the Air Vent

Because the world needs another opinion

Cycles In Temperature Trends

Posted by Jeff Id on October 15, 2011

Recently, a paper by Craig Loehle and Nicola Scafetta has come to my attention.   The simplicity of the math makes it attainable for the technical reader who doesn’t want to spend a month learning climate jargon.  This isn’t a Mannian stew.  The paper received a strong critique at WUWT, much of which was undeserved in my opinion.   What the authors did was recognize the widely discussed 60ish year signal in climate data and project it onto temperature records to determine the percent of recent trends are due to human contribution.   The problem is that measured temperatures are 150 years long so our recently invented thermometers (think about that sometimes folks) have only recorded two of the suspected 60 year cycles and those barely visible cycles are further masked by a  warming trend.   So the question becomes, are the 60 year climate cycles regular or are they not?  We don’t have proof of anything but if they are real, there is some additional realistic evidence of lower climate sensitivity to CO2.  A smaller amplitude high frequency component was also used in the work.

The abstract is a good place to start, and may have caused much of the problem in interpretation.

Abstract: The climate change attribution problem is addressed using empirical decomposition. Cycles in solar motion and
activity of 60 and 20 years were used to develop an empirical model of Earth temperature variations. The model was fit to
the Hadley global temperature data up to 1950 (time period before anthropogenic emissions became the dominant forcing
mechanism), and then extrapolated from 1951 to 2010. The residuals showed an approximate linear upward trend of about
0.66°C/century from 1942 to 2010. Herein we assume that this residual upward warming has been mostly induced by
anthropogenic emissions, urbanization and land use change. The warming observed before 1942 is relatively small and is
assumed to have been mostly naturally induced. The resulting full natural plus anthropogenic model fits the entire 160
year record very well. Residual analysis does not provide any evidence for a substantial cooling effect due to sulfate
aerosols from 1940 to 1970. The cooling observed during that period may be due to a natural 60-year cycle, which is
visible in the global temperature since 1850 and has been observed also in numerous multisecular climatic records. New
solar activity proxy models are developed that suggest a mechanism for both the 60-year climate cycle and a portion of
the long-term warming trend. Our results suggest that because current models underestimate the strength of natural
multidecadal cycles in the temperature records, the anthropogenic contribution to climate change since 1850 should be
less than half of that previously claimed by the IPCC. About 60% of the warming observed from 1970 to 2000 was very
likely caused by the above natural 60-year climatic cycle during its warming phase. A 21st Century forecast suggests that
climate may remain approximately steady until 2030-2040, and may at most warm 0.5-1.0°C by 2100 at the estimated
0.66°C/century anthropogenic warming rate, which is about 3.5 times smaller than the average 2.3°C/century
anthropogenic warming rate projected by the IPCC up to the first decades of the 21st century. However, additional
multisecular natural cycles may cool the climate further.

So if we have a global temperature record which looks like this:

The top pane is hadley temperature with an overlayed model, the bottom is the residual from a full length model fit showing a cyclic plus trend method.  Since there is long term signal still visible (kind of a U shape in the bottom data), the authors determined the model is absolutely misspecified.   The thick black line over the top temperature pane doesn’t seem entirely unreasonable though.  It is quite possible that there is a cyclic nature to the signal, but without more information, little can be concluded with certainty.

This is the comment Leif Svalgard left at WUWT:

The paper is cyclomania at its worst

A comment which I believe is way over the top.   Perhaps the conclusions are too strongly worded but the concept behind the paper is not bad, it actually could be perfectly correct.  Willis made more complete arguments with other fits having different timescales.  Nicola Scafetta replied with this comment which will sum up my own understanding.

Dear Willis Eschenbach,

I am sorry but you continue to say no-sense.

1) you cannot disprove our research by simply using a model slightly different from our, for example using a 62-year cycle instead of a 60-year cycle, and claim we are wrong because you get similar results! Why don’t you use a 100-year cycle instead of a 62-year cycle?

2) the existence of major quasi 60-year and quasi 20-year cycles are easily detected by power spectrum analysis, see figure 3, 10 & 11 in N. Scafetta, “Empirical evidence for a celestial origin of the climate oscillations and its implications”. Journal of Atmospheric and Solar-Terrestrial Physics 72, 951–970 (2010), doi:10.1016/j.jastp.2010.04.015

The bold is mine.  The paper by Loehle and Scafetta isn’t disproven by the mere mention that different cycles can be fit to the same data.   If the cyclomania is even close to a real process – say some cycles are 40 and some are 70 years, the fits address a substantial fraction of the climate warming we’ve seen.   Solar cycles experience some of the same imperfect variance as Leif will attest to himself.   Where the critics went really wrong was to fail to recognize the proxy data shown in the same paper, having a longer timeframe and similar cycles.

See, the 60 year cycle may not be only real since 1850.  It may not be a perfect cycle, but like sunspots, ice ages, and many other climate phenomena, there is potentially a generally repeated nature to the variance.  All that is required is for the cycle to exist and Loehle and Scafetta have a reasonable way to estimate the percentage of warming caused by CO2.

The last paragraph of the conclusion may be where they have gotten themselves in trouble.

It is also worth noting that by starting in a cycle trough in
1900 and ending on a cycle peak in 2000, graphical
depictions of warming in IPCC and other reports give an
exaggerated depiction of the rate of warming, a point noted
by Karlén [75] and Soon et al. [76]. By further failing to
mention solar effects and warm bias in the instrumental
record, it is implied that almost all the warming of about
0.8°C over this period is human caused, when really no more
than 0.4°C is likely to be. Reporting such a lower number
would probably cause much less alarm than what was
recently claimed by Rockström et al. [77]. Moreover,
because the 60-year natural cycle will be in its cooling phase
for the next 20 years, global temperatures will probably not
increase for the next few decades in spite of the important
role of human emissions (see Fig. 5A), as predicted by
multiple studies (reviewed in Loehle [69]; Scafetta [18]).

While the methods seem fine to me, they are far from definitive.   The paper did achieve an excellent fit to the data based on reasonable assumptions of observed natural cycles, like predicting the next sunspots, the total variation in the natural cycles is basically an unknown.   I would have prefered a less definitive conclusion on the order of, “if the previously observed cycles are an accurate component of recently observed temperatures and continue for the near future then we can make the following conclusions:”   Although similar statements are made in the paper, without more of that, a zealous reader can easily go too far critiquing the work.   The proxies are a longer record and do seem to support the cycles.

I don’t think that anything in the work is unreasonable though and the critiques of it in the WUWT thread were too severe IMO.  It is yet another paper which lends support to the ever more likely fact that climate models are running too hot and the potential that AGW is a far less severe effect than the IPCC likes to state.

61 Responses to “Cycles In Temperature Trends”

  1. Sean Houlihane said

    I have sympathy for most of the views which you refer to here. It is surely right to say that just because there is a fit to a periodic model doesn’t mean the model is in any way useful – but also interesting to note that a good fit can be made with a causal connection being physically possible. It seems to be a paper which doesn’t really lead anywhere though. Even if the models are running warm, and the paper encourages investigation of that, what would be the approach to test this assumption?

  2. Craig Loehle said

    Jeff, Thanks for the defense of the paper. I really do suggest people read it instead of giving a gut reaction. It is open access so free to download.

    The purpose of the paper was 1) to show that part of the warming since 1970 is natural, and therefore the GCMs are calibrated too hot, and 2) to attempt to detect the anthropogenic signal, which we think we did. If the cycles are really 62 yr this does not affect these conclusions. A phenomenological model suggests mechanisms to look for. Many aspects of science are “understood” only in terms of such description, without a mechanism yet being worked out.

  3. DeWitt Payne said

    It doesn’t matter if the cycles continue into the future. The really important point of the paper is that it is yet more evidence that the aerosol effect has been grossly exaggerated in the climate models. That alone is sufficient to reduce the estimated climate sensitivity to doubling CO2 to the lower end of the IPCC range.

  4. toto said

    The paper by Loehle and Scafetta isn’t disproven by the mere mention that different cycles can be fit to the same data.

    If there are enough ways to do it, then yes it is. That’s the concept of a p-value.

  5. DeWitt Payne said

    The other question is: why use a linear trend starting in 1942 for anthropogenic forcing? There is no reason to believe it’s linear and that it doesn’t start until 1942.

  6. Craig, I am not a climatologist but I have worked in space science for 50+ years.

    I suspect a great deal of the criticism is coming from those who have blindly endorsed the absolutely absurd “model” prediction that Earth’s climate is independent of Earth’s heat source – the Sun.

    As noted Professor Curry’s weekly review:

    A few old white males meeting at the Bilderberg probably generated the climate models, cosmology models, economic models, solar models, population models, etc that were worshipped by consensus scientists on government grants.

    That is the base of society’s problems today.

    A few scientists are waking up to the manipulation of science that occurred as the US space program was dismantled and US leadership in physical sciences faded:

    With kind regards,
    Oliver K. Manuel
    Former NASA Principal
    Investigator for Apollo

  7. Jeff Id said

    #4 Toto, The mere mention is not enough but a test may be. I wonder if you can suggest a proper test of significance? I’ve put a little thought into how you would statistically demonstrate that the 60 year cycle didn’t exist, or the odds that an existing cycle suddenly stopped, increased, decreased. It is a little beyond me at this point so any light you can shed on it would be interesting.

  8. Jeff Id said

    DeWitt, I thought the same thing. Shouldn’t we see some sort of capacitive charge curve as temperatures build? My eyes told me it might be there but it was a shoulder shrug moment.

  9. Jeff,
    thank you for your input and balanced comment.

    As Loehle said, people needs to read our paper before criticizing it.

    Criticisms such as those by Willis Eschenbach and Leif Svalgaard, and others, prove nothing but their own sophism. It is so evidently unfair and wrong that I believe their purpose is only to mislead people.

    Of course we never claimed that with our paper all issues are solved. However, our phenomenological model agrees with the data much better than what any GCM off the IPCC does, and suggests that important mechanisms and forcing are missing in those models. Science progress in this way, one step by time.

    Our poor Leif accussed our paper of “cyclomania at its worst”. Evidently Leif does not know that when Schwabe, who was not a professional astrophysicist, proposed that the Sun was oscillating with an 11-year cycle, he did that using only 17 years of careful observations, that is 1.5 cycles. Fortunately, Wolf, who was a professional and very smart astrophysicist, did not reason like our Leif, and understood the enormous importance of the finding.

    As you say one of the major issues that the critics have not acknowledged is the fact that a quasi 60-year cycle has been found by numerous scientists in numerous climatic records covering also several centuries and millennia. Our figure 4 just shows two records showing this cycle during the last three centuries. However, our paper contain numerous addition references where cycles ranging from 50 to 70 years have been detected. Moreover, in my paper “Empirical evidence for a celestial origin of the climate oscillations and its implications” I show that this cycle is a regular major astronomical cycle of the solar system that evidently is there since the solar system formation 4.5 billion years ago.

    People just need to read carefully our papers together with their references.

    A. Mazzarella and N. Scafetta, “Evidences for a quasi 60-year North Atlantic Oscillation since 1700 and its meaning for global climate change,” Theor. Appl. Climatol., DOI 10.1007/s00704-011-0499-4 (2011).

    C. Loehle and N. Scafetta, “Climate Change Attribution Using Empirical Decomposition of Climatic Data,” The Open Atmospheric Science Journal, 5, 74-86 (2011).

    N. Scafetta, “Empirical evidence for a celestial origin of the climate oscillations and its implications”. Journal of Atmospheric and Solar-Terrestrial Physics 72, 951–970 (2010).

    links of the papers are in my web-site

    Other papers that will go deeper into the issue will be published.

  10. Jeff Id said

    The evidence in the links Nicola provided does present a much more solid foundation for the paper’s conclusions.

  11. Craig Loehle said

    Toto: if different cycles can fit the same data, this would still show that the climate models are overfit (exaggerate how much warming in recent decades is human-caused).
    DeWitt Payne: Why linear and why 1942? If you increase GHG exponentially, the log-saturation of their effect gives a linear relation, as we stated in the paper. Outputs of the GCMs are mostly linear (though changing sulfate forcings distort this a little). 1942 was determined strictly from the residuals of the data from the first fit.
    As Nicola said above, we do not claim our model solves all problems but I think it is sufficient to show that the signal after subtracting reasonable estimate of natural cycles is not cause for alarm.

  12. gallopingcamel said

    The works of the graphologists and the climate modelers are interesting but useless if your purpose is to inform public policy.

    These approaches will remain useless until they become sophisticated enough to make correct forecasts of future climate.

    Many of the people wasting our time and money with such nonsense read Isaac Asimov’s “Foundation Trilogy” at an impressionable age. Now they imagine themselves as brilliant as Hari Seldon.

  13. steve fitzpatrick said

    There does appear to be a clear 60+/- year cycle of about +/-0.1C amplitude; the evidence of the 20 year cyclical component is probably not as strong, but it may also be present. The only real critique I have is the apparent disconnect from radiative forcing. Any secular trend ought to track radiative forcing better than time, especially considering the modest rate of ocean heat uptake. If you compare the temperature history, with a best-estimate of the +/- ~0.1C oscillation removed, then the trend is remarkably linear with respect to GHG forcing. Forcing just makes more sense than time.

    The CO2 history shows a gradual increase above the background (ice cores) starting in the late 1800’s; that would seem to imply forcing started to increase at about the same time; there is nothing magical about 1950. (The adjustment in Hadley SST’s after 1945 seems to make the correlation against forcing even stronger.)

    Further, the linearity of temperature vs forcing constrains aerosol offset to a low level: if aerosol effects are 1) large, and 2) non-linear with emissions (both of which are demanded by a hypothetical high sensitivity), then the temperature history ought not be proportion to forcing… yet is clearly is.

  14. Craig Loehle said

    Steve: compare the combined solar forcing model in the appendix and toward the end of the paper.

  15. Craig Loehle said

    Steve: also, it is possible that any warming pre-1950 is not separable (using our data and methods) from the long-term warming recovery from the little ice age.

  16. I did some googling, and discovered that there is also a 60 year cycle in interest rates. They are about to turn the corner, which may be useful information. Here’s the evidence!

    Maybe it’s also Jupiter.

  17. Yes, Nick Stokes,

    the 60-year economical cycle is well known since 1925.

    These economical cycles are very likely driven by climatic cycles which are originally driven by the Jupiter/Saturn 60-year oscillation that makes the entire solar system (including the Sun and the Earth) to oscillate in the same way.

    It is evident to me that the theory we are proposing fits a lot of multidisciplinary observations while the IPCC AGW GCMs do not explain anything in detail.

  18. DeWitt Payne said

    Re: Craig Loehle (Oct 15 11:26),

    If you plot the well-mixed ghg forcing data from GISS (column 2 only), it’s hardly linear and doesn’t start in 1942 either. See graph here. That’s what I would use as a proxy for anthropogenic forcing rather than a linear model. Any bets on whether the residuals would look better?

  19. Craig Loehle said

    DeWitt Payne: Most of the forcing is CO2, and it is not the forcings that are linear, it is the log of the forcings. What you showed is approximately exponential so the log of it is approx linear.

  20. Craig Loehle said

    DeWitt: also, please do not assume that the GISS forcings are correct–they have huge uncertainties about many of the components besides CO2

  21. Craig Loehle said

    I believe I may have misunderstood DeWitt’s graph. It is a graph of forcings which should be the result of the log of CO2 concentration–I can’t vouch for anything GISS and Hansen do, since they impose a sulfate aerosol cooling in the early part of the warming period which is unsupported by the data, in order to better match the historical record. They are addicted to aerosols in spite of the different models using different aerosol forcing histories and in spite of the data being made up basicly. Compare the forcing history implied by the solar reconstruction in our paper.

  22. Bill Illis said

    I’m clearly in the camp that there are 60 year cycles in the climate (which might not be quite as regular as 60 years all the time but nonetheless).

    And the comments above about Aerosols and the fact that Hansen has updated his Radiative Forcing data to 2010 has naturally persuaded me to chart the data up.

    Total Aerosols impact first. Some funny made-up straight lines and Hansen has actually increased the total Aerosols Forcing (from -1.8 W/m2 in 2003 previously to -2.2 W/m2 in 2003 in the current model, which declines even further to -2.4 W/m2 by 2010 – Asian aerosols I guess since the NH numbers are down considerably in the last two decades).

    Aerosols offset all the GHG forcing until about 1970 (obviously the warming from 1750 to 1971 was caused by Black Carbon and Solar forcing).

    And then all the Forcings (with Other being mainly Black Carbon and Solar).

    So, the climate is cyclical and/or driven by “forcings”. The forcings explanation, however, requires one to suspend rational judgement and just keep fitting assumptions until the math works. Not exactly a robust theory.

  23. Craig Loehle said

    Thanks Bill.
    Citation for how dodgy the aerosol numbers are:
    Schiermeier Q. The real holes in climate science. Nature 2010; 463: 284-287
    There is also a paper or blog post showing that the GCMs with the highest sensitivity (like Hansen’s) have the highest aerosol forcing to keep the model from warming too much vs data. very suspicious.
    Just looking at CO2, which is actually measured, it rises approx exponentially, and the log of exponential is a straight line.

  24. HaroldW said

    DeWitt –
    Why column 2 only? From the same source, total forcing (see graph here) is much less, and is a net zero as late as 1970. I’m not trying to vouch for this dataset, just saying that one should include all anthropogenic effects if one is trying to isolate natural cycles.

  25. Craig Loehle said

    HaroldW–I also don’t trust GISS “sum of all forcings” where they do funny things with volcanic forcings, plus this includes solar, which I am modeling separately, but note that after 1960 the blue line (GHG) is nearly linear.

  26. steve fitzpatrick said

    The first four columns of data are probably reasonably accurate…. the rest, well, not so much (more like speculation). The combined first four against temperature (after removing a +/- 0.1C cycle) correlate pretty well with temperature.

    I think that Wang et al is probably the more reliable estimate of solar variation. In any case, the correlation should be with total forcing; solar changes plus all GHG’s, not just solar.

  27. Craig Loehle said

    Steve, My point is that our model is decomposing the signal, and people are comparing the linear human impact portion to the GISS or other total forcings, which includes the solar we are modeling separately. I also think Hansen exaggerates volcanic forcings effects (it helps him get temps right that way).

  28. Carrick said


    If there are enough ways to do it, then yes it is. That’s the concept of a p-value.

    Agreed. Looking at atmospheric-ocean oscillation frequencies is of the few places I think paleo data are really useful for.

    See for example this.

    I agree that three cycles at 60-years of a quasiperiodic oscillation isn’t going to “nail” the frequency down. (Nor do we really expect these sorts of oscillations to be independent in frequency to the forcings that cause them.)

  29. Carrick said


    I did some googling, and discovered that there is also a 60 year cycle in interest rates. They are about to turn the corner, which may be useful information. Here’s the evidence!

    Probably a generational effect, I’m guessing it’s tied into the number of productive years in a human life. (Hopefully you are aware that there is cyclic behavior in socio-political-economic systems too.)

  30. Carrick said

    Nick, oddly if you think about it… if there were a circa 60 year cycle in GDP, anthropogenic emissions would act as a proxy for that productivity cycle.

    But emissions also modify climate.

    So weirdly enough… there is a link of sorts.

  31. Carrick said

    Here’s a comparison of the spectral content of GISTEMP, NCDC and HACRUT..

    All three have a roughly 60-year peak present:

    GISTEMP 58.5185 0.11476293
    NCDC 56.9369 0.13302032
    HADCRUT 56.2029 0.1139709

  32. steve fitzpatrick said


    Yes, everything about the GISS aerosol forgings is contrived. See
    You can support any climate sensitivity you want if you can adjust aerosols without constraint.

  33. Craig Loehle said

    31 & 32 thanks for those graphs. cool.

  34. Carrick said

    I’m with Steve on the contrived nature of aerosols, but unfortunately for the modelers, there isn’t enough “wiggle room” to get the full range of desired sensitivities. So they also adjust cloud forcings to fine tune the model (there is a much larger uncertainty range associated with that.)

    Back to Nick’s 60-year cycle in GDP… there is a feedback going the other direction too: Years with bad weather have an adverse effect on GDP and years with good weather have a net positive effect on GDP. If there is a 60-year cycle that modulates disruptive weather patterns (abnormally cold weather ironically is more disruptive), then that should actually show up in GDP metrics.

  35. Carrick said

    You’re welcome Craig. Did you notice the ~21 year solar-cycle period (not the 11 year period…that’s sunspot period not solar activity period) had the second largest amplitude in all three data products?

  36. steve fitzpatrick said

    Carrick #34,

    I don’t know. The IPCC’s AR4 credible aerosol offset range (direct and indirect) covers form 0.4 watt/M^2 to 2.7 watts/M^2. That seems sufficient to justify any sensitivity a particular model generates. Cloud parametrization gives added flexibility in ‘forcing’ a fit to the data I suppose.

  37. Craig Loehle said

    35 Carrick: yes that is the signal we detected.

  38. steve fitzpatrick said

    You are right about volcanoes. The GISS climate model is very simple to understand:

    Measured warming = {0.8 degree/(watt/M^2)} * {(accurately estimated GHG forcing)* 0.47 – 2*(actual OCEAN heat uptake rate) – (contrived volcanic aerosol offset)}

    With the contrived volcanic offset, it matches the temperature history almost perfectly. The 0.8 constant comes mainly from James Hansen’s dreams/nightmares back in the early 1980’s, but with a slight adjustment to better fit more recent temperature data (original constant was close to 0.9).

    It is utter rubbish.

  39. Carrick said

    SteveF, there’s a recent publication by Huybers (Journal of Climate 2010) that suggests that the “sensitivity knob” used by the climate models is cloud forcings. So they could use either one I suppose, but it’s my impression that the aerosol data are provided by just a few groups (the climate modelers don’t generally concoct their own aerosol histories, so that isn’t something they as much control too much….cloud forcings is a true “knob”, it’s just a few parameters that you vary).

    By the way, here’s a comparison of GDP growth to temperature.

    As I said above, it’s not surprising there is a correlation.

  40. steve fitzpatrick said


    The more tunable parameters the better the fit. With only four parameters, the diagnosed sensitivity could become almost elephantine. Wait… that is already the case. 😉

  41. To Carrick,

    the power spectra shown in 32 are essentially equivalent to my power spectra. See figures 3 and 6 in N. Scafetta, “Empirical evidence for a celestial origin of the climate oscillations and its implications”. Journal of Atmospheric and Solar-Terrestrial Physics 72, 951–970 (2010),

    Click to access scafetta-JSTP2.pdf

    Major peaks

    9.1 (solar/lunar tidal)
    10-11, solar/JS_planetarian
    20-22, solar/JS_planetarian
    ~ 60 solal/JS_planetarian

    But there are other peaks.

    About the GISS model performance see my figures 7 and 9 in the same paper paper.
    GISS model does not agree at all with the data!

  42. DeWitt Payne said

    Re: steve fitzpatrick (Oct 16 13:03),

    Column 4, assuming you don’t count the date column, is solar. That mainly adds wiggles. But ozone and stratospheric water vapor are pretty much guesses too. Even if you include everything else except stratospheric aerosols, reflective aerosols and aerosol indirect effect (columns 7,9 and 10), otherwise known as the fudge factors, the shape hardly changes. It’s still not linear and it still starts before 1942. Saying that the flat part of the temperature curve was caused by aerosols is at the heart of the kludge to obtain high climate sensitivity. If it’s actually all or mostly quasi-cyclic behavior, the aerosol forcings are wildly incorrect and shouldn’t be included in their current form. The stratospheric aerosol forcings from volcanoes are highly suspect, IMO, because Krakatoa in 1883 should have caused much more cooling than it did according to the temperature record of the time. Either that or the temperature record of the time isn’t very useful.

    In the two box model calculations using the GISS forcings, the main reason for the long time constant was to keep the stratospheric aerosol forcings in check.

    Here’s another graph where I took 65 year moving averages. The data plotted are GISSTemp, W-M ghg’s only and the sum of all 10 columns of the Model E forcing data. GISSTemp isn’t linear. Neither is W-M ghg’s. The forcing sum, however, is nearly linear. The W-M ghg’s are obviously going to correlate better. Yes, I know, eyeball and smoothing. But I’m not trying to publish it either.

  43. DeWitt Payne said

    Re: Carrick (Oct 16 15:37),

    On further thought, the aerosols are mainly there to make the models hindcast the twentieth century correctly and less for climate sensitivity. Still, without the aerosols or with much reduced aerosols, you put an upper bound on climate sensitivity.

  44. Carrick said


    the power spectra shown in 32 are essentially equivalent to my power spectra

    Agreed, I just did it on the three series to show it wasn’t an artifact of one of processing from one of the series. I tend to prefer ordinary Fourier transforms for quasiperiodic phenomena over say MTM or MEM, but regardless of the method of estimating the spectrum, it’s present in all series and in series such as the 1000-year duration proxy data linked in #28.

    I know Nick was being snarky with his comment on the 60-year peirod being present in GDP, but if you think about it, it needs to be there if it’s a real temperature fluctuation (and not just measurement noise).

    It’s not a surprise that GISS fail to reproduce the spectral characteristics of temperature measurements. This is due to the too coarse spatial and temporal resolution in that model. It’d be interesting to look at CESM or other models for which it is claimed the model was optimized to simulate ENSO and other atmospheric oscillations.

    I’d be surprised if any of the models can reproduce the 60-year cycle though (based on no particular insight on my part).

  45. Yes, Carrick
    the relation between the GDB and climate does exist. It is evident as you say and it was known since ancient times.
    I talk about it in my papers and has been observed by several authors.
    No IPCC model reproduce the 60-year cycle

  46. steve fitzpatrick said


    Nick was being snarky

    Some things can go unsaid.

    Nick is primarily an apologist for climate model failures, and an advocate for high climate sensitivity. I defy anyone to point to Nick expressing significant doubt about high climate sensitivity… in spite of all data to the contrary.

  47. steve fitzpatrick said


    I’d be surprised if any of the models can reproduce the 60-year cycle though (based on no particular insight on my part).

    I’d be surprised too. They would have to start by getting the oceans at least sort-of right, and they are presently a very long way from that.

  48. Carrick said


    They would have to start by getting the oceans at least sort-of right, and they are presently a very long way from that.

    Yep, that’s what I think too.

  49. RB said

    K-waves/GDP correlation with ~60 year cycles apparently appear to be a strong feature of the post-1870 world economy but not prior. There is the risk of being fooled by randomness.

  50. Carrick said


    K-waves/GDP correlation with ~60 year cycles apparently appear to be a strong feature of the post-1870 world economy but not prior.

    Which is not surprising, we didn’t have the linkages in the economy that would allow any weather related cycles to show up previous to about then, and GDP wasn’t as well measured in earlier times (noise was higher).

    For the record though, my GDP spectral analysis started in 1792 not 1870, and if you just look at the first 120-years, there is still a roughy 60 year period (I wanted two complete cycles, so some overlap is unavoidable). The amplitude is even slightly larger.

    There is the risk of being fooled by randomness.

    Which is why you look for agreement among multiple measures, as I did here.

  51. Carrick said

    Here are the peak frequencies and amplitudes:

    1792-1911 59.2593 0.22190808
    19792-2011 56.7742 0.15449478

  52. Carrick said

    (Make the last line 1792-2011 56.7742 0.15449478)

  53. Mark T said

    Hehe, it comes through as a (Skype) telephone # on my browser with the typo intact.


  54. Carrick,

    may you write me an email?

  55. Stephen Wilde said

    I’m sure of the existence of the 60 year climate cycling but underlying it is a cycling of about 500/1000 years from MWP to LIA to date.

    Getting a solar link to that would be a larger step forward and coincidentally it does seem to correlate with solar activity from LIA to date but the signal from one solar cycle to the next is rather weak.

    I would expect the oceans to modify the solar effects heavily on timescales of less than 60 years though. Not so much on the 500/1000 year timescales.

  56. DeWitt Payne said

    Then there’s the Dansgaard-Oeschger/Bond events that may be quasi-periodic with a period of ~1500 years. Heinrich events, if they are quasi-periodic, have an even longer period.

  57. RB said

    The climate driver meeting the Kondratieff wave technological innovation driver as a modern equivalent of the Biblical debt jubilee cycle … beware the western economy centered pattern seeker perhaps. Someone has even brought these elements together seemingly including Scafetta’s planetary attribution even.

  58. RB said

    Well, maybe not, the Hindus and the Chinese are on board for the 60-year cycle too. With Jupiter involved, as Scafetta says.

  59. @RB

    And what physical reality do the Hindi and Chinese 60-year cycles relate to? Fishing for cycles in a noisy signal is one thing, ascribing the supposed cycle to a real, measurable forcing is another. What is it exactly that’s on the rise for the last 30 or 40 years? Because if we’re talking about solar, then no measured parameter fits the picture. Solar has been stagnant or trending a little downward all that time (apart from the 11-year cycle).

  60. […] (pre-1976) data. The second is a recent paper by Loehle and Scafetta (L&S), which was discussed here by Jeff Id.  They noted, like other researchers, that the temperature series could be fitted by a […]

  61. Evidence of the 60 year cycle becomes much more apparent when looking at a plot of moving average 30-year temperature gradients as at the foot of my Home page

    This is the clearest plot I have seen which shows …

    (a) absolutely no anthropogenic effect above the long-term ~1,000 year cyclic trend – heading for a maximum within 50 to 200 years (yellow line)

    (b) a slightly decreasing rate of increase which gives a projection of about another 0.5 deg.C above current temperatures by 2100.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: