Evidence of a Significant Solar Imprint in Annual Globally Averaged Temperature Trends - Part 1
26 03 2008NOTE: This essay represents a collaboration over a period of a week via email between myself and Basil Copeland. Basil did the statistical heavy lifting and the majority of writing, while I provided suggestions, reviews, some ideas, editing, and of course this forum. Basil deserves all our thanks for his labor. This is part one of a two part series. -Anthony
Evidence of a Significant Solar Imprint in Annual Globally Averaged Temperature TrendsBy Basil Copeland and Anthony Watts
It is very unlikely that the 20th-century warming can be explained by natural causes. The late 20th century has been unusually warm.
So begins the IPCC AR4 WG1 response to Frequently Asked Question 9.2 (Can the Warming of the 20th Century be Explained by Natural Variability?). Chapter 3 of the WG1 report begins:
Global mean surface temperatures have risen by 0.74°C ± 0.18°C when estimated by a linear trend over the last 100 years (1906-2005). The rate of warming over the last 50 years is almost double that over the last 100 years (0.13°C ± 0.03°C vs. 0.07°C ± 0.02°C per decade).
Was the warming of the late 20th century really that unusual? In recent posts Anthony has noted the substantial anecdotal evidence for a period of unusual warming in the earlier half of the 20th century. The representation by the IPCC of global trends over the past 100 years seems almost designed to hide the fact that during the early decades of the 20th century, well before the recent acceleration in anthropogenic CO2 emissions beginning in the middle of the 20th century, global temperature increased at rates comparable to the rate of increase at the end of the 20th century.
I recently began looking at the longer term globally averaged temperature series to see what they show with respect to how late 20th century warming compared to warming earlier in the 20th century. In what follows, I’m presenting just part of the current research I’m currently undertaking. At times, I may overlook details or a context, or skip some things, for the sake of brevity. For example, I’m looking at two long-term series of globally averaged annual temperature trends, HadCRUTv3 and GHCN-ERSSTv2. Most of what I present here will be based on HadCRUTv3, though the principal findings will hold true for GHCN-ERSSTv2.
I began by smoothing the data with a Hodrick-Prescott (HP) filter with lambda=100. (More on the value of lambda later.) The results are presented in Figure 1.
Figure 1 - click for a larger image
The figure shows the actual data time series, a cyclical pattern in the data that is removed by the HP filter, and a smoothed long term low frequency trend that results from filtering out the short term higher frequency cyclical component. Hodrick-Prescott is designed to distinguish short term cyclical activity from longer term processes.
For those with an electrical engineering background, you could think of it much like a bandpass filter which also has uses in meteorology:
Outside of electronics and signal processing, one example of the use of band-pass filters is in the atmospheric sciences. It is common to band-pass filter recent meteorological data with a period range of, for example, 3 to 10 days, so that only cyclones remain as fluctuations in the data fields.
(Note: For those that wish to try out the HP filter, a freeware Excel plugin exists for it which you can download here)
When applied to globally averaged temperature, it works to extract the longer term trend from variations in temperature that are of short term duration. It is somewhat like a filter that filters out “noise,” but in this case the short term cyclical variations in the data are not noise, but are themselves oscillations of a shorter term that may have a basis in physical processes.
For example, in Figure 1, in the cyclical component shown at the bottom of the figure, we can clearly see evidence of the 1998 Super El Niño. While not the current focus, I believe that analysis of the cyclical component may show significant correlations with known shorter term oscillations in globally averaged temperature, and that this may be a fruitful area for further research on the usefulness of Hodrick-Prescott filtering for the study of global or regional variations in temperature.
My original interest was in comparing rates of change between the smoothed series during the 1920’s and 1930’s with the rates of change during the 1980’s and 1990’s. Without getting into details (ask questions in comments if you have them), using HadCRUTv3 the rate of change during the early part of the 20th century was almost identical to the rate of change at the end of the century. Could there be some sense in which the warming at the end of the 20th century was a repeat of the pattern seen in the earlier part of the century? Since the rate of increase in greenhouse gas emissions was much lower in the earlier part of the century, what could possibly explain why temperatures increased for so long during that period at a rate comparable to that experienced during the recent warming?
As I examined the data in more detail, I was surprised by what I found. When working with a smoothed but non-linear “trend” like that shown in Figure 1, we compute the first differences of the series to calculate the average rate of change over any given period of time. A priori, there was no reason to anticipate a particular pattern in time (or “secular pattern”) to the differenced series. But I found one, and it was immediately obvious that I was looking at a secular pattern that had peaks closely matching the 22 year Hale solar cycle. The resulting pattern in the first differences is presented in Figure 2, with annotations showing how the peaks in the pattern correspond to peaks in the 22 year Hale cycle.
Besides the obvious correspondence in the peaks of the first differences in the smoothed series to peaks of the 22 year Hale solar cycle, there is a kind of “sinus rhythm” in the pattern that appears to correspond, roughly, to three Hale cycles, or 66 years. Beginning in 1876/1870, the rate of change begins a long decline from a peak of about +0.011 (since these are annual rates of change, a decadal equivalent would be 10 times this, or +0.11C/decade) into negative territory where it bottoms out about -0.013, before reversing and climbing back to the next peak in 1896/1893. A similar sinusoidal pattern, descending down into negative annual rates of change before climbing back to the next peak, is evident from 1896/1893 to 1914/1917. Then the pattern breaks, and in the third Hale cycle of the triplet, the trough between the 1914/1917 peak and the 1936/1937 peak is very shallow, with annual rates of change never falling below +0.012, let alone into the negative territory seen after the previous two peaks. This same basic pattern is repeated for the next three cycles: two sinusoidal cycles that descend into negative territory, followed by a third cycle with a shallow trough and rates of change that never descend below +0.012. The shallow troughs of the cycles from 1914/1917 to 1936/1937, and 1979/1979 to 1997/2000, correspond to the rapid warming of the 1920’s and 1930’s, and then again to the rapid warming of the 1980’s and 1990’s.
While not as well known as the 22 year Hale cycle, or the 11 year Schwabe cycle, there is support in the climate science literature for something on the order of a 66 year climate cycle. Schlesinger and Ramankutty (1994) found evidence of a 65-70 year climate cycle in a number of temperature records, which they attributed to a 50-88 year cycle in the NAO. Interestingly, they sought to infer from this that these oscillations were obscuring the effect of AGW. But that probably misconstrues the significance of the mid 20th century cooling phase. In any case, the evidence for a climate cycle on the order of 65-70 years extends well into the past. Kerr (2000) links the AMO to paleoclimate proxies indicating a periodicity on the order of 70 years. What I think they may be missing is that this longer term cycle shows evidence of being modulated by bidecadal rhythms. When the AMO is filtered using HP filtering, it shows major peaks in 1926 and 1997, a period of 71 years. But there are smaller peaks at 1951 and 1979, indicating that shorter periods of 25, 28, and 18 years, or roughly bidecadal oscillations. There is a growing body of literature pointing to bidecadal periodicity in climate records that point to a solar origin. See, for instance, Rasporov, et al, (2004). A 65-70 year climate cycle may simply be a terrestrial driven harmonic of bidecadal rhythms that are solar in origin.
In terms of the underlying rates of change, the warming of the late 20th century appears to be no more “unusual” than the warming during the 1920’s and 1930’s. Both appear to have their origin in a solar cycle phenomenon in which the sinusoidal pattern in the underlying smoothed trend is modulated so that annual rates of change remain strongly positive for the duration of the third cycle, with the source of this third cycle modulation perhaps related to long term trends in oceanic oscillations. It is purely speculative, of course, but if this 66 year pattern (3 Hale cycles) repeats itself, we should see a long descent into negative territory where the underlying smoothed trend has a negative rate of change, i.e. a period of cooling like that experienced in the late 1800’s and then again midway through the 20th century.
Figure 2 - click for a larger image
Figure 2 uses a default value of lambda (the parameter that determines how much smoothing results from Hodrick-Prescott filtering) that is 100 times the square of the data frequency, which for annual data would be 100. This is conventional, and is consistent with the lambda used for quarterly data in the seminal research on this technique by Hodrick and Prescott. I’m aware, though, of arguments for using a much lower lambda, which would result in much less smoothing.
In Part 2, we will look at the effect of filtering with a lower value of lambda. The results are interesting, and surprising.
Part 2 is now online here
This truly has been a collaborative effort. Anthony should have changed all the first person singular references to a first person plural. His role in this becomes even more critical in Part II, when it comes to explaining some of the science we are seeing in all this. If he wants, he can put that part in the first person singular for himself. That way, if “we’re” wrong, he gets the blame, not me!
Maybe, when we’re done, we’ll put the two parts together into a single integrated PDF that can be hosted somewhere for easy download. http://www.icecap.us, maybe?
I suspect there will be lots of questions about the use of HP filtering. I mainly plan to let the results speak for themselves, and that includes the results in Part II. So I’m not going to expend a lot of effort defending the use of HP filtering. It is just a tool, like spectral or wavelet analysis to help us extract signals from complex data. It just happens to be one that is particularly useful for a certain kind of time series analysis (time series that are not stationary when differenced, but rather show non random first differences — like we see in Figure 2). Again, the results will speak for themselves, and I would encourage commenters not to get too distracted by the mere novelty of HP filtering until we see what it produces in Part II.
Basil
Thanks Basil,
I’m really beginning to see earth’s atmospheric processes more like that of an analog circuit with a variety of electrical components. There’s voltage, current, capacitance, reluctance and inductance in the irradiance-air-ocean systems.
I think an analog computer might very well model the earth’s atmosphere more accurately than a digital one. Digital signals do not exist in nature, but analog signals are abundant.
Wow.
A question I have is why didn’t the cooling of the 1940s-1960s cool us back to where we were in the early 20th century? Or is that what the whole surface record/UHI effect is about? We did, but we don’t see it because of an urban effect GISS/CRU refuse to admit exists?
you guys should publish this in a journal. E&E may take it if no one else will.
One question though–why did you use the particular filter that you used? I am merely curious, and I think your detractors will pick up on that and ignore the rest, merely saying that “you need to learn some climate science,” when clearly I think you know what you’re talking about here.
REPLY: When you are looking for a signal of a particular frequency, a bandpass filter excludes other frequencies outside of the range you specify. That is essentially what the HP filter does. As referenced in the text, it has other uses in meteorology, so it’s use here is not without precedence. -Anthony
Anthony: Are we allowed to get ahead of you and post links to graphs of, say, the PDO, which also illustrate (near to) 22 and 44 year cycles? Or would you prefer us to wait?
Here’s a graph, though, that won’t get the cart before the horse. It’s of the number of posts at Real Climate since it opened. The opening month with its 41 posts is excluded, since it was anomalous. Note the drastic decline in the trend over the last year that seems to coincide with the drop in global temperature.
http://tinypic.com/fullsize.php?pic=eupwg8&s=3&capwidth=false
REPLY: Lets wait, it is never good to spoil the end of the movie or book for others.
I too have used filters: to increase the slope of frequency bands surrounding the one I wanted in fast onset tone pips used in auditory brainstem research. (Side note: There are so many better uses for filters than in cigs.) But back to my thought about your use. I have been cogitating on the pattern of cycle change and sunspot overlap during normal periods and its definition, versus cycle change in “minimal” periods. Would this filter work on actual sunspot data between normal cycles versus historic “minimals” to discern a predictive pattern? Maybe even a mathematical pattern? By the way, all these pictures of waves have sent me back to my days of research in audiology with a great deal of sentiment. Unfortunately, I found the Ivory tower environment to be less than pure, as in “We only fund research that supports our belief”, kind of like the current crop of global warmers.
I thank you both, if you were politicians the world would be a much better place.
German News this morning is full of ” The Extreme Weather Conference” in Hamburg which started today, with apparently, more than 700 participants, it will be interesting to see if they produce any ‘extreme’ results.
A question I have is why didn’t the cooling of the 1940s-1960s cool us back to where we were in the early 20th century
Maybe it did. Just a thought.
(Or would you rely on the adjusted surface station metadata?)
Anthony, your comment about analog circuitry reminds me of a course I took in ecosystem modeling many years ago before desktop computer existed. The professor first trotted out a simple circuit board with a couple of voltmeters and potentiometers and demonstrated how fiddling with the input parameters (resistance settings) affected the output (voltage readings on the meters). Although it only modeled a simple prey-predator relationship, it sure made the point in an easily grasped way.
Thanks for your reply Anthony, I appreciate the work you and Basil did on this piece.
I thank you both, if you were politicians the world would be a much better place.
Consider that the Rev wound up making all these wonderful discoveries as a result of being hounded out during a local election.
MattN:
“A question I have is why didn’t the cooling of the 1940s-1960s cool us back to where we were in the early 20th century?”
I think the answer lies in the increased solar output during the 20th Century.
Here you can see the solar irradiance increase over a period of 50 years at the start of the 20th Cent and achieved a plateau for the latter half. Cyclical variations discussed in this post might have very well have set the mark back to pre-20th Cent, but for the increased energy pumped into the system.
UHIE is becomes more of a problem in the latter 20th Cent as urban growth accelerates, IMHO.
If we are going to use analog circuits as an analogy, would a hurricane be a short circuit in our weather machine?
Anthony,
Yes, I think you could model the weather system as a complex electical circuit
- with inductance, capacitance, resistance etc
- the weather system will have it’s own set of reasonances - like the El Nino/La nina cycle
- and (we postulate) it is being ‘pumped’ or stimulated somehow by the sun’s period oscillations
- which will produce reasonances with strong amplitudes when the cycles coincide & less strong, small reasonances when the cycles are anti-phase.
But I disagree that an analogue computer is better at modelling this than an digital one
- digital systems are perfectly capable of simulation these systems
- in fact their more reliable than analogue computers, and more accurate
- provided they’re properly programmed, and given good input data
- Garbage In - Garbage Out - is true if you’re talking about any sort of computer - analogue or digital!
Alan,
Tell me about about it! Every time I turned the radio on in the car, that’s what I heard. Not to disappoint you, they’ve produced lots of “extreme” results, while snow falls outside! On NDR radio I heard a report that they’ve been running every hour all day long with the following contents:
1. Global warming is continuing and accelerating.
2. A big 450 sq. km piece of ice broke off Antarctica this morning - another sign of global warming.
3. One scientist, I don’t recall his name, says Greenland will melt in the next 700 years. Causing sea levels to rise 1 meter per century, which he states will be catastrophic.
4. Global warming is going to be very expensive for Germany, costing up to 800 billion euros in the next 50 years.
5. Global warming will cause more severe droughts, more floods and more weather extremes.
6. Water shortages in Germany will lead to power outages,
7. and agricultural losses, etc. etc. etc.
And on it went for the entire day, and most likely the rest of the week.
The German weather service, DWD, is filled with Hansens and Schmidts.
The Germans have certainly taken a few pages out of Goebbel’s playbook. Apparently they’ve learned nothing from their previous follies.
Excellent work Gentlemen !
Thank you.
It is amusing to me that after all the noise that the AGW lobby have made while attempting to dismiss the solar effect over a period of years that they suddenly appear to have found merit in it as a partial explanation for why their predictions are so far out of whack with observed data.
They now have the nerve to postulate that natural variation has temporarily overridden CO2 forcing.
Hello this is Gaia calling NASA/GISS, come in NASA/GISS do you read me ?
Sorry to burst your bubble NASA/GISS but the changes we’ve seen have all been naturally driven, CO2 was merely along for the ride.
You need to find a free lunch counter somewhere else !
Over and out.
Gaia
Anthony — from the first time I heard the hypothesis that variations in the sun’s activity could alter cloud cover it struck me that it made the sun-earth system like a transistor. Small changes in cloud cover correlated with variations in the sun’s activity are like small signals applied to the transistor gate. They modulate the much larger flow of sunlight getting to the earth’s surface. The fact that the sun’s radiation output (especially UV but also TSI) is also varying in sync adds to the effect — but looking only at those small fluctuations missing the much larger “transistor amplification” mechanism. In that context, the ocean oscillations (PDO, AMO, NAO etc) help to “tune” the climate response to the amplified solar signal. The natural frequency of response of the ocean systems would, however, likely differ by ocean basin depending on the shape and size of each basin, the relative exposure to solar radiation, relative depths etc.
REPLY: Excellent analogy. Cloud cover is indeed like the gate of a PNP transistor. Given that the temperature chnage has been about .7 rather than .3, I’ll wager it is a silicon rather than a germanium transistor.
I think there’s another blog post in this. “Earth as Electric Circuit” perhaps.
Lets just hope no silicon controlled rectifiers (SCR) exists in the system.
DNorris says:
“I think the answer lies in the increased solar output during the 20th Century. ”
The graph you linked to is old data. Most solar scientists feel the sun’s radiative output has been stable for a long time.
I really appreciate this site and the level of discussion - I have just completed a review of climate science and the role of the IPCC (in house - my group advises a lot of conservation organisations who are frankly badly advised by computer modelling climatologists) - on the issue of why the 1940s dip did not take us back - the answer may lie in a combination of ocean dynamics and solar effects on cloud coverage (both overall percentage and spatial distrubution) - these are more easily followed in the recent warming/cooling period because of more extensive monitoring - of ocean heat content changes and sea surface changes, as well as ISCCP data - the period 1980-2001 shows quite clearly a 4% drop in cloudiness, and NASA GISS pick up the flux of SW radiation to the surface - my sense is that the oceans stored the previous heat wave, as with this last one - but not for long - they lose heat more rapidly than currently modelled following some kind of phase change - I can see such a change at the solar max of 2001/2002. We need to think cross-disciplinary, and I have to say, though IPCC try to, they don’t really get it - nor does Hadley - and until they do, all will refer to blips in general trends, rather than cycles and phase changes.
Thanks again,
Peter
Basil,
The science that says CO2 should cause some warming is well understood and not really questioned. Can your analysis be used to estimate how much CO2 related warming has occurred?
“A question I have is why didn’t the cooling of the 1940s-1960s cool us back to where we were in the early 20th century
Maybe it did. Just a thought.
(Or would you rely on the adjusted surface station metadata?)”
Good point, the US surface network, despite the the obvious problems documented at Surfacestations.org, is still the best in the world. It shows our current temperatures almost match those of the 1930’s (GISS USHCN). Perhaps this is a local phenomena isolated to the US, but it I suspect it is a closer match to the true global trend that the HadCRUT plot.
Regarding the TSI plot referenced by DNorris above, use caution as the author (Lean 2001) has since backtracked. This has been a point of discussion in the latest Svalgaard thread over at Climate Audit.
Basil and Anthony - thanks for the great write-up, very understandable and compelling. I suspect many of the readers/commenters are engineers like myself, not scientists. The comparisons to analog electrical circuits are very helpful to quickly grasp the concepts (as opposed to getting lost in the math).
Raven & Jeff C.
Can I get sources? If I am wrong, I need to change my opinion.
Thanks
OK. Let’s see. The climate system can be seen as an circuit system, complete with inputs and outputs. The Sun’s input is split unto 2 components: 1) Low freq rectified for baseline power, and 2) Higher frequency components filtered for input into a signal processing network. So far, the AGW folks have been concentrating on the power supply part of the input, and have smoothed out the signals that may be modulating the system output in major ways.
But, I speculate.
Great posts. Thanks.
DNorris,
Here is a link from the Climate Audit Svalgaard thread regarding TSI:
http://www.climateaudit.org/?p=2868#comment-227038
This is also discussed elsewhere in the thread. Note the Judith Lean trace (brown) is significantly out of family with the others. I don’t recall where I read she had retracted the data. I’ll see if I can find it and post a link.
Raven, see Pete’s latest graph, comment #207 in the Svalgaard #4 thread at climateaudit.org
============================
DNorris,
Try:
http://www.leif.org/research/
http://www.leif.org/research/TSI-LEIF.pdf
The difference between the past and the presence can be attributed to CO2 - but this only amounts to 0.4 degC. This represents of CO2 sensitivity < 1 degC/doubling which is much smaller than the IPCC estimates.
Peter Hartley: Isn’t the hypothesis your describe is what Lindzen formulated and what he and Spencer have been discussing recently?
Was the warming of the late 20th century really that unusual?
I think we need to ask ourselves how much of the warming of the late 20th century was really that . . . real?
“Chapter 3 of the WG1 report begins:
Global mean surface temperatures have risen by 0.74°C ± 0.18°C when estimated by a linear trend over the last 100 years (1906-2005). The rate of warming over the last 50 years is almost double that over the last 100 years (0.13°C ± 0.03°C vs. 0.07°C ± 0.02°C per decade).”
Umm, that sounds like a convoluted way to say there was no warming in the first half of the century. 5 x 0.13 = 0.65, 10 x 0.07 = 0.70, so 0.05 for the first 50 years. Error ranges are left as an exercise for the reader. That clearly disagrees with the hadCRUT data below.
Anthony:
“I’m really beginning to see earth’s atmospheric processes more like that of an analog circuit with a variety of electrical components. There’s voltage, current, capacitance, reluctance and inductance in the irradiance-air-ocean systems.
I think an analog computer might very well model the earth’s atmosphere more accurately than a digital one. Digital signals do not exist in nature, but analog signals are abundant.”
Back in my EE systems class the instructor generally drew analogies to mechanical systems, e.g. springs, masses, dashpots. Digital computers replaced analog computers for very good reasons, but if you do make an analog analog, make it mechanical - far more photogenic and TV news media would be thrilled to air it.
Right after that semester the Club of Rome report came out with its two page system model. Then I read a SF book (Greybeard, by Brian Aldiss) that looked at England after an economic collapse. I was depressed for the rest of the month, but eventually figured out the Club’s predictive powers on resources and adaptations reflected at static world.
“Given that the temperature change has been about .7 rather than .3, I’ll wager it is a silicon rather than a germanium transistor.”
I got it, I got it! Boy, are you old. I might have a Germanium transistor or two in the basement….
I apologize for wandering OT, but following up on DNorris’ question:
In Dr. Svalgaard’s comments linked above, he states:
“Note, that Lean herself [with Wang, 2005] has published later reconstructions where the change since the MM was only 1 W/sqm and since 1900 only 0.5 W/sqm, effectively halving the increase you calculate.”
See the trace labeled as “Wang” in the plot. Although Dr. Svalgaard does not specifically state that this is the 2005 reconstruction, I’m pretty certain this is the implication. It much more closely matches those of the other solar scientists.
The AGW true believers will sieze on any error to discredit the arguments of skeptics so it is important we keep each other up to date. I appreciate your open mind in requesting futher information.
MattN,
On your question about why the cooling of the 1940’s to 1960’s didn’t take us back to where we were at the beginning of the 20th Century, wait for Part II. If you want to imagine what’s coming, ask yourself about what’s happened with solar activity during the 20th century. Figure 2 only shows part of the puzzle. None of what we are doing discounts the role of UHI, or crummy siting of surface stations, in possibly biasing the raw data we are working with so that the absolute values are higher than they would otherwise be at the end of the 20th century. But were focusing on something more fundamental.
I do hope people realize that what’s being plotted in Figure 2 are “rates of change” in temperature, not absolute values. Basically, Figure 2 plots the time derivatives, dx/dt, of the smoothed series from Figure 1. So Figure 2 is telling us that from one peak of a 22 year cycle to the next, dx/dt’s initially drop off, reach a nadir, and then begin to climb back up to the next peak. As to the “why” of that, again, wait for Part II.
Terry,
I’ve suggested the same thing — trying to get this published eventually in a peer reviewed journal. I’ve published in peer reviewed journals in the past, and have served as a referree — in economics, though, not climate science. Since we’re not in academia, there’s a certain bias that exists in the peer review process that we have to overcome. Given how politicized this issue has become, if it stands up to the kind of scrutiny we’re going to get from those who have stakes to protect, what we learn from that kind of scrutiny will help us to prepare something that is more likely to get published. We’ll just have to wait and see.
As to why we chose the particular filter we chose — that’s going to be part of the prejudice or bias we have to overcome, because it has its origins in economics. But time series are time series, and economics has learned a great deal about how to explore the properties of time series. When I started all of this, like some other economists, my first thought was to use some kind of ARMA (Box-Jenkins) analysis. But that would be treating temperature time series like a stock market trend where differencing is used to achieve stationarity and the data follows a random walk. That will not do here, because I think it is fundamentally wrong to believe that the processes we are investigating follow a random walk. It boils down to whether we expect dx/dt to be random or not. We don’t, not where “x” relates to physical processes like the influence of, say, the solar cycle on climate. The solar cycle shows a very well known pattern of variation in time. If it is influencing temperature, then it is going to show that influence through the dx/dt of temperature. HP filtering is an excellent tool for this. As to the role of lambda in HP filtering, we treat that in Part II.
Bob Tisdale,
I’m not sure of what you’ve done, but if you wait, then maybe you can put what you’ve done into a larger context. We’re just getting started.
I will tell you, though, that I don’t envisage anything in our Part II that might steal your thunder, whatever it is. We’ve already alluded to the role of terrestrial dynamics, where large scale, long lived oceanic oscillations seem to sit on top of Hale cycle periodicities, as it were, and modulate the influence of the solar cycle. I think the longer 65-70 year climate cycle is just that — the influence of terrestrial dynamics interacting with solar influences, to produce a complex system. Again, you’ll see more of where we are going with this in Part II, and then I’d be delighted to see what you’ve done with the PDO (and see how it compares to what we’ve done, but will not be specifically including in Part II).
Pamela Grey,
Why don’t you ask your question again, after we post Part II? You’ll should see something about how we would propose using HP smoothing to look for a relationship between temperature and sunspots.
Note to Anthony:
I’m talking about the chart I sent you this morning. While you work on finishing up the “science” part of Part II, I’m going to work on how to integrate that chart, and something else I’m thinking about, into Part II.
pablo an ex pat,
LOL!
Raven,
If I understand the science we’ll be presenting in Part II, it will not depend on the Sun’s radiative output, but how it gets modulated by something else.
As to your question about CO2, I don’t really question the science that says it should cause warming. As you intimate, the question is how much? I don’t think we are going to be able to show anything with an annual time series like we’re looking at here. To repeat what I said above, we’re basically exploring relationships in “dx/dt’s” There is a well known pattern to the dx/dt of CO2 emissions. But I wouldn’t be at all surprised that when we “drill down” and look at it this way, we will find the “dx/dt” of CO2 levels follows temperature, it doesn’t lead it. I’m pretty sure that’s already been explored. Look for studies dealing with “interannual” variations in CO2.
Basil
Jeff C. - Thanks… I will digest that tonight.
Raven - C02 Effect… Thanks. I am familiar with the work Lindzen and others on the decreasing effect of CO2 with increasing concentrations, but was actually looking for your Solar Output sources.
Onwards:
This post triggered a vague recall of a paper I read by Landscheidt. I finally found it at Still Waiting for Greenhouse.
SOLAR ACTIVITY:
A DOMINANT FACTOR IN CLIMATE DYNAMICS
RIP John and Theodor
Raven,
http://www.junkscience.com/Greenhouse/co2greenhouse-X2.png
I don’t think anyone can say exactly how little CO2 drove the temp. in the 20th century. Personally I don’t believe anyone who claims to.
Landscheidt made climate predictions based on solar activity, and some were right on the money (though he never told us about the ones that were wrong).
I have qualms about the direction this is all going. Like the CO2 kooks, we are trying to find one single, easy factor that drives the climate. I think it’s far more complex than that.
I will say, however, that the sun is at least 10 times a greater factor than CO2. Research dollars spent on solar research would certainly be a much wiser investment.
Using a cascading algorithm such as the one you used, it goes without saying that the results of the second step (Fig. 2) are very sensitive to the accuracy of the filter applied in the first step — i.e., how good it actually was at distinguishing the short-term “noise” from the actual trend. While that partially depends on the accuracy of the tau and lambda terms employed, it also depends upon two more assumptions which I think are far more critical: (a) there is only one trend, and (b) the “noise” (i.e., short term variations) is/are symmetric in their shape. To the extent that any of those things are in error, then the results of the second step will represent an accumulation of that error. And just an eye-ball examination of the first figure in your “The Solar to Global Warming Connection - A short essay” post suggests the assumption of symmetry is violated: the solar cycle variations are positively skewed (for laymen: the rising edge of each “hump” is steeper than the falling edge). Since you are attempting to imply something about solar cycles from temperature records, at the very least it seems to me encumbent upon you to demonstrate that the temerature records don’t violate those assumptions as well. If they do, then the HP filter is not the most appropriate one to use. A non-symmetric algorithm (e.g., a wavelet) is more appropriate.
Succeding in that, then there’s the even more critical assumption of a single trend. For that to be viable you’d have to assume solar irradiance is coupled with temperature through a single mechanism. In other words, you’d have to assume that land masses, oceans, and the atmosphere all react the same way, symmetrically, and on the same time scale — or at least that one of the three dominates to the extent that it obviates the others. IMO, that stretches credulity to the breaking point. Thus, I would argue that using an HP filter, combined with a cascading algorithm is not the way to go.
By the way, have you guys read Scafetta and West, 2007? It seems to me their approach is similar to yours (they take a heuristic approach), but vastly superior in a variety of ways.
REPLY: Rather than speculating on results you haven’t seen yet, may I suggest waiting for part 2 before claiming the methodology inferior? -Anthony
Paddy: Actually, I was talking about the Svensmark hypothesis — ie. that the sun’s magnetic field modulates the high energy cosmic ray flux reaching the lower atmosphere and hence the condensation nuclei for low level clouds over the oceans in particular. The Lindzen iris hypothesis as I understand it is a much shorter-term negative feedback mechanism. Essentially, storms in the tropics reduce the amount of water vapor in the stratosphere allowing more IR energy to escape and thus cooling the earth. The transistor analogy relates more to the Svensmark hypothesis as a mechanism for amplifying solar effects — the turning on and off of the cloud cover by fluctuations in the sun’s magnetic field strength modulates the flow of light from the sun like a small signal on a transistor gate modulates the much stronger current flow through the transistor.
“Research dollars spent on solar research would certainly be a much wiser investment.”
That’ll never do. We can’t sell solar offsets….
MattN: Here’s an illustration that may help. Solar is a pain unless you smooth it. While solar irradiance (TSI) leveled off and dropped a little mid-century, it didn’t fall back to the levels it was at in the early 1900s-late 1800s.
http://tinypic.com/fullsize.php?pic=5oa929&s=3&capwidth=false
Fascinating. Do you have a mechanism where by the solar cycles should drive the rate of change of global temperature? As far as I, maybe mistakenly, understand the cosmic ray theory is that the clouds they generate will modulate the equilibrium point of the earth’s energy balance, directly effecting the temperature, not its rate of change. But you show a good correlation with the rate of change.
Have you been able to isolate the components that are directly attributable to the small variations in TSI and the increase of CO2?
The results are interesting. I followed the procedure and made a spectral analysis of the differences time series (figure 2). There are two cycles, a low intensity one with a 14 year cycle and one with a 21 year cycle:
http://tinypic.com/view.php?pic=s2cxva&s=3
Two suggestions:
1. For the next post try to make a spectral analysis of the signal
2. Try to apply the same procedure for the temperature of the southern hemisphere
Rico,
Ditto what Anthony said. Yes, I’ve certainly read Scafetti and West. And it is an elegant piece of work. But it doesn’t show what we’re showing. For all their effort, the most they can say about the 20th century is:
“During the 20th century one continues to observe a significant correlation between the solar and temperature patterns: both records show an increase from 1900 to 1950, a decrease from 1950 to 1970, and again an increase from1970 to 2000.”
We’ve shown something entirely different, and potentially more significant: that the rate of change in a smoothed global temperature series follows a pattern that tracks with almost pinpoint accuracy the 22 year Hale cycle. They claim a broad correspondence to three periods in the 20th century. Figure 2 shows close correspondence to six Hale cycles over the past 130 years.
And if you think that is something — I certainly do — wait until Part II.
Basil
I think CO2 levels are a cyclic function of ocean absorption. The ocean is the biggest sponge of CO2. But it only acts as a sponge under certain conditions that appear to be tied to cycles. The CO2 that is absorbed than gets deposited in deep ocean bottom material. It does not cough it back up. Then, in the non-absorbing part of the cycle, it stops soaking up CO2, leaving it to rise into the upper stratum of the atmosphere to get measured by alarmists. When the time comes for the ocean to once again be receptive to CO2, the process begins all over again. We should be entering an ocean absorbing phase right about now, if it hasn’t already begun. This long cycle is just a theory since we haven’t been measuring CO2 long enough to discover a pattern. CO2 probably does serve to warm us up a bit but we should be considering the possibility that it is also cyclic in nature.
Anthony: Rather than speculating on results you haven’t seen yet, may I suggest waiting for part 2 before claiming the methodology inferior?
I didn’t speculate on much of anything. I commented on what you wrote, and on the logic contained in it. The assumptions I questioned were your own, and by extention those contained in your method. That’s not speculation or opinion, it’s logic.
I did, indicate that your assumption that solar irradiance is coupled with temperature through a single mechanism (which is what the assumption of a single trend requires) strains credulity. And I suppose that could be considered opinion. Nonetheless, the assumption itself does logically require that land masses, oceans, and the atmosphere all react the same way to irradiative forcings. And that would be an extraordinary claim. As such it requires extraordinary evidence. And you didn’t present any.
REPLY: “And you didn’t present any.” Well then, again I suggest kindly stay tuned for part 2.
RE: An historic and stable value for TSI and Svalgaard.
I have high regard for Dr. Svalgaard’s encyclopaedic grasp of all details solar, but I cannot condone regarding his opinions and theories as mainstream. Try his old collaborators, Schatten and Hoyt if orthodox belief is your central concern.
JM,
Interesting. What do you see spectral analysis telling us that we’re not seeing with the use of HP smoothing and first order differencing?
I have nothing against other tools and techniques. We often come at the same thing by different means. It doesn’t mean necessarily that one is better than the other. But unless you can tell me what spectral analysis will show me that I’m not already seeing, I already have a full plate of things to do.
The suggestion to look at the Southern continent is a good one. But don’t look for it in Part II. We are just breaking some ground here, and don’t expect these initial posts to be the final word. More like a quick introduction to a new way of looking at things.
But I do appreciate the feedback. I think the technique is one that has broad application in climate science, where we have historical time series data. It sounds like Bob Tisdale may be using it to look at the PDO.
Looking at the global temperature trend is just a start, and for very little effort, is producing some interesting results. Applying it to the SH is something to do, at some point.
Basil
REPLY: I agree that the southern hemispshere might be interesting to look at separately. I think what JM is getting at is that a spectral analysis would show which cycle or multiple thereof is the dominate peak. I certainly have nothing against the idea. Perhaps when we get a variety of feedbacks from this two part presentation, we can look further. - Anthony
Now that the global warming community has bullied Judith Lean into reconstructing her solar reconstructions at least three times now and the latest numbers show hardly any solar variation whatsoever, it is clear that the global warming community is going to re-write every historical record there is until only CO2 matters.
Effectively, solar irradiance is useless now for any kind of analysis since it hardly varies at all. All that can be used is “sunspot number” which obviously does not account for all the variation in solar output across the entire EM spectrum.
Its a good thing we have Roy Spencer and John Christy to keep the land-based temperature records honest going forward (not the historical ones however) but they could be silenced at any time as well.
Rico,
If I may, let me back up a bit ask something for clarification. You say:
“Succeding in that, then there’s the even more critical assumption of a single trend. For that to be viable you’d have to assume solar irradiance is coupled with temperature through a single mechanism.”
I do not follow you on this. Any single trend can be the result of multiple independent variables or influences. Personally, based on what we will present in Part II, I’m not sure that we should have brought solar irradiance into the discussion at all. Actually, I don’t think we have to show at all what the physical mechanism is that is involved in the connection between the Sun and globally averaged temperature to justify the title of this series. To the extent that we can shed any light on the physical mechanism involved, that’s serendipity. And that’s what has been saved for Part II.
For now, how has what we’ve done any different, say, than paleoclimate studies of tree rings that show through MTM or spectral analysis evidence of a 22 year periodicity in the data? When such studies are reported, don’t they just usually claim that this is indication of a solar influence? Would a referee reviewing such a paper reject it because it doesn’t contain a theory about the solar physics involved?
Basil
Anthony: “And you didn’t present any.” Well then, again I suggest kindly stay tuned for part 2.
Okay, I’ll wait for part 2. Until then I will simply speculate (lol!) that you will (a) better explain the values you used for tau and lambda, (b) provide a response profile of how the resulting filter applies to real life time scales (or at least provide the numbers that would allow someone else to do it), and (c) attempt to quantify the apparent forcing relationship between the Hale cycle and the temperature response (including the necessary logic to justify yourselves). It seems to me that if you don’t do that, then merely showing some sort of a qualitative relationship between the Hale cycle and temperature — even if “tight” — wouldn’t be exactly surprising.
Further, if you’re seriously contemplating submitting your findings to a real, honest to goodness peer reviewed journal, then it seems to me you are at the very least compelled to discuss them in light of Scafetta and West, along with others who have noted an apparent secular relationship between one or another (or more) frequency of TSI oscillation and surface temperature (e.g., Eddy, 1976; Lassen and Friis-Christensen, 1995; Lean et al., 1995; Crowley and Kim, 1996; Hoyt and Schatten, 1997; White et al., 1997, 2003). But I assume you already know that — assuming your intent is to seriously submit your study for peer review. If not, then I guess anything goes.
REPLY: See Basils response.
««Interesting. What do you see spectral analysis telling us that we’re not seeing with the use of HP smoothing and first order differencing?»»
Basil,
Spectral analysis is a more objective method to find cycles in the time series. It allows the detection of cycles with frequencies that may be obscured by other cycles.
You state:
and
yet the graph in figure 2 does not show a peaks at 1926 and 1951, but 1936/1937 and 1958/1957
Could you clear this up, please?
Thanks
anomdebus,
That statement is in reference to the AMO, not temperature. We didn’t present a chart showing those numbers. In the section you are reading from, we are comparing what we’ve found with what others have found indicating a long ~66 year climate cycle.
Does that clear it up?
Basil
Basil,
Yes that does, thank you. I think I was letting my primate pattern making skills override my reading skills.
Thanks again.
Thanks for your very fine work. You have been putting a lot of work into finding the truth, and the lies, and all of us appreciate it.
On catholicfundamentalism.com there are frequent references to Global Warming, not because it’s real, but because those who tell lies for money, position, and power are condemning their souls to perdition.
There is a spiritual component to this, and we can’t forget to pray for those who lack the strength or morals to seek the truth.
Hi,
Basil what TSI source are you using? Hopefully not Hoyt. I think there are many current TSI sources available but Hoyt seems not to fit the current trend. Very nice piece of work though. I have also found that there might be a link to magnetism meaning that the earths magnetic field concentrates the PDE which breaks up the cloud forming nuclei when the TSI peaks. See the RSS map and magnetic field map. Very interesting and still very much work in progress.
http://www-atlas.usgs.gov/articles/geology/a_geomag.html#one
http://www.remss.com/msu/msu_data_monthly.html?channel=tlt
Basil, Anthony:
This is a great piece of work. I had just started playing around with taking the first derivative of smoothed time series, but had not explored filtered series as of yet. This will send me into whole other directions. But I want to see part II first!
Rico:
“indicate that your assumption that solar irradiance is coupled with temperature through a single mechanism (which is what the assumption of a single trend requires)…
i.e., how good it actually was at distinguishing the short-term “noise” from the actual trend.”
I don’t believe this analysis is at all cogent. If the purpose were to look at the incoming TSI and the PDO/AMO (or other/additional oscillations), as a TSI history, in a component analysis with the end being to create a polynomial fit, it would, but that is not here entertained. Indeed, a differential equation seems a better goal altogether than the one you envision.
Pattern recognition appears to be the sole objective here.
REPLY: “Pattern recognition appears to be the sole objective here”
That’s the best description one could make for what this is about. Find the pattern present first, detailed analysis comes later. - Anthony
Fascinating! You are skilled at what you do, guys! I’m currently revisting older work on solar right now. This new finding supports what I have generally found based on that work. Can’t wait for the conclusion!
Maybe I can shed some light on finding cycles within what appears to be steady state noise. In my research, synaptic brain response to auditory signals as measured by electrode pickup on the scalp can only be found by filtering out the background static noise of the brain. Our brains fire all the time in a fairly random and steady state when we are just chillin, as in not thinking or listening to anything (that’s how we know we are not brain-dead). This noise can be mathematically reduced (filtered out) to a narrow band of “zero” and even smoothed out to near zero (in simple terms adding negatives and subtracting positives so that random becomes zero), leaving space above and below for patterned or cyclic responses, if they occur. Now we add a series of clicks or pips to the ears, and presto: out comes a series of waves (measured electrical synaptic jolts) that can be measured for slope, peak amplitude, and time, as well as corresponded to the major synaptic junctions of the auditory neural pathway. We can even determine the frequency of the tone pip that was sent down the pathway. Higher frequencies result in earlier wave peaks, lower frequencies result in later wave peaks. Until filtering was used, the very small but PATTERNED auditory neural pathway responses hid behind a veil of random static noise. You just never know what you will find when you filter out random stuff. Gold miners used basically the same principal by filtering and washing out dirt to get to the nuggets.
Bill Illis (13:33:12) : says:
“Now that the global warming community has bullied Judith Lean into reconstructing her solar reconstructions at least three times now and the latest numbers show hardly any solar variation whatsoever, it is clear that the global warming community is going to re-write every historical record there is until only CO2 matters.”
There are good scientific reasons for the change that have nothing to do with climate science. In fact, the new solar data actually problems for the warmers because now they cannot argue that the LIA and MWP were caused by changes in the sun that are not occurring today. IOW - warmers cannot argue that CO2 is the only plausible explaination for the warming today if they don’t know why warming occurred in the past.
“IOW - warmers cannot argue that CO2 is the only plausible explaination for the warming today if they don’t know why warming occurred in the past.”
They used the history-erasing hockeystick to essentially eliminate all vaguely recent historical warming.
Basil .. Anthony …
I’ve seen this before!!!
When you mentioned a cycle of 65-70 years, Landscheidts Big Fingers immediately came to mind! I went back to John Daly’s site and re-read his article, and sure enough .. there it is.
“Cycles of big fingers have a mean length of 35.8 years (178.8 years [big hand] / 5 = 35.76 years [big fingers]). They are closely connected with solar activity. They coincide with maxima and minima in the Gleissberg cycle and open up the possibility of predicting these crucial phases many years ahead [62, 63]. As will be shown below, they also define the length of the 22.1-year magnetic cycle of sunspot activity (Hale cycle). ”
What was really interesting is the link between the 60-70 year cycle and an influence on the Hale Cycle. …. hmmmmmmm
I know some people think Landscheidt was a kook because he played with Astrology .. but I can’t help but wonder, why is it that I keep seeing other research unknowingly confirm what he did.
I am inclined to suspect that the recent stabilization in global temperatures may well merely reflect the launch of the Aqua satellite, launched 2002.
What happens is that whenever the data looks unfavorable to anthropogenic global warming, people cast about until they find a possible source of error. If you look hard enough, you can always find a *possible* source of error. Then they pull an error term out of their asses, and “correct” the data till it is politically correct. Eventually someone gets around to making a measurement that is proof against this potential source of error. (Aqua’s orbit does not drift) “Corrections” of older data sources then rapidly converge to agree this new more error resistant source of data - but the old “corrections” continue to be applied to historical data.
Satellites directly measure global temperatures against an absolute standard. The last remaining source of wiggle room was orbit drift, and Aqua does not drift, so since 2002, and *only* since 2002, we have unchanging thermometer in the sky measuring the whole globe, except for the poles.
Actually, more sunshine has been reaching the earth since the early 1990’s.
“Variations in solar radiation incident at Earth’s surface profoundly affect the human and terrestrial environment. A decline in solar radiation at land surfaces has become apparent in many observational records up to 1990, a phenomenon known as global dimming. Newly available surface observations from 1990 to the present, primarily from the Northern Hemisphere, show that the dimming did not persist into the 1990s. Instead, a widespread brightening has been observed since the late 1980s. This reversal is reconcilable with changes in cloudiness and atmospheric transmission and may substantially affect surface climate, the hydrological cycle, glaciers, and ecosystems. ”
http://www.sciencemag.org/cgi/content/abstract/308/5723/847
Anthony / Basil
Just FYI, the authors of this paper did something similar with the Chinese temperature records and also found cycles of similar lengths. So there is some support out there…
http://www.crikey.com.au/Media/docs/Zhen-Shan–Xiuan-MeteorAtmosPhys-2007-d1227bc1-3183-456f-a935-69c263af1904.pdf
It did not pass unnoticed as WC had a brief look at it on Stoat, and by the looks of it, Lubos also had something on it:
http://scienceblogs.com/stoat/2007/08/multiscale_analysis_of_global.php
cheers
Arnost
By the way, there is a difference between TSI and solar radiation reaching the earth.
http://www.timesonline.co.uk/tol/news/uk/article696586.ece
Consider the UK in 2006:
http://www.metoffice.gov.uk/climate/uk/2006/sunshine.html
As much as 70% above normal in 2006.
Or, long term:
http://www.metoffice.gov.uk/climate/uk/about/UK_climate_trends.pdf
“Table 12 shows the percentage change in sunshine, based on a linear trend starting from 1929. It shows that the greatest and most significant changes occurred in the winter season, when there has been an increase in sunshine of about 20% for central and northern England.
Sunshine has also increased in these areas by about 10% in autumn, and by 8% over the year as a whole for eastern and NE England. These increases could be a result of the Clean Air Acts of 1956 onwards, which has led to a decrease in air pollution.”
Cleaning up of air pollution …
Jim Arndt,
This discussion took a wrong turn somewhere. We’re not looking at irradiance. There’s casual mention of it by Anthony in a general remark at the beginning of the discussion, but if that’s the source, it reads something into what Anthony said that I don’t think he meant. Some other commenters have gone off on their own tangent about irradiance and what the data is for it, and so forth, but we’re not going to infer anything about irradiance that I know of. In fact, in an early reply in the discussion I put a little
after a remark about infering a rather different physical basis for what we think our findings may be showing.
Rico,
We’ll explain more about lambda in Part II. But why do you ask about tau? What is your understanding of the role it plays in HP filtering?
I understand well what is required to pass peer review, and we’re only just getting started on this. I can see where this work needs to interact with some of the references you cite, but not others. And I have a long bibliography I’m compiling of numerous others. Which brings me back to “c” in your list of expectations for Part II. What sources might I have overlooked that show relationships between Hale cycles and temperature trends? Even if there are others, are you saying that corroborating work that may only advance the state of knowledge or understanding incrementally is a waste of time? Actually, based on what we’ll present in Part II, I think what we are doing goes beyond that, but I don’t really understand the basis for your skepticism.
Basil
REPLY: I never meant to imply irradiance was used in this essay. I used the term in talking about how I think the climate system can look much like an electrical circuit, with irradiance being part of that, voltage for example…and that all came from thinking about bandpass filters. I’m an amatuer radio operator so I tend to think along tose lines. - Anthony
Hi,
Basil and Anthony, Thank you for making that clear. Many people use the Hoyt and it not in line with the current thinking in TSI reconstructions. Just wanted to make sure. Keep it up and I am waiting in earnest for part II.
Anthony,
Our friends at NCDC shut off the managing parties field again. They have also taken down the obstruction and exposure information.
REPLY: I’ll look into it.
Related to the latest ice breakage horor story. A puzzle for number phobic environuts. According to
wikipedia
the Ross Ice Shelf moves 1.5 to 3 *meters* a day. Given that rate calculate amount of time it would take to reach New Zealand *if* there is no iceberg breakages.
Pamela Gray: “Then, in the non-absorbing part of the cycle, it stops soaking up CO2, leaving it to rise into the upper stratum of the atmosphere to get measured by alarmists.”
Beautiful line! m’am.
You mean the LIA and MWP they say didn’t happen?
Basil, the spectral analysis also shows the exact amplitude of each frequency component which you won;t get from a time domain analysis.
BTW I recently favor GaAs transistors
Basil (14:13:53): I do not follow you on this. Any single trend can be the result of multiple independent variables or influences.
It can, but it’s not a necessity. If nothing is known about the nature of the data it could just as easily be the opposite: multiple independent variables or influences could result in different trends. But by using an HP filter you are imposing the single trend assumption when in fact it might not be valid. In math speak, the HP algorithm assumes the signal upon which it is applied is composed of two distinct frequencies whose periodicities are “sufficiently distinguishable” from each other. The filter is highly dependent upon that assumption. That might be appropriate for economic data, but I doubt it’s appropriate for climate data. But if you think it is, then it seems like you should argue your case in some detail. Moreover, it appears to me the results could be highly dependent upon the value of lambda, which essentially serves as the cut-off frequency. In both those regards, I think you should read this paper if you haven’t already.
And perhaps the essential question is: how dependent are the conclusions you make after the second step in your analysis on the lambda value you use in the first step? If they aren’t, then it seems to me you will have to defend your use of a particular value on a technical level. And if you can’t, then I think it could be rightly argued that you are attempting to interpret artifact. But I’ll wait for what you say in Part 2.
As for your later question about the tau (19:42:02), I withdraw my statement. Upon first glimpse of the formula for the HP filter I misinterpreted its meaning. I think I have a better handle on it now. Regarding the question of corroboration, my point was just that you have to fit your results in with those of others that have attempted to analyze the same or similar phenomena. But it sounds like you have a good handle on that.