RealClimate logo

The Rise and Fall of the “Atlantic Multidecadal Oscillation”

Filed under: — mike @ 4 March 2021

Two decades ago, in an interview with science journalist Richard Kerr for the journal Science, I coined the term the “Atlantic Multidecadal Oscillation” (AMO) to describe an internal oscillation in the climate system resulting from interactions between North Atlantic ocean currents and wind patterns. These interactions were thought to lead to alternating decades-long intervals of warming and cooling centered in the extra-tropical North Atlantic that play out on 40-60 year timescales (hence the name). Think of the purported AMO as a much slower relative of the El Niño/Southern Oscillation (ENSO), with a longer timescale of oscillation (multidecadal rather than interannual) and centered in a different region (the North Atlantic rather than the tropical Pacific).

Today, in a research article published in the same journal Science, my colleagues and I have provided what we consider to be the most definitive evidence yet that the AMO doesn’t actually exist.


Back in the 1980s and 1990s, a number of articles pointed to a pattern of North Atlantic warming during the 1930s-1950s, subsequent cooling in the 1960s and 1970s, and warming thereafter, which seemed to resemble a natural oscillation in the climate system. I co-authored an article (Mann et al, 1995) in the Nature demonstrating the apparent persistence of these multidecadal oscillations several centuries back in time based on the analysis of paleoclimate proxy data (our analysis of similar proxy data would ultimately lead, a few years later, to the now well-known “hockey stick” curve that shows the warming of the past century to be anomalous in a long-term context).

In 2000, in an article that led to Kerr’s commentary in Science, my collaborator from the climate modeling group at the Princeton Geophysical Fluid Dynamics Laboratory (GFDL) Tom Delworth and I argued that this observational evidence for an AMO-like climate oscillation was supported by an analysis of an extended (thousand+ year long) control simulation of GFDL’s state-of-the-art (at the time) coupled ocean-atmosphere model (Delworth and Mann, 2000) (pdf). Since it was a control simulation with no external “forcing” (no greenhouse gas changes, no variations in solar output, no volcanic eruptions, etc.), any oscillation that was produced has to be internally generated. And indeed we established that this model did produce such an internal oscillation, with a multidecadal timescale, centered in the extratropical North Atlantic, tied to coupled ocean-atmosphere processes involving the ocean “conveyor belt” circulation (sometimes called the thermohaline circulation and sometimes equated with the “Gulf Stream”, though the latter in fact is a wind-driven current in mid-latitudes, while the thermohaline circulation/conveyor belt represents its extension on into the higher latitudes of the North Atlantic).

About five years later, analysis of an extended simulation of yet another climate model–the coupled ocean-atmosphere model run by the Hadley Centre within the UK Meteorological Office, yielded evidence for a similar oscillation, albeit with a longer (roughly 100 year) period, and a more global signature (Knight et al., 2005).

As I’ve stated elsewhere, at times I feel like I created a monster when I gave a name to this putative climate oscillation in 2000. The concept of the AMO has since been misapplied and misrepresented to explain away just about every climate trend under the sun, often based on flawed statistical methods that don’t properly distinguish a true climate oscillation from a time-varying trend: If you assume that all trends are a simple linear ramp, and call everything left-over an “oscillation”, then the simple fact that global warming flattened out from the 1950s through the 1970s driven by the ramp-up in cooling sulphate aerosol pollution masquerades as an apparent “oscillation” on top of a simple linear trend. We’ve published a number of articles over the years (see e.g. here, here, here, here, here, and here) demonstrating that studies that use such an approach to define the AMO end up mis-attributing to a natural “oscillation” what is actually human-caused climate change. Such analyses have been used by some to dismiss, among other things, the impact climate change is having on increasingly active and destructive Atlantic hurricane seasons, attributing the increase in recent decades to a supposed upturn in the AMO.

But if the AMO is simply an artifact of studies that misinterpret the time-varying pattern of human-caused climate change as a low-frequency oscillation, what about the studies mentioned previously that identify an internal oscillation in control simulations of climate models? It turns out that they are the exception (indeed the rare exception), rather than the rule. Decades ago there were only a handful of long control coupled model simulations and two in particular (the GFDL and UK Met Office coupled models, as noted earlier) did generate an AMO-like oscillation (though the spatial patterns, timescales, and apparent mechanisms in the two cases were different enough that one could question whether they truly represented the same thing).

Today, by contrast, there are many dozens of coupled models around the world, and the Coupled Model Intercomparison Project (CMIP) makes the ensembles of simulations widely available to other researchers. So what does an analysis of these multimodel ensembles reveal?
Analysis of Historical and Control Simulations

In an article we published a year ago, we showed that the AMO does not in general exist in current generation models. We analyzed a set of long control simulations with more than 40 different state-of-the-art climate model simulations from the CMIP5 multimodel climate archive. We used a type of “spectral analysis”, a statistical procedure that identifies whether there is evidence for truly oscillatory variability (in the form of a spectral “peak”, i.e. a spike in the plot of amplitude of variation as a function of frequency/period) at some particular timescale or narrow range of timescales. The “MTM-SVD” method identifies whether an entire spatiotemporal dataset contains an oscillatory signal (as indicated by a spectral peak that is correlated across the dataset, in this case, surface temperatures spanning the globe) that is distinct from simple background noise (i.e. random variability).

As shown below (Figure 1–left), the CMIP5 control runs show strong evidence of oscillatory behavior on 3-7 year interannual timescales associated with ENSO. The vast majority of simulations exhibit a statistically significant spectral peak within that range of periods, and even the average across all simulations crosses the 90% (“p=0.1”) significance threshold, despite the tendency for cancellation when averaging (since different models yield significant ENSO timescale peaks at different periodicities within the 3-7 year range). By contrast, the models show no evidence for a spectral peak on the 40-60 year AMO timescale (see the inset which zooms in on decadal and longer timescales). The behavior on that timescale is largely consistent with what one would expect from random data.

Yet, actual historical surface temperature data from 1850-present (see thick blue curve), in addition to showing significant interannual ENSO timescale peaks, do indeed yield a spectral peak centered at a multidecadal 40-50 year period. Do the results for the historical data, then, contradict what we find with the model simulations? As it turns out, no.

Consider a parallel analysis (Figure 1-right) of the CMIP5 historical simulations. These are simulations that have been driven with the same external factors (the human factors of rising greenhouse gas concentrations and sulphate aerosols, and natural factors such as solar output changes and volcanic eruptions) that have driven temperature changes over the historical period of 1850 to present. In this case, the models reproduce the 40-50 year multidecadal spectral peak seen in the actual observations. Indeed, the peak is seen in nearly every simulation and is significant at the 99% (“p=0.01”) level even for the average over all simulations. Clearly the multidecadal spectral peak is a very robust feature of both the historical simulations and the historical observations themselves. But it’s not indicative of an AMO.

An analysis of the simulations shows that the multidecadal peak is tied to the same pattern of 1930s-1950s warming, 1960s and 1970s, cooling and warming thereafter that is seen in the actual observations as discussed earlier. That alone raises suspicions, since if this were an internally-generated oscillation, there is no reason that the warming and cooling phases in the models would coincide in timing with those in the observations. Instead, as discussed earlier, the alternating warming and cooling is tied to the competition between steady greenhouse warming and the prominent sulphate aerosol cooling of the 1950s-1970s. The aerosol cooling is especially pronounced in the North Atlantic (particularly during summer) as noted by Mann and Emanuel (2005) helping to explain why the apparent “oscillation” is most pronounced in that region. 

Figure 1. Spectra from MTM-SVD analysis of global surface temperature fields from CMIP5 control simulations (left) and historical simulations (right). Shading with mean over all model simulations is shown by black curve and historical result is shown by blue curve. Dark grey region bounds 68% of the simulations while light grey region bounds 95% of the simulations. Inset zooms in on the decadal (f=0.1 cycle/year) and longer periodicities. Horizontal dashed lines correspond to median (p=0.5) and p=0.1, 0.05 and 0.01 significance levels relative to colored noise null hypothesis.

At this point prospects for the existence of an internal AMO climate oscillation might be starting to look rather bleak. But there is still one seemingly compelling argument left for AMO advocates: If the AMO is indeed an artifact of competing anthropogenic greenhouse gas and sulphate aerosol forcing during the historical era, then why is it that AMO signals, as noted earlier, have been detected in paleoclimate proxy data that predate the historical period? In fact, a recent critique (Müller-Plath, 2020) of our analysis of the CMIP5 control and historical simulations published in the somewhat controversial journal Frontiers rests on that very argument. Is there merit to this argument?
Analysis of Last Millennium Simulations

In our new article, my co-authors (Byron Steinman, Daniel Brouillette, Sonya Miller) and I analyzed the CMIP5 “last millennium” simulations (16 in total) that span the interval 850-1849 CE. These simulations precede the historical period and are driven by natural (primarily volcanic, solar and orbital) radiative forcing alone. Application of the same (MTM-SVD) analysis as above to the model surface temperature fields yields evidence for a multidecadal (~60 year period) AMO-like spectral peak in the majority of simulations (12 out of 16 at the 90% significance level and 11 out of 16 at the 99% significance level). The results are shown in Figure 2 where they are compared against the previously discussed “control simulation” results. The difference is striking. In the control simulations, where no forcing is applied, there is no evidence (as we’ve already seen) for a distinct multidecadal spectral peak. Yet there is robust evidence for just such a peak in the “last millennium” simulations, where natural (volcanic and solar) radiative forcing has been applied. This comparison suggests that the natural radiative forcing must be responsible for the apparent AMO-like signal.

Figure 2. Spectra from MTM-SVD analysis of global surface temperature fields . A. CMIP5 control simulations. Individual colored curves depict results for all N=44 simulations while the ensemble mean is shown by the thick black curve. Horizontal dashed lines correspond to median (p=0.5) and p=0.1, 0.05 and 0.01 significance levels relative to colored noise null hypothesis. B. Same as A. but for the N=16 CMIP5 Last Millennium simulations (the more prominent blue curve denotes the GISS-E2-R simulation examined in Figure 3) below.

One convenient feature about the MTM-SVD method is that it allows you to reconstruct the spatial and temporal pattern of the signal associated with a particular spectral peak. We take, for example, the individual simulation (a simulation of the NASA GISS-E2-R model) that produced the most prominent ~60 year spectral peak in Figure 2B. The spatial and temporal pattern of the corresponding signal is readily reconstructed and shown in Figure 3 below.
There are a number of revealing features in the signal pattern. The spatial pattern (Figure 3A) displays a high-amplitude response in the tropical regions that is reminiscent of the pattern of response to explosive tropical volcanism (temperatures cool the most in the tropics after an explosive tropical eruption since the more sunlight you have in the first place, the more that is reduced by a volcanic dust veil). It is notable that there is also some evidence of enhanced signal amplitude along the Gulf stream extension in the North Atlantic, suggestion some possible coupling to North Atlantic ocean dynamics. A detailed analysis of the evolving spatial pattern of the signal over a typical ~60 year cycle (see article) shows a close correspondence between the initial phase (coinciding with peak global cooling) and the pattern of response to explosive tropical eruption established in some past studies (e.g. Shindell et al, 2004).

The temporal pattern (Figure 3BC) shows that the major cooling excursions coincide with several of the largest explosive tropical eruptions of the last millennium (e.g. the 1258 CE, 1331CE, and 1453 CE eruptions), which happen to be paced in a manner that projects onto an apparent multidecadal (60-70 year period) “oscillation”. Past studies have indeed noted the coincident multidecadal pacing of explosive volcanic activity in past centuries (i.e. Ammann and Naveau (2003)).

Figure 3. Spatial and temporal characteristics of multidecadal “signal” (centered at f=0.016 cycle/year/ ~63 year period) for CMIP5 GISS E2-R Last Millennium simulation. A. spatial pattern of % resolved variance associated with signal and B. reconstructed time-domain signal for representative equatorial eastern Atlantic grid box (grid box centered on longitude 35W and latitude 0; location denoted by the large black ‘+’ in panel ‘A’).

A spectral analysis of a simple “energy balance” climate model driven with volcanic-only, solar-only, and volcanic+solar forcing (see article) shows that there is indeed a multidecadal spectral peak in the response of surface temperatures to natural radiative forcing and that peak arises from the volcanic forcing alone. We conclude that the apparent AMO-like signal during the last millennium is a consequence of the coincidental multidecadal pacing by episodes of explosive volcanic forcing.

The available evidence both from observations and current-generation climate models, in summary, does not provide any support for an internal AMO-like oscillation in the climate system.

That doesn’t mean that there aren’t interesting questions left to address here. Our planned future work will examine the next generation (CMIP6) coupled models to see if improved representations of ocean and atmospheric dynamics might somehow lead to modes of internal decadal/multidecadal variability that are not evident in the CMIP5 models. And it will be instructive to analyze the considerably more extensive networks of paleoclimate proxy data that have become available since our 1995 analysis. Of particular interest is whether the multidecadal oscillations in the actual paleoclimate proxy data show the same relationship with volcanic forcing demonstrated in the last millennium simulations. 
Concluding Thoughts

There are several lessons in this tale. One is that scientists must always be open to revising past thinking. That is part of the critical scientific process—what the great Carl Sagan referred to as the “self-correcting machinery” of science. Two decades ago there seemed to be both observational evidence and modeling evidence (if rather limited) for the existence of a multidecadal AMO in the climate system. My own work supported that interpretation, and indeed it was I who gave this beast a name. The scientific community ran with the concept, and numerous scientists—even some at our leading research laboratories like the aforementioned GFDL—continued to misapply it in a way that downplays some critical climate change impacts like the warming of the North Atlantic and the increase in Atlantic hurricane activity associated with it.

Now we have come full circle. My collaborators and I, over the past decade, have continued to investigate the origins of the putative AMO signal and have been led inescapably to the conclusion that the AMO (unlike, say, R.O.U.S.) doesn’t actually exist. It’s an artifact, during the historical era, of competing anthropogenic (greenhouse warming and sulphate aerosol cooling) drivers and, during the earlier period, an artifact of the fact that volcanic forcing happens to have displayed a roughly multidecadal pacing in past centuries.

A scientist has to admit when they are wrong. Unfortunately for all of us, my colleagues and I weren’t wrong about the unprecedented warming revealed by the now iconicHockey Stick” curve, despite the unrelenting attacks on it by climate change deniers over the past two decades.
But I was wrong about the existence an internal AMO oscillation when I coined the term twenty years ago.

Nonetheless, some very good science has been done by a number of researchers and groups around the world in pursuing this matter. And we have learned quite a bit, for example, about the true origins of multidecadal climate variability, and prospects for long-term climate prediction.
That, in fact, is science (and Science) working the way it’s supposed to.


  1. M.E. Mann, J. Park, and R.S. Bradley, "Global interdecadal and century-scale climate oscillations during the past five centuries", Nature, vol. 378, pp. 266-270, 1995.
  2. T.L. Delworth, and M.E. Mann, "Observed and simulated multidecadal variability in the Northern Hemisphere", Climate Dynamics, vol. 16, pp. 661-676, 2000.
  3. J.R. Knight, "A signature of persistent natural thermohaline circulation cycles in observed climate", Geophysical Research Letters, vol. 32, 2005.
  4. G. Müller-Plath, "Internal Multidecadal and Interdecadal Climate Oscillations: Absence of Evidence Is No Evidence of Absence", Frontiers in Earth Science, vol. 8, 2020.
  5. D.T. Shindell, "Dynamic winter climate response to large tropical volcanic eruptions since 1600", Journal of Geophysical Research, vol. 109, 2004.
  6. C.M. Ammann, and P. Naveau, "Statistical analysis of tropical explosive volcanism occurrences over the last 6 centuries", Geophysical Research Letters, vol. 30, pp. n/a-n/a, 2003.

81 Responses to “The Rise and Fall of the “Atlantic Multidecadal Oscillation””

  1. 51
    Alan Lowey says:

    Oops! I’ve just seen my mistake

  2. 52
    zebra says:

    Follow-Up to #37,

    (Thanks to the author for the reply.)

    I looked at the papers, although I didn’t have full access to two of them, and as best I can understand it, the underlying (proposed) ’cause’ of periodic variation of AM temperature is periodic variation in the AMOC. I was reminded of that thing about turtles…

    But seriously, there was also just the discussion of ‘contradictory’ AMOC papers (changing, not changing), and to me, it all reinforces the need for better communication with the public. The actual science is fine; it does what science is supposed to do, but I keep having to point out, even here, lack of clarity in language about very basic stuff.

  3. 53
    Killian says:

    45 William B Jackson says:
    11 Mar 2021 at 2:44 PM

    No 43. You got that right if we are speaking of Florida for instance it won’t take all that much sea level rise before it becomes a few marshy islands. Though I do think the developer just outside Little Rock was “probably” stretching things some years ago with the sign advertising future ocean front property! That was about 20 years ago if I remember right.

    It depends on the time frame. Baton Rouge being beachfront this century is not out of the question.

  4. 54
    Killian says:

    20 Liz Hanna says:
    7 Mar 2021 at 12:09 AM

    Chris Crawford nailed it when he spoke of “truth” or “correctness” being a function of ‘the time’, when forming a thesis based on best available science.

    Crawford needs better nails, perhaps. Truth is. It has no time function. The perception or belief of what is true and/or correctness *is* a function of the context it occurs in, however.

    it is the way science works… and moreover, why it is the best method we have to begin to understand the world around us.

    Unfortunately, too many who praise science think it is only done by university students and PhDs. As the discussion about the Kogi people on another thread shows, science is done not just in a lab.

  5. 55
    Matt Skaggs says:

    “The behavior on that timescale is largely consistent with what one would expect from random data.”

    Is there any aspect of this analysis that is demonstrably inconsistent with autocorrelated random data? That is, can you go beyond statements like “strong evidence” and show how you were able to conclude that any fluctuations in any of the data were oscillatory rather than random?

    I presume the answer is in the paywalled paper, but maybe you could share a few details with the public?

  6. 56

    Alan says:

    “Mike: why is the correlation of the AMO with the Pacific MJO not discussed?
    So, unlike ENSO, which is stationary, the MJO is an eastward moving disturbance of clouds, rainfall, winds, and pressure that traverses the planet in the tropics and returns to its initial starting point in 30 to 60 days, on average.

    MJO correlates quite closely with a 21 day lag shift of the high-resolution SOI data (which is an accurate proxy for -ENSO). This time series correlation in the following chart should be spread far and wide

    So, likely yes indeed, its the MJO which is a traveling wave offshoot from the ENSO disturbance along the equator, as the link is showing

  7. 57

    KIA 49: Humans have been collecting large quantities of accurate and global climate data for less than 100 years, probably for only 70 years.

    BPL: Don’t just make stuff up, KIA. It’s too easy for people to check.

    We’ve had enough reports for reliable global temperatures, pressures, cloudiness, wind direction and speed, and rainfall since 1850. We have carbon dioxide back to 1959 and proxies for it in ice cores back 800,000 years with some data from a million years ago. We have seafloor and lake sediments that go back as much as three million years. BTW, we have a Central England temperature series that goes back to 1650, when Oliver Cromwell was Lord Protector of Egypt.

    Anyone can look this stuff up on the internet. Are you lazy, or are you deliberately trying to mislead people?

  8. 58
    Ray Ladbury says:

    Mr. KIA: “Humans have been collecting large quantities of accurate and global climate data for less than 100 years, probably for only 70 years.”

    Utter and complete bullshit. There are some local climate records that go back hundreds of years. There are proxies that go back hundreds of thousands if not millions of years.

    Your failure to understand the data does not undermine its cogency. Do better.

  9. 59

    Sorry, Lord Protector of England, of course, not Egypt. I don’t think Cromwell ever made it to Cairo.

  10. 60
    CCHolley says:

    RE. Mr. Know Nothing @49

    Perhaps in another 100 years we’ll have enough data to make high probability assertions about what the climate is doing, but not yet.

    What a bunch of baloney.

    Oh, the arrogance of Mr. Know Nothing who knows nothing about science and the statistical handling of data making an absurd proclamation as if he has expertise. What a doofus. A clueless amoral climate denier who is unwilling to learn the science and cares nothing for future generations.

  11. 61

    KIA says:

    “Perhaps in another 100 years we’ll have enough data to make high probability assertions about what the climate is doing, but not yet.”

    There is enough empirical data already to do detailed cross-validation of many climate models, which means that we don’t have to wait another 100 years. The issue is that viable approaches to cross-validation aren’t widely known yet — I gave a presentation at AGU a few years ago describing several examples. And even more will come from machine learning experiments.

  12. 62
    William B Jackson says:

    No 60 I would humbly suggest that the case KIA is not that of a know nothing, it is rather a case of one who ignores factual reality to make his point. In other words that of an willful liar! If I am in error and he is really the ignorant person who can learn nothing he pretends, I must apologize and express great sadness.

  13. 63
    Ray Ladbury says:

    BPL: ” I don’t think Cromwell ever made it to Cairo.”

    Given what he did in Ireland, that’s probably lucky.

  14. 64
    barn E. rubble says:

    RE: 35
    Kevin McKinney says:
    “Not to mention the fact that on current climate trends, the noble tradition of outdoor hockey is at risk . . .”

    Not to mention they won’t even know what snow is . . .

  15. 65
    barn E. rubble says:

    As the comments seem to have drifted from the point of article, I’ll ask again:
    Has anyone from the NOAA weighed in on this with their thoughts? The NOAA has (still) an AMO Index page on their site.

  16. 66
    nigelj says:

    barn E. rubble says @64, yes the climate problem cannot be adequately solved without some government control. For example things like carbon taxes and subsidies for renewable energy, phasing out ICE cars, which are all indirect forms of control. The purpose of government is primarily to exert control, for example the police force, the criminal law, and environmental law. If that troubles you, the rule of the jungle awaits you with open arms.

  17. 67
  18. 68
    zebra says:

    barn E rubble #65,

    The NOAA site has one or two sentences (literally) which might be changed, like

    “AMO is probably a natural climate oscillation”.

    And the reference to the AMOC as what is causing the temperature to vary as measured.

    So, as I pointed out above, the issue is with what is meant by “oscillation”. The OP is not saying that the data doesn’t exhibit a pattern that could be characterized (perhaps loosely) as periodic.

    What has changed is the causal narrative… the suggestion is that it isn’t energy moving back and forth within what is defined as the climate system, but the result of external forcing.

    So, of all the things on NOAA’s plate right now, trying to get back to normal after the last 4 years, perhaps they feel that a breathless reaction to one paper on a very complex topic is not a priority.

  19. 69

    zebra said:

    “What has changed is the causal narrative… the suggestion is that it isn’t energy moving back and forth within what is defined as the climate system, but the result of external forcing.”

    Quite a few possibilities for external forcing — the daily (diurnal) cycle, the annual or semiannual cycle, any tidal forcing is external, other very long orbital variations, volcanic emissions, and everything related to GHG emissions is external of course.

    The non-external “unforced” variations are typically assumed to be natural resonances that are eigenvalues or attractors of the geophysical fluid dynamics.

    The arguments are analogous to whether sunspots are externally forced by planetary orbits (this submitted paper) or due to the magnetodynamics of the molten layer, see this recent review paper for the latter

  20. 70
    Brett Schmidt says:

    I’ve read that Einstein said that anyone who has never made a mistake has never tried anything new.

    I’m also curious as to what this means for other possible multidecadal climate oscillations, such as the PDO (Pacific Decadal Oscillation) and the IPO (Interdecadal Pacific Oscillation), which, along with ENSO (El Nino/ Southern Oscillation) are thought to have a major impact on the weather over a large area of the globe, including where I live on Eastern Australia.

  21. 71

    #64, BER–

    Not to mention they won’t even know what snow is . . .

    I suspect this was intended as an attempt at cutting irony, but if so, it fails by displaying ignorance of the basics of the topic at hand. I’d elaborate, but it would be tedious and mostly either OT or of limited relevance for most readers, who can hardly be expected to care much about the necessary climatic conditions for building a practical backyard skating rink versus those necessary for the occurrence of occasional snowfalls.

    Though on second thought, they might care about the difference between an actual research study and an offhand opinion.

    (And by the way, speaking of offhand opinion, I wonder why such as Barn E. never quote the bit where the much-ridiculed Dr. Viner said “Snow will probably cause chaos in 20 years time”?)

  22. 72
    MA Rodger says:

    Brett Schmidt @70,

    The PDO & the IPO describe very similar phenomena and at the heart of both is ENSO. If you smooth out the ENSO time series the resulting time series is effectively the same as the PDO/IPO.
    ENSO itself I would describe more as ‘an instability that results in an oscillating situation’ rather than an actual oscillation. Unlike AMO, ENSO is not the outcome of number-crunching but of actual events in the real world. You can actually measure the change in height of the western Pacific resulting from ENSO.

  23. 73
    Richard Caldwell says:

    BPL: Don’t just make stuff up, KIA. It’s too easy for people to check.

    RC: So? It is far easier to not check. And one gets to own the libs as a bonus.

    So, why on Earth would KIA’s target audience check diddly when laughing at you is easier and way more fun?

    Remember: KIA is here specifically to make people laugh. That you ain’t included in the entertained is your problem, not KIA’s.

  24. 74
    Richard Caldwell says:

    Killian: . As the discussion about the Kogi people on another thread shows, science is done not just in a lab.

    RC: Yep. And way too much science is done specifically so as to exclude holistic meta-truths.

    Them: Isolate and exclude everything except this single factoid we’re pondering

    Me: Ponder wider.

  25. 75
    Richard Caldwell says:

    Killian: Crawford needs better nails, perhaps. Truth is.

    RC: You’re speaking a different language. Crawford is concerned about perceptions (though I don’t know Crawford, whoever that is). So interpret his/her words accordingly, while excluding whatever “inherent intellectual conflict” your connecting with him/her causes.

    So, educate me about Crawford while channeling a Crawford supporter (or accepter).

    Tell me about Crawford in the gushingest words you can muster

  26. 76
    Dan says:

    re: 73 “Remember: KIA is here specifically to make people laugh.”

    Oh we definitely laugh at him and his complete failure to understand basic scientific laws. And how science is conducted. And that he thinks he knows something that literally every professional climate science organization in the world (including the National Academy of Sciences) does not; the height of his scientific arrogance and ignorance. We also laugh at his cowardice and insecurity when it comes to his inability to admit his brethren killed people in DC on 1/6, in Charlottesville, at the synagogue in Pittsburgh, and many other locations.

  27. 77

    MAR says:

    “If you smooth out the ENSO time series the resulting time series is effectively the same as the PDO/IPO.”

    Actually there is a higher-frequency component to the PDO which is more like the NAO. So what I would say is that the PDO may be a composition of some of ENSO and some type of NAO applied to the north Pacific.

  28. 78
    nigelj says:

    I don’t think you can assume KIA is here just to annoy liberals and play to his supporters. He sounds more like he’s a climate denialist doing work for a lobby group. He has the exact characteristics they employ. Read a media article by some expert going through this, and it made me think of KIA. Annoying liberals and making his denialist supporters laugh is a bonus. And yes KIA is amazingly ignorant.

  29. 79
    Yannick Peings says:

    What about low-frequency atmospheric variability as a driver of the AMOC/AMO ? I.e. low frequency variability, or trends, in the North Atlantic Oscillation ? Persistence in NAO and associated wind stress/heat flux anomalies can force AMOC and AMO variability, doesn’t it ? Low frequency variability in the NAO and jet speed/position is typically underestimated in the models. This is an important driver of AMOC/AMO fluctuations that the models are missing, which may lead to overestimation of the role of external forcings (versus atmospheric internal variability). Did you consider that ?

  30. 80

    I used to run with the “lukewarmer” crowd. I’ve written some dumb things on my blog in the past about climate. I’ve indirectly and directly insulted the scientists who contribute to this blog, for which I humbly apologize.

    This thread is cracking me up. Dr. Mann has revisited his own work, and come to the conclusion that he was not correct, that the AMO actually may not exist… And published a paper on that result.

    Some 15 years, the folks at Watt’s Up With That, and other lukewarmist / denialist blogs (myself included), were crowing the “obvious”, that world temps were going to start cooling due to the AMO turning negative. In short order, that prediction failed.

    Question: Where was the post from my former cohorts acknowledging they were wrong?

    It didn’t happen. All that happened is we / they moved on to the next pet hypothesis. This is a pattern that happened over and over again. Being a former geology / seismology major (calc killed the dream) I had gained a better appreciation for the scientific method. A blog post, even if it looks pretty thorough, is no substitute for science done by experienced professionals.

    Consider this my apology for the crap I wrote a decade ago. :-)

  31. 81

    MA 80,

    That is one of the most mature and humble posts I have ever seen on the internet. Thank you for an illustration of how to do it right. The ability to change one’s mind with new evidence is pretty rare these days.