.
Greg Goodman
That does not mean that is how IPCC got there , neither does it mean the second result gives the correct sensitivity.
The data and the formulas I used are the same ones used by the IPCC. The chance of arriving at the same value using the same formula and data but different method is almost nil.
The formula you used for CO2 ‘forcing’ yes. That is not how they calculate sensitivity. That comes from a number of sources mainly climate models. I don’t think they are right , just that it is incorrect to say what you presented is how they do it.
Though what you present is a fairly good caricature.
You have not commented on the fact that your method is still including “cosine warming”:
http://climategrog.wordpress.com/?attachment_id=209
If you go from peak to peak you may get what you were intending to get , and it will be a lot less.
That may be interesting.
Sorry, you’re not far off peak to peak in your third plot, I was seeing the dates on the early ones.
Girma says:
My result is based the following climate pattern of the 20th century:
http://bit.ly/15FKX0n
What result does your calculation give for the previous cycle 1875-1940?
Greg Goodman
Here are published results for the the long-term warming rate:
“…the rapidity of the warming in the late twentieth century was a result of concurrence of a secular warming trend and the warming phase of a multidecadal (~65-year period) oscillatory variation and we estimated the contribution of the former to be about 0.08 deg C per decade since ~1980.”
http://bit.ly/10ry70o
It is IDENTICAL to my value of 0.08 deg C/decade long-term warming trend.
I’m with Girma. I don’t know why everyone else has got their panties in a bunch.
The Team thought that they had the whole “CO2 causes terrifying warming” thing sorted because the warming from about 1970 to about 2000 matched their calculations of sensitivity (assuming that temps would be flatlining if it wasn’t for the increase in CO2.)
But if you look at the whole pattern from about 1880, there’s a very obvious oscillation:
http://www.woodfortrees.org/plot/gistemp/plot/hadcrut4gl/from:1880
If you look at the overall trend then clearly the Team’s sensitivity sums were nonsense.
Girma says:
May 18, 2013 at 8:20 am
First, let me start with the fact that if you have one peak and two troughs, you don’t have enough data to say whether there is or is not any “natural oscillation”. You’d need several cycles, perhaps a large number, to establish that.
Next, you claim that the oscillation has a 60 year period. Your graph in support of your claim shows the bottom of one “cycle” at about 1905, the top of the “cycle” at about 1950, and the next bottom at about 1965 …
So the first upwards half of your “cycle” seems to last about 45 years, and the next half of the cycle seems to last 15 years … say what? Yes, those add up to 60 years, but it hardly looks like an “oscillation” …
Also, you’re using an averaging period (756 months) that is half of the length of your dataset … not recommended.
Next, you’ve spliced in CO2 data on top of the temperature data, and in a very similar color, which is misleading at best. Plus, for unknown reasons, you’ve left the first 19 years of HadCRUT data off your graph. Here’s your graph, back to the start of the HadCRUT data …
The early data is already beginning to show problems with your theory, so let’s look at a longer dataset …
Here’s the BEST data, treated the same way. You can see that from 1850 on the two are quite close … but the earlier data illustrates why your method is … well, let me call it far less than optimal …
This is a totally typical situation in climate science. You have what looks like a solid cycle, it lasts for a couple of complete swings … and then it fades out and is replaced by something entirely different. You’ve been suckered by one of mother nature’s oldest tricks, my friend …
Overall, I’d say you have not even come close to establishing either of your two claims in the title. You haven’t said a word about how the IPCC got 3°C for the sensitivity, and you haven’t provided any justification, either theoretical or observational, for removing what you admit to be part of the natural signal. Handwaving about a hypothetical 60 year cycle won’t do it …
w.
Like I said, crude but effective. If you can back it up by a more rigorous study that makes it more credible. That backs up you 0.08 K/decade.
IPCC also bring all sorts exaggerated volcanic cooling, parametrised water vapour feedbacks at stuff in as well and it all goes wrong.
You are assuming no such effects in the way you calculate your sensitivity. So they are not doing the same thing but ignoring the cyclic variation. Saying they do is inaccurate, as a few people have pointed out.
You also stated you think temp causes CO2 but carry on calculating CO2 causing temps.
Since the mechanism is very different and depends on d/dt CO2 these are not interchangeable.
I think you need to look at that.
Greg Goodman on May 18, 2013 at 9:15 am posted:
http://www.woodfortrees.org/plot/hadsst2gl/from:1958/mean:12/mean:9/mean:7/derivative/normalise/plot/esrl-co2/derivative:0.003/derivative:-1.03/mean:12/mean:9/mean:7/normalise
Greg, what the heck are you trying to show there?
You have two “derivative” calls for the CO₂, but the function takes no value, read the WFT Help page. Leaving off the values gets the same result, which is “subtract each sample from one before it”, which you do twice:
http://www.woodfortrees.org/plot/hadsst2gl/from:1958/mean:12/mean:9/mean:7/derivative/normalise/plot/esrl-co2/derivative/derivative/mean:12/mean:9/mean:7/normalise
Then for both, you convert the months into 12 month running means, then convert those running means to 9 month running means, then convert the 9 month running means of the 12 month running means into 7 month running means.
WHY?
Willis
I have not claimed a constant pattern before 1869. As the climate forcing changes, the pattern also changes.
If you include the time before 1869, the residual will have a trend.
Here is a simple fit to the 21-years and 63-years moving average since 1869 with R^2 = 0.998 and R^2 = 1, respectively.
http://orssengo.com/GlobalWarming/ClimateSensitivityOfOnePointThreeDegC.png
The above graph shows clearly the climate pattern since 1869.
144 years is sufficient to establish climate relationship during that time.
@Joseph Olson –
Yes, the AGW scam is the biggest fraud in history, making Bernie Madoff’s peculations look like pocket change by comparison.
When the bubble bursts, I’d like to see the AGWers forced to reimburse the taxpayers for all the money paid to them in so-called “research” (translate: propaganda) grants and all the monies spent subsidizing environmentally as well as economically disastrous “renewable energy” schemes likie bird-killing wind turbines and habitat-destroying solar arrays.
Girma , look at rate of change of hadcrut and best.
http://www.woodfortrees.org/plot/hadcrut3vgl/mean:252/mean:188/mean:141/derivative/plot/best/mean:252/mean:188/mean:141/derivative/plot/esrl-co2/scale:0.003/offset:-1.03/detrend:-0.22/from:1982/derivative/mean:12/mean:9/mean:7
The hadcrut that you chose to use has a acceleration that would not be explained by the ln(co2) formula.
You have to have a credible model for you data. What you are trying to do is fit a cos+linear and use linear for your sensitivity. There is a significant upward curve in your plot an that is the acceleration we see here. Your cos plus linear would have a level cosine as derivative
BEST show a deceleration and a strong 30 y as well a 60y then goes sky high (probably residual UHI)
The whole idea of linear CO2 AGW does not fit the data, not even with a 60 year cosine. I agree having one is whole lot better than not but the model is just wrong. No sense in trying attribution until there is a reasonable model.
Now have a look at CO2 (I’ve filtered out 12m seasonal). There’s a lot of action in there too and guess what … when you look it actually shows it’s caused by temperature and not the other way around.
You actually suggested this was the case in one of you comments but it’s not what your post is about. I again suggest you try to take a look.
http://climategrog.wordpress.com/?attachment_id=233
kadaka, don’t worry about the params to diff ( like you said they have no effect) I trimmed down Girma’s plot and forgot to remove them.
The triple running mean is to provide a filter that does not mess up the data.
Each step is reduced by a factor of 1.3371 or as near as you can.
If you don’t, this sort of thing can happen where the runny mean inverts peaks in the data.
http://www.woodfortrees.org/plot/rss/from:1980/plot/rss/from:1980/mean:60/plot/rss/from:1980/mean:30/mean:22/mean:17
Running means are a disaster , the triple is quite a good low pass filter.
BTW, its not “converting” anything, it is three successive filters
Natural variability doesn’t appear to have fixed periodicity. As far as can se from the CET and the N. Atlantic geological records oscillations vary in length between 46 and 65 years, however the average since 1650 happens to be ~ 60 years: http://www.vukcevic.talktalk.net/NVb.htm
from 1890 I found periods 52, 62, and 65, so in a short-ish period of data FT analysis may this identify around 60yr.
Its the feedback assumptions/theories that take the 1.1C/1.2C to a total of 3.0C per doubling.
Here us a little table showing how it is actually calculated. The feedbacks are over 2.2 W/m2 per 1.0C increase in temperatures. 1.1C of doubled CO2/GHG causes a feedback increase of 8% in water vapor, a 2% decline in cloud optical thickness and reduced Albedo as ice melts. The first round of these feedbacks produces another increase in temperatures of 0.7C,
That 0.7C increase in temperature from feedbacks produces another round of feedbacks increasing the temperature another 0.45C, Then there is round #3 of feedbacks on the feedbacks so on and so on. Pretty soon, the tropopause has increased its forcing by 11.7 W/m2 and temps here are 3.0C higher. The theory assumes the surface will increase by something like the tropopause (although there is small increase in the lapse rate counted in the feedback numbers already).
http://s2.postimg.org/xkjw426dl/Stefan_Boltzmann_3_0_C_doubling.png
However, if the feedbacks are much less than this 2.2 W/m2/K, we get much less warming. And here is something rarely talked about, if the feedbacks are much more than this 2.2 W/m2/K, there is actually a runaway greenhouse effect. The actual feedbacks only need to be something like 3.0 W/m2/K and we are at 11C of warming per doubling. Bump them up to 4.0 W/m2/K or so and the oceans boil off eventually.
The feedbacks numbers have been carefully chosen to keep the numbers at 3.0C per doubling.
So far, water vapor is coming in less that 50% of that predicted, and cloud optical thickness is probably Zero as far as we can tell.
http://s13.postimg.org/7dk4nfh6f/Hadcrut4_vs_TCWV_Scatter.png
http://s9.postimg.org/y8o23z2rz/UAH_RSS_vs_TCWV_Scatter.png
Feedbacks strength versus how much warming we get.
http://s24.postimg.org/7jjj2kcgl/Feedback_Strength.png
A couple of commenters have said that isn’t how the IPCC arrived at tis ECS, and to read the IPCC report to see how they really did it. Well, I have read the IPCC report (AR4) – some parts of it many times – and it is clear that Girma Orssengo’s article is basically correct in this respect. For example, in TS.4.5 . it says “Large ensembles of climate model simulations have shown that the ability of models to simulate present climate has value in constraining climate sensitivity.”. In other words, they have matched climate sensitivity to observed temperature in the way Girma describes. Since the models are clearly driven mainly by the warming period of late 20thC, Girma’s actual calculation is pretty reasonable, though AR4 is opaque in areas like this so it is difficult to be certain.
IPCC’s big problem arose when it couldn’t find enough forcing in CO2 to match their value for ECS. The (well-known) calculations for CO2 gave them an ECS of only 1.2. What they came up with was “feedbacks”. Water vapour and albedo was easy to explain, but that could only bring ECS up to 1.9 (AR4 8.6.2.3 page 633). Ignore at this point the fact that there was no actual evidence for this “feedback”. Clouds were and still are one of the big unknowns in climate science, so that is where they created a “feedback” to complete the picture. There is no known mechanism for cloud “feedback” and no empirical observation, but by coding suitable parameters into the models [yes, they say that is what they did, see AR4 8.2.1.3] they could bring ECS up to the required 3.2.
So, in summary, temperature increase of late 20thC ==> ECS 3.2 ==> feedbacks …
… which brings me to Willis Eschenbach’s (“w”) comment:-
w says “How are you “removing the warming rate due to the multidecadal oscillation” by calculating its linear trend?“. What Girma did was to take the linear trend over 60 years, which is the length of a cycle that is very visible in the temperature record of the last 120+ years. His unspoken argument is that you have to take a temperature trend over complete cycle(s) in order to remove the cyclical effect: “To remove the warming rate due to the multidecadal oscillation of about 60 years cycle, least squares trend of 60 years period from 1945 to 2004 is calculated“. This is entirely logical as far as it goes – to check, simply calculate some linear trends in a sinewave.
I say “as far as it goes”, because although it is probably the best and only way to remove the cyclical effect, nevertheless w is correct to point out that we don’t know much at all about this cycle. We can’t assume that its amplitude is constant, that its shape is symmetrical, that it is the only factor to be removed in order to find CO2-forced temperature change, etc, etc.
So, although Girma’s explanation is a good one, and it does correctly show that the IPCC’s ECS is totally unreliable, nevertheless it cannot be used as a basis for further calculations. Before climate science can move ahead, it is necessary to understand much more about climate’s major components.
Correction: last sentence in my post above is misleading and should be ignored. Sorry, my bad!
Girma
“I am not saying CO2 is causing the warming. I believe it is the warming that is causing the increase in CO2 concentration, as the vostok ice cores show. The CO2 concentration will drop when the temperature falls.”
Although I agree with this statement, that isn’t the way the IPCC is using temp as a function of CO2. They are saying that if CO2 goes up by X amount, the temp will rise Y amount and that is simply not a valid conclusion. My point is that temp is NOT a function of CO2 when you look at the past history (long term or short term) of temp and CO2. You can calculate virtually any CS you want (including warming or cooling) just by selecting a different time span as the basis for your calculation. So what use is CS at all?
Not all of the 60 year cycle temps are the same–there are big differences, so you can’t just use one number to subtract them out of your equation.
Mike Jonas
Thank you for your reasoned comments.
Here is a published paper that arrive at recent warming rate of 0.08 deg C/decade:
“The underlying net anthropogenic warming rate in the industrial era is found to have been steady since 1910 at 0.07–0.08 °C/decade, with superimposed AMO-related ups and downs that included the early 20th century warming, the cooling of the 1960s and 1970s, the accelerated warming of the 1980s and 1990s, and the recent slowing of the warming rates.”
Tung and Zhou (2012)
Using data to attribute episodes of warming and cooling in instrumental records
http://www.pnas.org/content/110/6/2058
Gentlemen Look at Fig2 on the top post at http://climatesense-norpag.blogspot.com
if you think that CO2 drives temperatures you are forced to admit that CO2 cooled rather than warmed the world for thousands of years during the Holocene.The 60 and 1000 year cycles are the key ones for the present discussion. They are clearly the controlling factors.For the 1000 year cycle see later figures and earlier posts on my blog.
Girma says:
May 18, 2013 at 10:51 am
Huh? It’s not a “pattern” if it is constantly changing.
What you have claimed is that there is a “multidecadal oscillation”, one which you have not identified with any physical phenomenon.
Now you are saying that it is really a temporary “multidecadal oscillation” which will change in some unspecified manner with some unidentified quantity called the “climate forcing”.
Gotta say, my friend, you are not making things clearer …
Again, I fear your meaning is totally unclear. What does it mean to “establish [a] climate relationship”? Because from your graph, you have one peak and two troughs in the data. It sounded for a while like you were claiming a regular cycle, one that could legitimately be removed from the data in the manner in which we deal with say the cyclical annual variations.
But no, now that I’ve shown that your “multidecadal oscillation” pattern doesn’t exist further back than 1869, now you are saying it’s just a temporary pattern, might disappear tomorrow … but if so, what is the justification for removing it?
And despite not knowing what we’re looking for, just some ephemeral “multidecadal oscillation” that appears and disappears, you claim that 144 years of data (two troughs and one peak) are enough to “establish [a] climate relationship” … ‘fraid I can’t help you with that one …
My friend, I fear you are pursuing a blind alley. You can’t just arbitrarily decide to filter out some parts of the data because you like the result …
w.
Heres a quote from the blogpost linked earlier.
Having some passing acquantance with the above literature I would suggest that the currently most useful compilation for thinking about the record of the last 2000 years is.
Christiansen and Ljungqvist 2012
http://www.clim-past.net/8/765/2012/cp-8-765-2012.pdf
Fig.3 Im not sure how to import figs into WUWT check the original post at http://climatesense-norpag.blogspot.co
The point of most interest in Fig 3 is the present temperature peak and the MWP peak at 1000 AD which correlate approximately with the solar millenial cycle seen in Fig2. The various minima of the Little Ice age and the Dalton minimum of the early 19th century also show up well.
The general principal is to perfom spectral and wavelet analysis on the the temperature and any possibly useful driver associated time series to find any quasicyclic patterns which can be cross correlated. (possibly with appropriate time lags)
For a general review of this approach see several Scafetta papers eg
http://www.fel.duke.edu/~scafetta/pdf/scafetta-JSTP2.pdf
For decadal scale variations a 60 year cycle ,which seems to correlate temperatures and the PDO, is well established see the post” Global Cooling -Methods and Testable Decadal Predictions” at
http://climatesense-norpag.blogspot.com.
Willis, is Girma really your “friend”? If not, the repeated use of that title may be construed as somewhat condescending and, with your stature, that is really not necessary.