Warming in Last 50 Years Predicted by Natural Climate Cycles
by Roy W. Spencer, Ph. D.
One of the main conclusions of the 2007 IPCC report was that the warming over the last 50 years was most likely due to anthropogenic pollution, especially increasing atmospheric CO2 from fossil fuel burning.
But a minority of climate researchers have maintained that some — or even most — of that warming could have been due to natural causes. For instance, the Pacific Decadal Oscillation (PDO) and Atlantic Multi-decadal Oscillation (AMO) are natural modes of climate variability which have similar time scales to warming and cooling periods during the 20th Century. Also, El Nino — which is known to cause global-average warmth — has been more frequent in the last 30 years or so; the Southern Oscillation Index (SOI) is a measure of El Nino and La Nina activity.
A simple way to examine the possibility that these climate cycles might be involved in the warming over the last 50 years in to do a statistical comparison of the yearly temperature variations versus the PDO, AMO, and SOI yearly values. But of course, correlation does not prove causation.
So, what if we use the statistics BEFORE the last 50 years to come up with a model of temperature variability, and then see if that statistical model can “predict” the strong warming over the most recent 50 year period? That would be much more convincing because, if the relationship between temperature and these 3 climate indicies for the first half of the 20th Century just happened to be accidental, we sure wouldn’t expect it to accidentally predict the strong warming which has occurred in the second half of the 20th Century, would we?
Temperature, or Temperature Change Rate?
This kind of statistical comparison is usually performed with temperature. But there is greater physical justification for using the temperature change rate, instead of temperature. This is because if natural climate cycles are correlated to the time rate of change of temperature, that means they represent heating or cooling influences, such as changes in global cloud cover (albedo).
Such a relationship, shown in the plot below, would provide a causal link of these natural cycles as forcing mechanisms for temperature change, since the peak forcing then precedes the peak temperature.
Predicting Northern Hemispheric Warming Since 1960
Since most of the recent warming has occurred over the Northern Hemisphere, I chose to use the CRUTem3 yearly record of Northern Hemispheric temperature variations for the period 1900 through 2009. From this record I computed the yearly change rates in temperature. I then linearly regressed these 1-year temperature change rates against the yearly average values of the PDO, AMO, and SOI.
I used the period from 1900 through 1960 for “training” to derive this statistical relationship, then applied it to the period 1961 through 2009 to see how well it predicted the yearly temperature change rates for that 50 year period. Then, to get the model-predicted temperatures, I simply added up the temperature change rates over time.
The result of this exercise in shown in the following plot.
What is rather amazing is that the rate of observed warming of the Northern Hemisphere since the 1970’s matches that which the PDO, AMO, and SOI together predict, based upon those natural cycles’ PREVIOUS relationships to the temperature change rate (prior to 1960).
Again I want to emphasize that my use of the temperature change rate, rather than temperature, as the predicted variable is based upon the expectation that these natural modes of climate variability represent forcing mechanisms — I believe through changes in cloud cover — which then cause a lagged temperature response.
This is powerful evidence that most of the warming that the IPCC has attributed to human activities over the last 50 years could simply be due to natural, internal variability in the climate system. If true, this would also mean that (1) the climate system is much less sensitive to the CO2 content of the atmosphere than the IPCC claims, and (2) future warming from greenhouse gas emissions will be small.
<!– This entry was posted on Sunday, June 6th, 2010 at 6:51 AM and is filed under Blog Article. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed. –>
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
sphaerica says:
June 9, 2010 at 5:18 am
phlogiston says:
June 9, 2010 at 12:42 am
Here’s the point that you are missing: phenomena like the PDO, AMO and SOI are cyclical. This means that they increase and decrease in a repetitive and somewhat predictable way.
Fine… try it. Please produce for me a simple projection of the PDO, AMO and SOI into the future. If they are regular and predictable enough then that should be easy.
But you can’t, because they’re not.
Are so.
http://wattsupwiththat.com/2009/10/08/new-paper-barents-sea-temperature-correlated-to-the-amo-as-much-as-4%C2%B0c/
Tim Clark says:
June 9, 2010 at 5:40 am
Googling that name turns up a lot of entries. Can you please provide a link?
dr.bill, June 9, 2010 at 7:02 am
You make insulting, very public accusations, but you will not support them with facts.
jbar,
The large t-stat on CO2 suggests you have a collineartiy problem. 2 or more of your dependent variables are highly correlated. This is common in time series models. Try removing one independent variable and see if the coefficient on CO2 changes substanially. Also, you can check the eigenvalues and variance inflation factors using your stats program. For further info search for ‘Collinearity Diagnostics’. It also looks like you may have a problem with serial correlation. Most stat programs will provide diagnostics and alternatives to OLS that will correct for the problems. The bottom line is that you should not draw inference from you results until these problems are addressed.
The larger problem I see is that we need to let theory be our guide with respect to selection of explanatory variables. We can’t fall into the trap of grasping at whatever variables appear to improve our model fit. Theory suggests that all 4 of your explanatory variables belong in the regression, but they are not the ONLY 4. The 2 competing models in this thread both suffer from this problem. The 3 or 4 variables are serving as proxies for the true set of explanatory variables.
For example, your statement that CO2 explains 52% of variation is only true if your model is correctly specified and contains ALL explanatory variables. Theory suggests CO2 and other GHG’s are important so it belongs in the regresssion. However, as you add the true set of explanatory variables to your regression, the amount of variation explained by CO2 will decline.
I work in finance and there is a nice example. For decades, stock returns for a given company (percent change in price) were regressed on a constant and a single factor, the ‘market portfolio’ (think a broad stock index like the S&P 500). The results were used to measure beta, the amount of variation in a company’s stock returns explained by variation in the overall market. However, Fama and French showed that when you added a set of ‘true’ explanatory variables to the regression, the coefficient on the market portfolio was no longer statistically different from zero. My intuition is that CO2 is analogous to the market portfolio. Once climate researchers develop better theories, it will guide them toward the set of true explanatory variables and the role of CO2 in climate change theories will be greatly diminished.
stephen richards says:
June 8, 2010 at 3:17 am
Carlo says:
June 8, 2010 at 3:09 am
> Is this so that your mediocre climate school can adjust it’s message to
>convince more of us that AGW is a valid science? or are you genuinely
>trying to engage in debate?
I am not from a “climate school”, but we are studying tool to improve online deliberation and discussion and in this experiment Climate Change is the topic faced.
It is well established that oceanic cycles and pressure cell systems developed over the oceans that then impinge upon the land add heat and allow heat to be reduced over land. The mechanism from ocean heat to atmosphere heat is known and can be calculated.
Oceanic cycles and SST are based on well known layered temperature cycles between warm and cool occurrences (warm means the upper layer is fairly still and is warm, cool means the upper layer has been mixed with the cooler layer below and is choppy).
To add CO2 to this model of oceanic indices, you must calculate the affect of IR LW heat penetration into the warming of the oceans. The air temps are the dependent variable, the oceanic conditions are the independent variable. Therefore you must consider CO2 also an independent variable. However, simply adding in the rise in CO2 is meaningless in this study. You must calculate the greenhouse affect on the warming/lack of warming in the oceanic hydrological cycle. And those of you who know about the ability of LW IR heat penetration into the oceans know that this independent variable does not make a differences. Its effect is WAAAYYYY below the SD of the natural oceanic cycles. LW IR can only heat the thin surface tension skin of the oceans and is evaporated away almost as soon as the thin skin warms. The bottom line, oceans cannot store the affects of LW IR heat penetration. Hell, it has a hard enough time keeping in the SW IR.
Dave Springer says:
June 9, 2010 at 2:05 am
The average temperature of the ocean is 5C. It has hundreds of times more thermal mass than the air.
We’re basically at the mercy of the mixing rate between warm shallow surface water (16C, 10%) and the vast cold deep (4C, 90%). If it mixes too fast – booyah! – bye bye interglacial, hello ice age. There’s nothing puny man can do to warm the deep ocean and unless that gets warmed it’s a giant heat sink inevitably sucking us towards glacier city. “””
This may sound like a silly question (maybe it is !) ?
If the earth core is somewhere in the 5,000 to 10,000 deg C range; and the surface/lower troposphere is 15 deg C; and you say that the deep oceans are at 4 deg C; and are sucking in ‘heat’ from the warm surface waters; where the hell is all that heat piling up down there.
It would seem that ‘heat’ is flowing away from the earth core; and you say it is flow down away from the surface; so somewhere down there, must be a humdinger of a heat storage gizmo.
And why do you say that the deep oceans are at 4 deg C and the average is at 5 deg C ? Have we measured those values.
I’m not quibbling with your numbers (I don’t don’t have any idea); I’d just like to understand where you get them from.
I DO get nervous, when people say the deep oceans are at 4 deg C. I know that is common for deep fresh water lakes; as a result of the 4 deg C maximum density of fresh water; but salt water does not have a maximum density short of freezing; at least for the levels of salinity in the oceans. Now I don’t know if there is maybe a pressure effect that alters that sea water behavior; but I get skittish when people say the deep oceans are at 4 deg C.
But I also want to know where all theat heat is ending up if it is going both in from the surface and out from the core.
“”” Wren says:
June 8, 2010 at 9:26 pm
Sphaerica says:
June 8, 2010 at 9:54 am
Dr. Spencer,
Various web sites define these indexes as:
The PDO Index is calculated by spatially averaging the monthly sea surface temperature (SST) of the Pacific Ocean north of 20°N.
The Southern Oscillation Index (SOI) is calculated from the monthly or seasonal fluctuations in the air pressure difference between Tahiti and Darwin.
[Note that air pressure is dependent upon, and therefore a proxy for, temperature.] “””
“”” [Note that air pressure is dependent upon, and therefore a proxy for, temperature.] “””
Just wanted to be absolutely certain that I quoted you correctly.
So air pressure is a proxy for Temperature.
I’m vaguely familiar with the gas laws; equations of state and all that; but let’s just for the moment stick with the perfect gas from which we get:- pV = nRT
So the number (n) of mols of gas in the atmosphere is constant (maybe) and R is a Universal constant; and p is a proxy for Temperature (you say), so presumably p is proportional to T.
That seems to leave me with V must be constant (well at least as constant as (n) is.) Well I suppose you can argue that the atmosphere is bounded by the surface of the earth and the outer edge of the universe; but then maybe that Volume is expanding; isn’t space supposed to be expanding.
Maybe there really isn’t any fixed volume to the eartyh atmosphere; My ancient Radio-Physics training says that the ionosphere keeps moving up and down, in response to the sun’s action; so I am not used to thinking of the atmosphere as being constant Volume; so how can pressure possibly be a proxy for Temperature; well other than in the Michael Mann approximation of proxy ?
Jbar says:
June 9, 2010 at 3:50 am
“However, I am convinced that if Dr. Spencer did a regression against ALL available variables including CO2 (instead of “cherry picking” only three), then his regression would put CO2 in its proper place as driving the very long term trend while the AMO would assume its proper place in driving much of the multidecadal variation and boosting the recent warming.”
You mean, it would show the relationship which you favor?
You appear to be missing the elephant in the room: it does not matter to the skeptics if you can provide multiple models which fit the data. It is the AGW advocates who have the burden of proving CO2 is the culprit. If there are multiple models which produce the same result, then how can you pin the rap on CO2?
Dr Bill:
With respect, I think you are playing into the hands of the troll who is posting under the alias of ‘Sphaerica’.
The troll’s blatant purpose is to deflect attention and discussion away from consideration of Dr Spencer’s excellent analysis and its implications. He/she/they are doing this by posting ridiculous assertions (including untrue paraphrases that grossly distort the words of others) and by changing the subject whenever the daft nature of the assertions is pointed out.
You, me and others have demonstrated that the troll’s postings are factually in error, illogical in content, and often not related to Dr Spencer’s work in any way.
So, I suggest that the troll be ignored. Let him/her/they post rubbish and allow others to evaluate it for themselves. He/she/they will soon get tired of being ignored and unintrrupted discussion of Dr Spencer’s work can then occur.
Richard
George E Smith
Google any reference, the water column 2000-10000 has temperature 0-3 C. Huge gravitational pressure keeps water close to temp of minimum volume (max density) i.e. for seawater, zero. Heat budget takes second place to gravity.
The Earth’s core heating affect is already in the equation as a constant. No need to consider it. The only thing needed is to determine how much SW IR gets passed the outer atmosphere it do its heating job (which is highly variable exactly due to our atmosphere), where that heat goes once it warms the oceans, and how it is dissipated, in terms of climatological affects. Most of the equation has been figured out. It’s the dissipation that we wonder about when SW IR is getting through the atmospheric conditions in higher amounts (IE clear sky, calm “doldrum” winds). It is leaving Earth (or we just can’t find it being stored anywhere), but how is it leaving? Through storm cells? Polar regions? That process is being figured out.
George E. Smith says:
June 9, 2010 at 9:44 am
One short, simplistic reference
The point is that the two are tied together. One is not necessarily a direct, proportional, linear proxy for the other, but they are certainly interrelated… and can you think of any other mechanism, any other source of energy other than heat/temperature, which would cause consistent, measurable variances in air pressure?
George E. Smith says:
June 9, 2010 at 9:44 am
See also Wikipedia: Southern Oscillation
Richard S Courtney: June 9, 2010 at 10:16 am
Richard, I agree wholeheartedly with your attitude towards trolls, and it is the approach that I normally follow myself. If you track back some of the links and comments from sphaerica’s website to other places, there’s little doubt of the troll status, so once in a while I find it good practice to explicitly ‘call BS’, just in case some honest citizen might be taking them seriously. Other than that, and I might rightly be accused of puerility in this respect, I occasionally do it just for fun, somewhat in the manner of a kid poking at an anthill. Mea culpa. ☺
/dr.bill
Dave Springer says
June 9, 2:05 am
We’re basically at the mercy of the mixing rate between warm shallow surface water (16C, 10%) and the vast cold deep (4C, 90%). If it mixes too fast – booyah! – bye bye interglacial, hello ice age.
On an earlier thread I made a back-of-envelope calculation that, for the deep ocean to “suck” all the heat down from the surface, so that the sea surface layer and troposphere were nowhere more than 3C, the mean temperature of the deep ocean would need to increase only 0.4C. Maybe someone could check this.
There is a little known quantum effect in very deep water or any other fluid in which due to intense gravitational pressure, space-time becomes slightly “crinkled”. When a baryonic particle passes through one of the convoluted space-time nodes associated with this crinkling, vibrational trajectory becomes folded and thus thermal Brownian movement is negated or annihilated. The consequence is the disappearence of a small amount of heat from such gravitationally pressured fluid volumes, at a steady rate. The deep ocean is, in this way eating Trenberth’s missing heat, and it aint going to give it back.
(not)
Thanks for this nugget Roy. 🙂
I’ve always thought that the natural variability of our ‘various’ climates was more than the IPCC would give credence to. However, I think that there is more than just the phase differences between Earth’s differing reactance entities. Solar output is also important.
If solar output doesn’t ‘resonate’ at a ‘reactance peak’ of Earth’s reactance entities there will be little energy transfer (I refer to attractors here).
Best regards, Ray Dart.
rogerkni says:
June 8, 2010 at 11:16 am
OMG did I write all that?
Tom and jbar,
Keep up your discussions please as I’m very much enjoying them.
May I also make a request that in your discussions that you also include the GCMs as after all it is the ‘without CO2 forcing included versus with CO2 forcing included’ ensemble model runs that the IPCC relies upon to make there case that the late 20th century ‘unprecedented’ warming trend can only be explained by invoking CO2.
jbar, please note that this is vital to the IPCC’s case for projected future catastrophic global warming and to the cry that we ‘must act now to save the planet’. The IPCC make no specific claims for catastrophic warming due to the effects of CO2 from man’s emissions of GHG prior to the late 20th century. If you like the late 20th century warming trend is like the curved ramp at the end of British aircraft carriers. Without it (unless they are Harrier VTOL jump jets) the planes can’t take of. So it is with the GCM’s and their virtual reality world predictions of CAGW. IOW the hockey stick (the acceleration in warming rate towards the end of the 20th century) matters a great deal to the IPCC.
vukcevic wrote:
“Northern Hemisphere temperatures trend follows closely the Arctic, with a possible few years phase difference.”
My sense is that many climate enthusiasts overlook the correlation between Arctic precipitation and Arctic temperature. The key is the reversal of temperature-precipitation relations either side of the freezing point. (We need to bear in mind the continentality of the winter ice sheet. SH is a whole different story and SH dynamics are not so nonlinear.)
vukcevic wrote:
“Correlation between variation in the Earth’s magnetic field and the Arctic anomaly is very convincing, but physical relationship as yet unknown.
http://www.vukcevic.talktalk.net/NFC1.htm“
Consider that temperature & pressure are not independent. See the work of Barkin on pressure-related north-south decadal/multi-decadal dynamics of Earth shells. The mystery dissolves.
Very interesting graph vukcevic. I encourage you to send the graph to Barkin & Sidorenkov.
M77 wrote:
“The training period doesn’t seem to fit very well before 1920.”
Stratospheric volcanoes:
http://www.sfu.ca/~plv/SAOT_SO_SEP_MSI_IVI2.png
More here (including definitions):
http://www.sfu.ca/~plv/VolcanoStratosphereSLAM.htm
SSTs are not independent of insolation (not to be confused with irradiance), pressure, & circulation, so talk of “ocean cycles” is basically synonymous with talk of “cloud & optical extinction cycles” (aside from some spatial turbulence).
Good observation.
The residuals for the model training period are 100% nonrandom — & related to the following:
http://www.sfu.ca/~plv/PolarMotionPeriodMorlet2piPower.PNG
Beyond that, although the model stimulates valuable discussion & thinking about natural cycles, it only conveys that temperature is related to temperature.
Just a point of teaching: Roy has developed a classical statistical model in which a segment of data conditions is used to statistically “train” future (but known) temperatures.
A case in point: The El Nino models are divided between predictive statistical models, considered to be the gold standard, and newer predictive models called “dynamical” models. Statistical models take the parameters of the past that produced known temperature trends and then developed probability outcomes, were these parameters to show up again. The more the historical parameters produce similar trends time after time, the better the statistical models are at being right. Dynamical models are created based on proposed mathematical constructs of how climate conditions work and the resulting temperature trends. Several models are created (in fact not a few of the dynamical El Nino models have GHG influences calculated in), each with its own set of “how climate works” mathematical scenarios, which are then compared to the statistical models. The El Nino models are still in the research stage. It will be interesting to see which models came closest to actual events.
THE best place to look at the difference between statistical and dynamical models is here:
http://iri.columbia.edu/climate/ENSO/currentinfo/SST_table.html
I keep tabs on this site monthly. It is informative and instructive as a teaching tool that I use to develop my informed opinions about our current set of climate models. Too bad Hansen does not see the value in such open comparisons between his dynamical models and statistical “gold standard” constructs like the one Roy has put together.
Paul Vaughan
June 10, 5:41
so talk of ocean cycles is .. synonymous with cloud and optical extinction cycles …
it only that temperature is related to temperature.
I am surprised that you seem to be among those who consider the ocean only 10m deep. It is likely the THC circulation and Kelvin waves drive cycles such as ENSO and the ocean oscillations. If as some imply deep to surface mixing was limited and insignificant, then the oceans would be anoxic and we would be in the middle of an extinction event such as the end-Permian.
But were not. Climate causation is probably more ocean to atmosphere than the reverse.