Prospects for a Prolonged Slowdown in Global Warming in the Early 21st Century

by Nic Lewis

[W]e estimate that the warming slowdown (< 0.1 K/decade trend beginning in 1998) could persist, due to internal variability cooling, through 2020, 2025 or 2030 with probabilities 16%, 11%, and 6%, respectively. – Knutson et al.

A new paper by Tom Knutson, Rong Zhang and Larry Horowitz of NOAA GFDL has just been published in Nature Communications [link to full manuscript]. The authors take a well-balanced approach to seeking possible explanations for global mean surface temperature (GMST) increasing at a much lower rate from around the turn of the century than over the late 20th century, and consider the possibility that it may continue for some time.

The abstract reads:

Observed global mean temperature over 1998 to 2015 increased at a slower rate (0.1 K decade-1) compared to the ensemble mean (forced) projected by Coupled Model Intercomparison Project 5 (CMIP5) models (0.2 K decade-1). Here we investigate the prospects for this slower rate to persist for a decade or more. The slower rate could persist if the transient climate response is overestimated by CMIP5 models by a factor of two, as suggested by recent low-end estimates. Alternatively, using CMIP5 models’ warming rate, the slower rate could still persist due to strong multidecadal internal variability cooling. Combining the CMIP5 ensemble warming rate with internal variability episodes from a single climate model – having the strongest multidecadal variability among CMIP5 models – we estimate that the warming slowdown (< 0.1 K decade-1 trend beginning in 1998) could persist, due to internal variability cooling, through 2020, 2025 or 2030 with probabilities 16%, 11%, and 6%, respectively.

The authors say that CMIP5 and earlier generation climate models’ historical forcing simulations, in the ensemble mean, reproduce well observed late 20th century warming and a hiatus during around 1940–1970, in response to anthropogenic and natural forcings. However, these models do not reproduce the rapidity of the observed early 20th century warming, suggesting that internal (unforced) climate variability played a minor role in the two later episodes but a significant one in the early warming.

Although this comparison with observations does at first sight suggest such a conclusion, the satisfactory reproduction of the 1940–1970 hiatus and late 20th century warming by the CMIP5 ensemble could have other explanations. Whilst in the ensemble mean of climate models the mid-century hiatus and late century warming no doubt are almost entirely forced, there is reason to believe that in the real world multidecadal internal variability –linked to the Atlantic multidecadal oscillation (AMO) and the Pacific Decadal Oscillation (PDO) – played a significant role, depressing the GMST trend from 1940 to the early/mid 1970s and increasing it from then until the turn of the century.[i] In many CMIP5 climate models, excessive strengthening of negative aerosol forcing (cooling) during the 3rd quarter of the 20th century may have served to mimic the effect on GMST of missing multidecadal internal variability. And excessive transient climate response (TCR) levels in CMIP5 models may have compensated, in the late 20th century, for their lack of a positive influence of internal variability on GMST.

As cited by the authors, some recent studies suggest that misrepresentation of forcing accounts for the rapid post-2000 warming in CMIP5 models exceeding the observed warming.[ii] However, these studies focused on individual forcings, such as solar and stratospheric (volcanic) aerosols. When all forcings are taken into account – some of which grew more positively after 2005 than in the RCP scenarios – there is little evidence that the rate of increase in actual total net forcing was overestimated in CMIP5 models. In a thorough investigation of the situation in the NorESM1-M model, it was estimated that over 1998–2012 the actual forcing trend was about 20% higher than that in the RCP8.5 scenario.[iii]

Sensibly, Knutson et al. effectively discount explanations of the early 21st century warming slowdown based on misstated forcing changes, missing temperature data in warming regions such as the Arctic, and the like. Instead they “explore how long an early 21st century hiatus or slowdown could last, assuming a strong internally generated cooling superimposed upon ongoing anthropogenic warming.” They focus on the period 1998–2015, pointing out that:

The warming rate from El Niño peak to peak (1998 to 2015) in observations was about 0.11 K decade−1, and similar (0.10 K decade-1 ) if the two El Niño years at each end are excluded (i.e., 1999 to 2014).

Further, they state:

Reducing the CMIP5 forced warming rate by 50% (to 0.1 K decade-1 ) yields a trend close to the observed warming rate. … if the CMIP5 models overestimate TCR by a factor of two, we might expect global mean temperatures over the next few decades to continue (on average) along the observed trend trajectories since about 1998.

The in press Barcikowska et al. J. Climate paper, of which Knutson and Zhang are co-authors, also points (Fig. 1) to the prospects of the AMO, after peaking during the 2000s, contributing to global surface warming through to 2030 being perhaps ~0.05 K decade-1 slower than the underlying forced trend. This is also consistent with the findings in Delsole et al (2011).

Further, the Stadium Wave paper predicted the same. According to its press release:

The stadium wave hypothesis provides a plausible explanation for the hiatus in warming and helps explain why climate models did not predict this hiatus. Further, the new hypothesis suggests how long the hiatus might last.

“The stadium wave signal predicts that the current pause in global warming could extend into the 2030s,” said Wyatt, the paper’s lead author.

Apart from the influence of multidecadal internal variability, Knutson et al. raise the possibility that the average TCR of CMIP5 models exceeds the transient GMST response of the real climate system. In this connection, it is interesting how consistent the temperature response to forcing (per AR5) has been since 1850. If one excludes years with significant volcanic forcing, which produces an abnormally small temperature response (Gregory et al 2016),[iv] plotting global annual temperature against forcing gives points that cluster closely around a straight line, particularly if the AMO[v] component is removed. Figure 1 shows a plot that I have produced to demonstrate the remarkably linear relationship between AMO-adjusted GMST and forcing.[vi] Gregory and Forster (2008)[vii] used the same method to estimate TCR, excluding years with volcanic forcing stronger than –0.5 W m⁻², although without adjusting for the AMO’s influence.

slide1

Figure 1. Annual AR5 forcing and AMO-adjusted GMST data for 1850–2015, excluding years with volcanic forcing stronger than –0.6 W m-2. Black: OLS regression best fit line, formula and R2.

The line slope equates to a TCR of 1.38 K,[viii] close to the 1.33 K estimated in Lewis & Curry 2014.[ix] An article recently published here,[x] which used post-1950 GMST time series adjusted to remove the influences of ENSO and volcanic and solar forcing variability, came up with a median TCR estimate of 1.35 K. Whilst such observationally-based TCR estimates are uncertain, I find unconvincing arguments that “energy budget” observationally-based TCR estimates like these, which take care to minimise volcanic and multidecadal internal variability influences, are depressed by the mix of forcings acting during the instrumental period having less warming effect than the same amount of CO2 forcing: the evidence used seems flawed to me.[xi]

A TCR in the range 1.33–1.38 K is well below the mean TCR of ~1.8 K for CMIP5 models. This supports Knutson et al.’s suggestion that the transient response of CMIP5 models, on average, exceeds that of the real climate system. Based on these estimates, the average TCR of CMIP5 models likely exceeds that in the real world by about 30% – Knutson’s factor of two is based on the lower uncertainty bounds of observational estimates, not on their medians. However, the future warming projected by CMIP5 models is on average 65% or more above that projected by simple but physically-consistent models with a TCR of ~1.35 K.[xii]

JC Reflections

This is a remarkable paper, and it is gratifying to see the GFDL group making a concerted effort to sort out the multitude of modes and time-scales of natural internal variability. For the first time, we have a coupled global climate model (AOGCM) that seems to be approaching a reasonable treatment of the ocean and the coupled modes between the ocean and atmosphere, including strong multi-decadal variability. The GFDL CM3 model is an outlier in this regard among CMIP5 models, apparently causing some to reject this model in their analyses. Personally, I think this model comes closest to a realistic treatment of internal variability. According to Laepple and Huybers (2014), AOGCM-simulated sea surface temperature variability is systematically smaller than instrumental and proxy-based estimates, increasingly with timescale, the discrepancy reaching two orders of magnitude for tropical variability at millennial timescales.

Further, the study actually addresses the implications of lower TCR values that have been determined from observational analyses using simple energy balance models. Lower TCR values, combined with a greater amplitude of natural internal variability, have substantial implications for attribution of 20th century warming as well as the projections of 21st century climate change.

The spike in global temperatures from the 2015/2016 El Nino has caused some to declare the warming slowdown to be over. This is not necessarily the case; we will need at least another 5 years of observations to determine whether the slowdown is over and warming resumes at a pace of at least 0.2°C/decade, or whether the slowdown will continue for another decade or two.

Update from Tom Knutson:

Tom Knutson has commented by email in relation to this post, as follows :

“One minor clarification is that we are not yet convinced that the mid-twentieth century hiatus and the accelerated warming of the late 20th century are simply forced responses.

Referring to the mid 20th century hiatus (p. 7) we note that “an important role for internal climate variability is suggested, as the CMIP5 forced response deviates from the observed spatial pattern.” As to the late 20the century accelerated warming period, we note (p. 7) that “…if models overestimate the TCR (forced warming rate), internal variability may have also played a significant role in the late 20th century global warming.”

We don’t explicitly say that the mid-20th century hiatus of global mean temperature may have been largely due to internal climate variability cooling offsetting anthropogenic warming (as opposed to aerosols and volcanic forcing providing the offsetting cooling). We probably should have said this explicitly as a possibility.

However, once one accepts that internal variability likely played a substantial role in early 20th century warming, it follows that internal variability may have well contributed similar negative or positive trends to global mean temperature for the mid 20th century hiatus and the late 20th century accelerated warming periods.”

Moderation note:  As with all guest posts, please keep your comments civil and relevant.

—————

End notes

[i] DelSole T, Tippett MK, Shukla J (2011) A significant component of unforced multidecadal variability in the recent acceleration of global warming. J Clim 24:909–926

[ii] Huber, M. & Knutti, R. Natural variability, radiative forcing and climate response in the recent hiatus reconciled. Nature Geosci., 7, 651-656 (2014); Schmidt, G. A., Shindell, D. T., & Tsigaridis, K. Reconciling warming trends. Nature Geosci., 7, 158-160 (2014); Santer, B. D., et al. (2014), Volcanic contribution to decadal changes in tropospheric temperature, Nat. Geosci., 7, 185–189.

[iii] Outten, Stephen, et al. “Investigating the recent apparent hiatus in surface temperature increases: 1. Construction of two 30‐member Earth System Model ensembles.” Journal of Geophysical Research: Atmospheres 120.17 (2015): 8575-8596. The estimated excess positive linear trend relative to RCP8.5 is 0.08 W m−2 dec−1, compared with a linear trend under RCP8.5 of ~0.38 W m−2 dec−1 per the RCP forcings dataset when no post-2005 volcanism is assumed.

[iv] Gregory, J. M., et al. “Small global-mean cooling due to volcanic radiative forcing.” Climate Dynamics (2016): 1-13.

[v] Enfield DB, Mestas-Nunez AM, Trimble PJ (2001) The Atlantic multidecadal oscillation and its relationship to rainfall and river flows in the continental US. Geophys Res Lett 28:2077–2080

[vi] The AMO values used are 5-year moving average residuals from regressing non-detrended annual AMO values5 on the same AR5 forcing time series as is used to regress temperature against, for the same years, so they do not carry any forcing signal. The removed AMO component was given a peak-to-peak amplitude of 0.25 K.Years with volcanic forcing stronger than –0.6 W m⁻² have been excluded. These magnitudes were selected to minimise the regression errors.

[vii] Gregory JM, Forster PM (2008) Transient climate response estimated from radiative forcing and observed temperature change. J. Geophys. Res. 113.

[viii] When multiplied by the standard forcing value of 3.71 W m-2 for a doubling of CO2 concentration.

[ix] Lewis N., Curry J.A., 2014. The implications for climate sensitivity of AR5 forcing and heat uptake estimates. Clim. Dyn. , 45 (2015), 1009–1023. DOI 10.1007/s00382-014-2342-y

[x] https://judithcurry.com/2016/10/26/taminos-adjusted-temperature-records-and-the-tcr/

[xi] See https://niclewis.wordpress.com/appraising-marvel-et-al-implications-of-forcing-efficacies-for-climate-sensitivity-estimates/

[xii] http://www.thegwpf.org/content/uploads/2014/03/A-Sensitive-Matter-download.pdf

 

 

 

272 responses to “Prospects for a Prolonged Slowdown in Global Warming in the Early 21st Century

  1. Reblogged this on Climate Collections and commented:
    “This is a remarkable paper, and it is gratifying to see the GFDL group making a concerted effort to sort out the multitude of modes and time-scales of natural internal variability. For the first time, we have a coupled global climate model (AOGCM) that seems to be approaching a reasonable treatment of the ocean and the coupled modes between the ocean and atmosphere, including strong multi-decadal variability. The GFDL CM3 model is an outlier in this regard among CMIP5 models, apparently causing some to reject this model in their analyses. Personally, I think this model comes closest to a realistic treatment of internal variability. According to Laepple and Huybers (2014), AOGCM-simulated sea surface temperature variability is systematically smaller than instrumental and proxy-based estimates, increasingly with timescale, the discrepancy reaching two orders of magnitude for tropical variability at millennial timescales.

    Further, the study actually addresses the implications of lower TCR values that have been determined from observational analyses using simple energy balance models. Lower TCR values, combined with a greater amplitude of natural internal variability, have substantial implications for attribution of 20th century warming as well as the projections of 21st century climate change.

    The spike in global temperatures from the 2015/2016 El Nino has caused some to declare the warming slowdown to be over. This is not necessarily the case; we will need at least another 5 years of observations to determine whether the slowdown is over and warming resumes at a pace of at least 0.2°C/decade, or whether the slowdown will continue for another decade or two.” — Dr. Judith Curry

  2. How many model analysis papers have been published this year, just to put this single paper into perspective? I think either you or Dr. Curry or both should have a good idea the number without spending much time researching it.

  3. If you take the 30-year temperature and its trend, that has shifted little from 0.17 K/decade since 1980. This is 85% of the 0.2 K/decade mentioned for average CMIP5 models. The plot shows the 30-year temperature (green), the trend for the last 30 years (blue), and the derivative in K/decade (red). It shows no sign of slowing. While 15-year temperatures are all over the place, 30-year temperatures provide a steady background.
    http://www.woodfortrees.org/plot/gistemp/from:1900/mean:120/mean:240/derivative/scale:120/plot/gistemp/from:1900/mean:120/mean:240/plot/gistemp/from:1985/trend

  4. The climb down continues. Here by GFDL. And Trump,has not yet been inaugurated.

    • “The climb down continues. Here by GFDL. And Trump,has not yet been inaugurated.”
      Pretty silly. The paper was submitted on 2 April 2016.

      But it isn’t any kind of climb down. It says there is a 16% chance that unforced variability might counter AGW rise, to limit trend to 0.1K/decade till 2020. That might be a slightly higher probability than others would assign, but it’s still not much. In fact, that chance seems very small, since HADCRUT 4, which they are using, shows trend 0.16 K/dec from 2001 to date. However, they may be talking of trend adjusted to remove ENSO.

      • stevefitzpatrick

        Nick,

        the underlying trend (from the mid 1970’s) seems consistent with a background rate of warming in the range of 0.13C per decade. I agree that accounting for the influence of ENSO gives a clearer picture. Accounting for major volcanoes clarifies further. The models are still too sensitive; most likely about 50% above reality. That might be a good point to start discussions of likely consequences and what mitigation efforts, if any, are economically justified.

      • Have to say that from the non science perspective, 16% is not the sort of odds I’m betting the mortgage on.

        Point to Nick

      • “It says there is a 16% chance that unforced variability might counter AGW rise, to limit trend to 0.1K/decade till 2020.”

        Technically what it says is that there is a GCM that gives that result (whereas others would presumably give a lower chance because they are less accommodating of alternative sources of variation). As Nic notes none do a good job of early 20th C. warming, so you pays your money and takes your chances. Model output is to be tested against reality rather than vice versa.

      • “Combining the CMIP5 ensemble warming rate with internal variability episodes from a single climate model – having the strongest multidecadal variability among CMIP5 models”

        If I read this correctly, they take an individual model that is arguably the most accurate, dilute its results with the drivel from the rest, and derive these silly probabilities.

    • ristvan,
      The paper was submitted back in April 2016, and accepted in October. It is not a ‘climb-down’ that is related to Trump. It may be a ‘climb-down’, or at least a face-saving effort, due to the divergence of models from reality. I do expect the modeler’s budgets will suffer cuts in the next few years, so I hope the more junior modelers are polishing their CV’s.

      • Smart people would shift budgets from bad models to good.

        Political folks will cut everything because they don’t care about science.

      • Mosher, political folks rarely are capable of cutting everything. Too much entrenched interests and plain old inertia. In the case of multiple GCM’s, I think any cutback will be money well saved.

      • stevefitzpatrick

        Steve Mosher,
        Sure, culling of the worst models is something that should have been going on for the last decade; but it hasn’t. Models which are wildly divergent from reality continue to be funded and used to project future warming. I think consolidation of the multiple modeling groups funded by the taxpayers, and shifting of most of that funding to more important work…. like replacement of the failed Glory satellite…. is a reasonable first step. There is a legitimate public need to better define the Earth’s sensitivity to forcing. GCM’s have shown no capability to do that. Better empirical data on aerosols and ocean heat uptake will better define Earth’s sensitivity; that is where most funding should go.

      • Steven Mosher,

        “Political folks will cut everything because they don’t care about science.”

        I see you are trolling with a broad brush today.

      • Maybe it works better as: I see you are trolling with a wide net today.

      • “Political folks” do the research then let “policy folks” re-write the conclusions.

      • Models that happen to fit some favored part of the global temperature trend are not necessarily the best ones. That’s exactly the issue of implicit tuning of GCMs to fit the late 20th century series that many have decried. At the least, goodness of fit ought to look at multiple aspects of model output to reduce the chances of pulling up spurious correlations as indicators of merit.

      • charlieskeptic

        Politics demanded that CMIP5 models more accurately reflect significant Arctic ice declines missed by CMIP3 models, Steve. Adjustments necessary to achieve that (even then falling short) resulted in model failures on other metrics.

        This highlights an issue I previously put forth: If models tuned to late 20th Century temperature trends were to be tuned to 21st Century trends, then the modeled 20th Century would not reflect actuals.

        Even though CMIP5 modelers knew the 2000 through 2005 average temperatures, models did not replicate them. If they had, models would have not replicated late 20th Century values.

        Catch 22 for the modelers.

  5. “We estimate that the warming slowdown (< 0.1 K/decade trend beginning in 1998) could persist…"

    What 0.1 K/decade warming trend? There's no warming trend from 1997-2015

    And the steepest drop in temperature after the 2015-2016 El Nino

  6. That anthropogenic global warming during the 2nd half of the 20th century has been overrated doesn’t get the headlines– even so, many scientists have come to a similar conclusion: the cooling trend is because the activity of the Sun has significantly weakened much like it did several hundred years ago, during the so-called, Little Ice Age, and that does not even include volcanic eruptions.

    • Sorry–didn’t Real Climate have a recent post saying that the Earth was warming despite a cold sun?

      • It’s Baaaack!

        Old Man Winter Returns!

        While last winter was a reprieve from shoveling and high fuel bills, the party is over. According to the 2017 Farmers’ Almanac, “winter is back!”

    • Don’t bet on cooling.

      The sun’s activity is in free fall, according to a leading space physicist. But don’t expect a little ice age. “Solar activity is declining very fast at the moment,” Mike Lockwood, professor of space environmental physics at Reading University, UK, told New Scientist. “We estimate faster than at any time in the last 9300 years.”

      Lockwood and his colleagues are reassessing the chances of this decline continuing over decades to become the first “grand solar minimum” for four centuries. During a grand minimum the normal 11-year solar cycle is suppressed and the sun has virtually no sunspots for several decades. This summer should have seen a peak in the number of sunspots, but it didn’t happen.

      But Lockwood says we should not expect a new grand minimum to bring on a new little ice age.Human-induced global warming, he says, is already a more important force in global temperatures than even major solar cycles.

      • AGW True Believers deny the effects of the Sun on the Earth’s climate in the same way murderers try to obliterate evidence of the crime.

  7. The climate models fail miserably at basic mathematical levels involving solutions that are unique and useful. This is the direct result of the complex non-linear partial differential equations which the models are trying to solve.

    Not sure trying to incorporate more appropriate characterizations of natural phenomena can salvage distint predictions/forecasts/prophesies that are well beyond the inherent capabilities of the climate models.

  8. “‘And the steepest drop in temperature after the 2015-2016 El Nino”
    One needs a little bit more than a one off observation.
    When the other data sets kick in will be the time to use this argument.

  9. Pingback: Prospects for a Prolonged Slowdown in Global Warming in the Early 21st Century – Enjeux énergies et environnement

  10. The 10 years prior to the 97-98 El Niño had a downward trend in warming.

    In the 8 years after the 97-98 El Niño ended surfaces temperatures appear to have launched upward like a rocket.

    Hey, maybe Trump can shut down Wood for Trees!

  11. richardswarthout

    Perhaps GISS can find a home in the Ivy League, much more suited for progressive world government elites. Wood for Trees cannot be blamed for the garbage going in to it’s database. And Trump speaks and acts for the 80% of Red Counties in the USA. Your problems are with those red counties that cover nearly all of America, unseen from inside the Ivy League bubble.

  12. The chances of a fair coin coming up heads on the next toss is 0.5.

    It doesn’t matter whether the last fifty tosses came up heads, tails, or a mixture of the two. The past does not influence the future.

    Attempting to assign a probability to future temperatures on the basis of past temperatures seems a little futile. Given a constant heat input from the Sun, less atmosphere will result in higher direct Sun temperatures – more, lower.

    Likewise, anything that reduces the opacity of the atmosphere will result in a direct Sun temperature increase – reduced particulates, reduced cloud, reduced CO2 etc.

    Increased heat from human activities demonstrably increases temperatures across the globe. More in some places, less in others. Measuring increases is fraught with difficulties – air moves and changes its opacity moment to moment.

    Simple phenomena such as mirages, heat shimmer, or fading radio transmissions demonstrate the ability of the atmosphere to interfere with straight line energy transmission. It doesn’t matter whether the wavelengths are short, long, or in between. All are affected in different degrees.

    It may be that one pole heats while the other cools. Or that hot spots in the crust move around, continuously rearranging already chaotic flow patterns in oceans, and the atmosphere.

    The magnetic poles wander at random – nobody can say why with any certainty. The Earth itself wobbles, gyrates, pulses and generally gibbers about. There appear to be cycles, but nobody knows why, or whether they will continue, or indeed, whether the observed cycles are merely beat frequencies of faster cycles which have escaped notice to date.

    0.2 C per decade is 100 C in 5000. I think not. After 4,500,000,000 years, the surface temperature has dropped. Reality. Maybe one day someone will figure out how to usefully predict future temperatures. At the moment, it seems one guess is as likely as another.

    Cheers.

    • Predicting future temperatures is really easy, the temperatures have followed the same pattern for ten thousand years. Project that same pattern forward and you will get similar cycles of temperature and you will stay inside the same upper and lower bounds.
      This is Occam Razor Simple. What has happened will happen again, and again, and again, etc. . . . . .

    • Attempting to assign a probability to future temperatures on the basis of past temperatures seems a little futile.

      Attempting to assign a probability to future temperatures on the basis of past temperatures seems to really be the only reasonable way to go. What has happened will happen. what has never happened is unlikely to ever happen. Climate alarmism is based on the theory that what has caused climate changes in the past stopped working, for unknown reasons and suddenly man made CO2 is now the only cause of climate change. That is total BS. Nothing has changed and whatever caused past climate changes is still working and changing man made CO2 will never cause enough difference to be measured.

    • Flim Flam Flynn is the grand dragon of Alt-Science

  13. The Little Ice Age was cold because ice extent increased after it snowed more during the Medieval Warm Period.

    Look at the actual real ice core data.

    The Little Ice Age ended because when it is cold, sea ice covers much of the Polar oceans and it does not snow enough and ice depletes and then retreats.

    Look at the actual real ice core data.

    This is Occam Razor Simple stuff.

    • More Alt-Science

      • Come on Steven! It is either science or it isn’t. Nowhere has the insidious use of “alt” as a pejorative been more stupid than in medicine. Something is either medicinal or it isn’t and there is no alternative medicinal choice.

      • “Something is either medicinal or it isn’t and there is no alternative medicinal choice.”

        Boy, that was poorly worded. I meant to say there is no alternative medicine only alternative to medicine or alternative choices in medicine.

      • charlieskeptic

        Oh, crap. Are we now going to be forever subjected to Mr. Mosher’s Alt-Madness meme? Bad enough that he consistently Wanders out of his personal Weed Patch in commenting on stuff he doesn’t understand.

      • Steven Mosher

        “It is either science or it isn’t. ”

        Alt-philosophy

      • Steven Mosher

        I think the Alt Science label will Stick

        This brought to you by

        mr Lukerwarmer
        and
        Mr Treemometer
        and
        Mr, Piltdown Mann

      • Too funny. I am still relatively new to this site and have wondered why Mosher is so relentlessly attacked for being stupid. Tragically, I now understand why.

        People get what they give.

      • Alt-Science is a political term or art. Using it brands you as a political being, rather than a science-guy. Like the folks who write the IPCC Summary for Policymakers and Media. I don’t think that’s where you want to go, Stephen.

      • charlieskeptic

        But that is where Mr. Mosher has always been, Grant. A political being. Glib recitation of others’ dogma.

        At first glance, Mr. Mosher’s Weed Wandering has always seemed to go with the flow. The King’s coin, doncha know?

      • davideisenstadt

        Wisdom courtesy of a guy who works for BEST, a group that has as much in common with Berkley as the Princeton Review has in common with Princeton.
        Hey…I went to Oxford University…
        yeah Oxford University dry cleaners, in Hoboken…they had a two for one sale on dress shirt cleaning.

      • Roger Knights

        Steven Mosher: “I think the Alt Science label will Stick.”

        I do too. But that’s a double-edged sword. (That is, If warming ceases or cooling commences, it will boost the credibility of “alt”-ers in all fields.)

  14. [W]e estimate that the warming slowdown (< 0.1 K/decade trend beginning in 1998) could persist, due to internal variability cooling, through 2020, 2025 or 2030 with probabilities 16%, 11%, and 6%, respectively

    Or in IPPC speak, “It is unlikely that the warming slowdown could persist through 2020, and very unlikely it could persist through 2030″

    Hardly revelatory, surely?

    • VTG

      Yes, you are right, hardly revelatory.

      The take home message probably centres found this;

      ‘The authors take a well-balanced approach to seeking possible explanations for global mean surface temperature (GMST) increasing at a much lower rate from around the turn of the century than over the late 20th century, and consider the possibility that it may continue for some time.’

      So, it is likely that temperatures will increase, albeit at a slow rate, or might remain broadly where they are, but that is not the same as actual cooling.

      So all in all, perhaps an era of relative temperature stability which is a familiar story when looking at our historical climate back to the 11th century. Is this 1975 article the first mention of a ‘Stadium wave?’

      ‘…a third prominent feature of the European climate during the high middle ages was its general stability, there was none of the great storminess which characterised the period after about 1250 (lamb, Britton)

      All of these conditions occurred in various degrees in different parts of medieval Europe and it is at least possible that they occurred in regular patterns shifting from west to east and back again…”

      http://images.library.wisc.edu/WI/EFacs/transactions/WT1975/reference/wi.wt1975.rrring.pdf

      tonyb

      • So, it is likely that temperatures will increase, albeit at a slow rate, or might remain broadly where they are

        You disagree with the study then? It concludes that the likelihood of this happening is only 6%

      • VTG

        No I am agreeing with you as ‘might’ is quite a slight chance. Also remaining broadly where they are surely signifies the same thing

        tonyb

    • VTG, you missed on of the key-sentences of the paper I’m afraid. It goes “Nonetheless, if CMIP5 models overestimate the TCR (forced warming rate), internal climate variability may have also played a significant role in the late 20th century global warming.” As Nic Lewis showed in the fig. 1 a great part of this internal variability in question could be the AMO which had a big increase during 1990…2000. The paper discusses this possibility from model-outputs and is on the right track IMO. BTW: Thank you to Nic Lewis citing my small article here from a month ago. I included the AMO-adjustment (illustrated by Nic) in the C/W -Record – some special freinds like this more than HadCRUT4 ;-)- from 1951 on (Enso, solar and vulcano adjusted by “Tamino”) and got a TCR of 1.34 with a R² of 0.92 with regressing on the AR5-forcings which is very high IMO. The trend slope of the (mostly GHG) forced rate of the GMST converges vs. 0,11 K/ decade.

    • Ahhh, yesssss: We use IPCC climate models that cannot predict the past to predict the future. Ad hoc fudges will fix anything, though.

  15. Pingback: New Paper: Global Warming Hiatus May Last Until 2030 | The Global Warming Policy Forum (GWPF)

  16. [W]e estimate that the warming slowdown (< 0.1 K/decade trend beginning in 1998) could persist, due to internal variability cooling, through 2020, 2025 or 2030 with probabilities 16%, 11%, and 6%, respectively

    Conversely if the warming slowdown (< 0.1 K/decade trend beginning in 1998) persists through 2020, 2025 or 2030, the probability that the models are correctly explaining it is 16%, 11%, and 6%, respectively.

    There is no way we can assign any value to these models' projections for decision making.

    • Conversely …

      This is the point. The higher the probabilty they find in the models for a slowdown (no clue about the real world), the better the chance of saving face in case temps don’t follow the storyline.

    • Again, the warming slowdown did not begin in 1998; it started after 2006, and it ended in after 2012: a duration of ~7 years.

      • Depending on the dataset that you trust there has been no pause or the pause was already set at the very latest in 2003

        El Niño and La Niña do not allow to see the real warming trend while they interfere, so if your trusted dataset shows a pause, you cannot affirm that the pause is over until temperatures clearly show that they are no longer affected by El Niño/La Niña, probably by 2019-20, and are clearly above the pre-El Niño trend line.

      • When it comes to the surface, the chief scientist of RSS does not trust RSS.

        But it doesn’t matter.

        The 30-year trend to 2006 is greater than the 30-year trend to 1999, so it warmed a lot after 1998.

      • When it comes to the surface, the chief scientist of RSS does not trust RSS.

        Right, because we live on a planet where surface temperatures and lower troposphere temperatures are completely independent of each other.

        At the start of my career I learned to distinguish between empirical evidence and a scientist opinion. The first is to be trusted, the second can be shared or argued at will.

        If you want to play cherry picking trend dates, you don’t need me for that.

      • The 30-year trend went up after 1998, but warming stopped in 1998? Good luck with your science.

      • If I was the chief scientist in an organization and I didn’t trust one of its products, I would get a new job, change the product or get a new line of B.S.

      • The 30-year trend went up after 1998, but warming stopped in 1998?

        The warming rate started decreasing on 1998.

        You are not talking about science, just arguing your own cognitive bias. In the process you produce noise.

      • 30-year trends to 1999 versus to 2006:

        to 1999: slope = 0.0176786 per year

        to 2006: slope = 0.0184476 per year

        CK – I don’t pay any attention to people cannot reason their way out of a paper bag. He does not trust RSS for conclusions about the surface of the earth. It’s quite good if you’re a bird.

      • charlieskeptic

        RSS was just great until it started disagreeing in the 21st Century with politicians and their bureaucratic appointees.

    • “There is no way we can assign any value to these models’ projections for decision making.”

      Sure there is.

      We do this all the time.

      • At your peril. Tarot seems equally implausible but much cheaper.

      • Steven Mosher

        Peril?

        Miss me with your Alt-Alarmism

      • It is curious how your comments appear to have been getting shallower and shallower with time. The disintegration of a web character.

      • Javier

        I posted a graphic overwhelmingly directly using Met Office data and a quote from respected scientist Pfister and Mosh dismissed it as alt-science. Curious.

        tonyb

      • Javier and Tonyb,

        I agree. Mosher has lost the plot. His comments are mostly silly and snide. It seems to me he is incapable of objective analysis – he’s just an advocate of climate alarmism.

      • “he’s just an advocate of climate alarmism.”

        Cui bono?

      • Steven Mosher,

        I’m sure you and your ilk use model projections for decision making.

        That’s no doubt why you’ve managed to scale the dizzying heights of nonentity. Tell me all about the model guided decisions you’ve made, and how well they’ve worked.

        You’ve become rich, powerful, influential, and respected by all? You wish.

        I’m content. I back my assumptions. So far, so good.

        Cheers.

      • Tonyb,

        Upton Sinclair — ‘It is difficult to get a man to understand something, when his salary depends on his not understanding it.’

        You will never convince Steven Mosher, don’t matter the evidence. It is not a scientific characteristic, but Steven Mosher is not a scientist anyway.

        I think it is better to ignore him, as he is not contributing anything of value.

  17. Let’s get a bit of perspective on this. The 20th century warming that caused all the fuss was about 0.7 deg C. The climate models predict warming at 0.2 deg C per decade. Earth’s temperature stopped rising around the turn of the millenium. If Earth’s temperature continues to fail to rise until 2030 as the paper suggests, then it will have fallen behind model prediction by 0.6 deg C. What a joke. 0.7 total in a century. 0.6 error in 30 years.
    Earth’s temperature is almost all natural. This paper is just a pathetic attempt to remain in control of the dialogue.

      • I started with the view that (almost all) was natural – but then I falsely concluded that it could not be predicted. I’ve now adapted my view to “pseudo predictable randomness”.

      • Strictly speaking, it is certainly all natural. Joking aside, let me put it this way:

        It is extremely likely that more than half of the observed increase in global average surface temperature from 1951 to 2010 was natural (non anthropohenic). The best estimate of the human induced contribution to warming is practically zero.

    • “The climate models predict warming at 0.2 deg C per decade. Earth’s temperature stopped rising around the turn of the millenium. “
      No. HADCRUT4 trend since 1/2000, 0.16C/dec. GISS 0.194C/dec. NOAA 0.192C/dec. Certainly rising, and not much short of 0.2C/dec.

    • “20th century warming that caused all the fuss”
      Made me think of the the song …
      “we’ve got to sink the Bismark ’cause she’s making such a fuss”
      Reminded me of global warming political hysteria.
      A beautiful state of the art behemoth, faith blind to obsolescence, a committed crew bonded to the cause, meandering rudderless awaiting the inevitable.

  18. March 2015 I wrote:

    “My Global Temperature Forecast: cooling ~0.35C by 2030 years”

    http://scottishsceptic.co.uk/2015/03/17/my-global-temperature-forecast-cooling-0-35c-by-2030-years/

    My gut feeling is that if we don’t see actual cooling by 2020 I would have to have another look.

  19. In many CMIP5 climate models, excessive strengthening of negative aerosol forcing (cooling) during the 3rd quarter of the 20th century may have served to mimic the effect on GMST of missing multidecadal internal variability. And excessive transient climate response (TCR) levels in CMIP5 models may have compensated, in the late 20th century, for their lack of a positive influence of internal variability on GMST.

    That is just a polite way of saying that both aerosol forcing and models have been tuned so as to reproduce past data. Such a procedure could continue indefinitely, but then models would lose any predictive credibility.

  20. “…the satisfactory reproduction of the 1940–1970 hiatus and late 20th century warming by the CMIP5 ensemble could have other explanations.”

    Indeed. This just shows that if you give progressive scientists enough money and enough time to tinker with their incompetent computer models, they can reproduce the past with at least a hint of accuracy..

    Which tells us nothing about the future. No amount of post-scientific modelling will bring about the thermageddon they have been predicting for almost 30 years.

    Maybe the Trump administration will stop funding the climate hysteria machine. I wonder what will all these “scientists” will do if the gravy train suddenly stops?

    • They’ll create another policy-coupled gravy train

    • “…the satisfactory reproduction of the 1940–1970 hiatus and late 20th century warming by the CMIP5 ensemble could have other explanations.”

      Oh yes indeed!

      Tuning the climate of a global model Thorsten Mauritsen et. al 2012

      “2.1. The Tuning Process
      [14] We tune the radiation balance with the main target to control the pre-industrial global mean temperature by balancing the TOA net longwave flux via the greenhouse effect and the TOA net shortwave flux via the albedo affect. …. After a problem has been identified in the coupled climate model, we iterate the following steps until a satisfactory solution is found:

      1. Short runs of single months, or if possible one or more years, with prescribed observed SST’s and sea ice concentration; first with reference parameter settings, and then altered parameter settings.

      2. A longer simulation with altered parameter settings obtained in step 1 and observed SST’s, currently 1976–2005 from the Atmospheric Model Intercomparison Project (AMIP), is compared with the observed climate.

      3. Implement the changes in the coupled model setup to run under pre-industrial conditions and evaluate the altered climate. Frequently, we make small parameter changes in this step to fine-tune the climate, without first revisiting steps 1 and 2. “

  21. The AMO is strongly solar forced.
    The association between sunspot cycles, and the AMO:
    https://www.linkedin.com/pulse/association-between-sunspot-cycles-amo-ulric-lyons?trk=pulse_spock-articles

  22. What they fail to realize is that the internal changes that take place on the the earth which influence the climate are influenced by solar/ geo magnetic field strengths.

    It is amazing how this is left out. In addition the GHG effect has never lead the temperature change in the historical climatic record . Never.

    When solar/geo magnetic fields are very weak they push the terrestrial items that control the climate into a cooling mode and vice versa.

  23. The question to them is what causes internal variability cooling?
    Anyone can say it, but what is the reason and that is what matters.

  24. And I do not mean changes in the AMO/PDO because one has to go further and ask what causes them to change.

    • BINGO! All the modeling in the world cannot change the fact that we fundamentally do not understand the climate system.

      • Steven Mosher

        Alt right is largely defined through opposition

        Alt science is the same..
        some folks deny the data
        some folks deny the methods
        some folks deny that you can study it at all

        Charlie is Alt-Science

      • charlieskeptic

        Judith Curry, in her report to a group of attorneys, used my suggested wording in her introduction:

        “Key summary points:
        ▪ GCMs have not been subject to the rigorous verification and validation procedures that is
        the norm for engineering and regulatory science.
        ▪ There are valid concerns about a fundamental lack of predictability in the complex
        nonlinear climate system.
        ▪ There are numerous arguments supporting the conclusion that climate models are not fit for
        the purpose of identifying with high confidence the proportional amount of natural versus
        human causes to the 20th century warming.
        ▪ There is growing evidence that climate models predict too much warming from increased
        atmospheric carbon dioxide.
        ▪ The climate model simulation results for the 21st century reported by the IPCC do not
        include key elements of climate variability, and hence are not useful as projections for how
        the 21st century climate will actually evolve.
        Climate models are useful tools for conducting scientific research to understand the climate
        system. However, the above points support the conclusion that current GCM climate models
        are not fit for the purpose of attributing the causes of 20th century warming or for predicting
        global or regional climate change on timescales of decades to centuries, with any high level of
        confidence. By extension, GCMs are not fit for the purpose of justifying political policies to
        fundamentally alter world social, economic and energy systems. It is this application of
        climate model results that fuels the vociferousness of the debate surrounding climate models.”

        My contribution was: “GCMs are not fit for the purpose of justifying political policies to
        fundamentally alter world social, economic and energy systems.”

        Bite that Alt-Science, Mr. Mosher. It is clear that you are Wandering in the wrong Weed Patch.

  25. Given that neither the climate models nor the surface temperature statistical models are trustworthy this seems to be a strange dance. Models dancing with models, each tuned to the other.

  26. The AMO is one of most successful teases in the history of physics. Uri Geller would bend spoons. Uri Geller was an expert spoon bender.

    Feynman actually went to investigate. And… Uri bent no spoons.

    The AMO is a spoon bender. Its adherents are mystics… witch doctors.

  27. The basics in the abstract mirror the findings of Scafetta of Duke Univ. from his work 5-10 years ago.

  28. We run a purely statistical model that forecasts global temperatures for the next 5 years. The skill scores are quite high versus persistence.

    Right now, these are the forecasts for the Wood for Trees (WTI)

    1 year -0.28
    3 years -0.06
    5 years -0.14

    This suggests that temps will continue unwinding from el nino in the next 12 months and then rise again… but be net lower in 2021 than now.

  29. Novenber 28, 2017:

    This does not portend cooling…

  30. Thank you for reporting evidence of the slowdown in global warming. If the forces controlling Earth’s climate are ever understood, it will be because the late Professor Paul K. Kuroda (1917-2001) risked his life to preserve nuclear secrets from WWII for the public to eventually re-discover.

    http://joannenova.com.au/2016/11/climate-psalm-101-the-co2-god-causes-quakes-tsunamis-and-volcanoes/#comment-1862836

  31. The greatest AMO scientist of all time:

    Followed closely by Michael E. Mann. Yeah, that Michael E. Mann.

  32. There are efforts to describe “irregular natural cycles”, with varying amplitudes and periods. Perhaps we are looking at a chaotic attractor – it resembles a cycle, and it changes its amplitude and “period” irregularly. Is there an equivalent of harmonic analysis for chaotic attractors?

  33. Pingback: New paper: No Global Warming at least till 2030 - Principia Scientific International

  34. Good essay.

    Thank you to Nic Lewis and Judith Curry for providing links to the full papers, and to the publishers for making them available for download.

  35. I found the evidence proving that water vapor is temperature regulating nightly cooling, just like I’ve been saying for a number of years now.
    Here’s the data

    The key is when the rel humidity goes up to near 100%, the net rad stops being radiated, and when rel humidity drops, it starts radiating again.

    This is a temperature regulation of nightly cooling. This allows any accumulated energy to radiate to space regardless of how much co2 there is. It explains deserts and rain forests. It explains why we had the pause, co2 is not the control knob, water vapor is, and when the oceans cycle changed in the 70’s, it altered how water vapor got distributed to cool.

    https://micro6500blog.wordpress.com/2016/12/01/observational-evidence-for-a-nonlinear-night-time-cooling-mechanism/

  36. Pingback: STUDY: Global Warming ‘Slowdown’ Could Last Another 15 Years | Conservative Mind

  37. “This is a temperature regulation of nightly cooling. This allows any accumulated energy to radiate to space regardless of how much co2 there is. It explains deserts and rain forests. It explains why we had the pause, co2 is not the control knob, water vapor is, and when the oceans cycle changed in the 70’s, it altered how water vapor got distributed to cool.”

    It is hardly a surprise that saturated air (usually but not always fog) at the surface stops LWIR escaping to space from the ground.

    However the fog-top can still see space if not cloudy above.
    Therefore the fog top radiates
    to space, and cools.
    This is how fog thickens vertically with slight stirring mixing down the cold fog top air.

    Also if the fog continues to cool – A long polar night FI – The the fog will “drop out” as frost/rime.
    H2O is a condensing GHG.
    CO2 is not it is NON-condensing.
    If the atmosphere did not have NON-condensing GHG’s there would be nothing to stop the Earth cooling to a GMST of -18C.

    That is why CO2 is most assuredly the “control knob”.
    Without it there would be nothing to stop H2O from precipitating out to the ground.

    https://pielkeclimatesci.wordpress.com/2010/11/03/guest-post-co2-the-thermostat-that-controls-earth’s-temperature-by-andy-lacis/

    • Also if the fog continues to cool – A long polar night FI – The the fog will “drop out” as frost/rime.
      H2O is a condensing GHG.
      …….
      If the atmosphere did not have NON-condensing GHG’s there would be nothing to stop the Earth cooling to a GMST of -18C.

      Wrong, more water is evaporated everyday to replenish what condenses out. Note when the sun is shining mostly on the other hemisphere it does get cold here.

      CO2 is not it is NON-condensing.

      Not at Earth’s current temps, and if it was it would cause warming because water isn’t available to control cooling.

      • “Wrong, more water is evaporated everyday to replenish what condenses out. Note when the sun is shining mostly on the other hemisphere it does get cold

        OK:
        The total radiative forcing of all GHG’s in the atmosphere is ~155 W/m2. This is what raises Earth’s surface temp from 255k to 288k.
        CO2 contributes around 32 W/m2 of that.
        What do you think the lack of 32 W/m2 of heating would do to Earth’s GMST?
        Then, of course, temp is what regulates evaporation.
        So less evaporation of surface water.
        So less radiative forcing from atmospheric WV.
        Feed-back loop.
        Further FB loop comes in as albedo increases due accumulation of snow over NH land masses (first and the SH will follow).
        Further falls in temp, and less evap and so on.

        The hydrological cycle is underpinned by the non-condensing nature of the likes of CO2.
        It is is balance because of that 32W/m2 of extra forcing. Without it WV would steadily and increasingly leak out of the atmosphere.

      • Doesn’t matter that you or Mosher thinks, the outgoing radiation at the surface almost stops near 100%rel humidity, look at the actual measurements. And the optical window is still open to space, I’ve measured it.
        You can get the same data just about any place it’s not 100% rel humidity all the time.

    • “That is why CO2 is most assuredly the “control knob”.”

      No it isn’t.

      If the noise is eliminated, atmospheric CO2 concentration lags temperature. It does not lead it.

      Stop making stuff up.

    • Tony Banton,

      You wrote –

      “It is hardly a surprise that saturated air (usually but not always fog) at the surface stops LWIR escaping to space from the ground.”

      I know you don’t actually believe what you wrote (because nobody could be so stupid as to believe that IR pictures taken from space don’t exist, or that IR from the Sun doesn’t make things hotter), so I am curious why you wrote it.

      Did you really intend to write something else, or do you just parrot whatever the nearest Climate Cultist is mumbling? Only joking – I assume you don’t actually understand physics, or that you find it difficult expressing yourself clearly. I’m always willing to help with either, if you wish.

      Cheers.

  38. Fig. 2. Time evolution of global surface temperature, top-of-atmosphere (TOA) net flux, column water vapor, planetary albedo, sea ice cover, and cloud cover, after zeroing out all the non-condensing greenhouse gases. The model used in the experiment is the GISS 2°´ 2.5° AR5 version of ModelE with the climatological (Q-flux) ocean energy transport and the 250 m mixed layer depth. The model initial conditions are for a pre-industrial atmosphere. Surface temperature and TOA net flux utilize the left-hand scale.

    • The figurative word is model. Since you’re explanation denies the physical evidence it must be rejected.
      Unless it too showed co2 wasn’t the control knob, back before you guys parameterized water evaporation so you could warm them up. It’s run hot ever since, maybe it was correct back then, but it wasn’t politically useful then.

      • “The figurative word is model. Since you’re explanation denies the physical evidence it must be rejected”

        Wot a Sky-dragon slayer rejecting science.
        Well-I-never!

        All based on empirical science my friend.
        Failing accepting/understand that – then just use common-sense.

      • BTW:
        What “physical evidence” is that then?
        Have you found a planet with a WV laden atmosphere but no non-condensing GHG’s to study?
        That’s why we have models.
        Because we only have the one planet to look at that’s anywhere similar.

      • This is what’s wrong with all you climatologists, you have thousands of places around the planet everyday that currently log the daily dynamic response to the Sun, everyone of those is a test sample. only one planet.The planet you have has a wide range of parameters and regular test cycles, all you have to do is pay attention to it.
        I’ll ask you Tony, do you know what the implications are for that drop to near zero outgoing radiation in the middle of the night? and why the ghg effect of co2 doesn’t have any affect on global cooling?

  39. I see that the alt-science team is out in force today

    • The new term for Sky-dragon slayers.
      Even Watts abhors those.

      • Steven Mosher

        its more pleasant than denier

      • Yes, but mosh, consider all the consensus science that has become alt-science (and visa versa). Confirmed false paradigms abound and before too long AGW may join their ranks…

    • Steve even you should be able to understand this chart.

      • Steven Mosher

        micro does alt-science

      • charlieskeptic

        To Mr. Mosher, anything outside his little weed patch is alt-science.

      • At least I do science!

      • Steven Mosher

        “At least I do science!”

        no micro you do alt-science.

        No data
        No methods
        No scepticism
        No Publishing
        No Review
        No uncertainty.

      • No data
        No methods
        No scepticism
        No Publishing
        No Review
        No uncertainty.

        lol Do you understand what it means or not?

      • Steven Mosher,

        Before you “do science”, you need a falsifiable hypothesis.

        And of course, you can publish in all the predatory vanity journals you wish, endlessly play with historical figures, and call yourself a scientist whilst having no scientific qualifications or evidence of any involvement in real science – involving reproducible experimentation and recording results of physical experiments. People might not care what you call yourself, but instead, look to see if you have actually done anything which could be considered scientific.

        In the same way, Gavin Schmidt of NASA may be incorrectly regarded as a scientist by some. I suppose if you believe that a Bachelor of Arts degree and a PhD in mathematics, automatically confers a scientific halo, Gavin is indeed a scientist. He certainly is in his own mind, even though he has produced precisely nothing which advances humanity’s knowledge of anything at all!

        If you have a falsifiable GHE hypothesis involving CO2, you could have produced it, I suppose. Maybe it’s tucked away with the missing heat and Michael Mann’s Nobel Prize.

        Sciencey words maketh not a scientist.

        You seem to deny that climate is merely the average of weather, and then attempt to gallop off in all directions at once. Certainly confusing, but not at all productive.

        Until you have some science to back up your assertions about the future, your speculations remain in the realm of fantasy – where they belong.

        You might care to start with a falsifiable hypothesis, if you want even the slightest respect from those outside the Cult of Climatology. If I sound cynical, it’s because the Cultists are long on promises, warnings, and threats, but conspicuously lacking in useful results.

        Cheers,

      • Steven Mosher

        The grand Dragon of Alt Science— Flym Flamming Flynn

        “Before you “do science”, you need a falsifiable hypothesis.”

        1. actually not. Much of observational science is just that. Observational.
        2. One of the unsolved problems of philosophy is the demarcation problem.. That is, how do you separate science from non science..
        without using science. In short, how do you avoid circular or unfalsifiable
        definitions of science.
        3. The biggest issue is how one defines what it means to be “falsifiable”

        But lets ignore those problems and pretend that we have a “consensus” definition of what science is. We dont, but lets pretend that the problem that has challenged philosophy has been solved.. Lets pretend that debate is settled. Its not. But lets pretend.

        The falsifiable hypothesis is simple. It was articulated in 1896 and again in 1938, and again in 1988. The fact that the science has been correct since 1896 doesnt make it necessarily right. And the fact that its falsifiable hypotheses have not been falsified doesnt make it right either.
        But it is falsifiable nonetheless. For example, if your crazy theory about the earth was true, AGW would be false. If Popes theory was correct, AGW would be falsified. if the Skt dragons were correct it would be falsified. If Bib Tisdale were correct, if Monkton were correct, if bob carter were correct, if Lindzen were correct, if ANY of these people who proffer falsifiable hypotheses were correct, AGW would be falsified.

        In fact, the strongest argument for the falsifiability of climate science is the existence of alternative skeptical theories.

        Focus on that. The fact that some skeptics actually can offer MEANINGFUL alternative theories, demonstrates the falsifiability of the theory they oppose. If you were right about the climate, climate science would be wrong and therefore falsifiable. If its not falsifiable, then your in the same boat.

      • The falsifiable hypothesis is simple. It was articulated in 1896 and again in 1938, and again in 1988.

        And what exacts is that Steve? How much warming is the hypothesis for a doubling?

      • Hey Mosh, still confused? Go look at what’s happening, looking at annual average temperature throws away the good stuff.
        You’d think a rocket scientist like you are would get that.

    • “its more pleasant than denier”

      Maybe not so much more pleasant given the context of the times. The “alt-right” meme seems to have found success, and that meme is used so recklessly as a weapon intended to silence opposing views. “alt-science” can too easily be conflated with “alt-right” and before you know it your more pleasant than denier pejorative comes to mean racist science.

      I wonder,if the “free will is an illusion” advocates achieve, or manufacture, a consensus in neuroscience, should we call the free will advocates deniers, brain cell slayers or alt-science? What if one is simply skeptical of the claims being made?

      Many neuro-scientists rely upon the LIbet studies to justify their advocacy of free will as illusion despite the fact that Libet himself did not reach those conclusions, and the Libet studies, certainly as currently interpreted or consensusly interpreted, deserve a fair amount of skepticism.

      • Steven Mosher

        it’s simple
        Alt right is largely defined through opposition
        Alt science is the same..
        some folks deny the data
        some folks deny the methods
        some folks deny that you can study it at all

        That is why folks have struggled to define what “climate denial” or climate scepticism or what ever you want to call it. Its defined by
        opposition

        Opposition to adjustments
        Opposition to models
        Opposition to the consensus
        Opposition to any and all policy

        Alt-Science works

        As for the collateral associations.. meh.. language is chaotic.. I cant predict how folks will take it

      • JPZ, science is rife with paradigms proven false. (and that fact of the matter alone reduces mosh’s alt-gibberish to naught)…

      • “Alt science is the same”

        When did Newton go from “alt-science” to science? When did Einstein go from “alt-science” to science?

        “That is why folks have struggled to define what “climate denial” or climate scepticism or what ever you want to call it. Its defined by
        opposition”

        If “climate change” was actually about the science, I might be inclined to buy your argument, but it is clearly so political by nature and your argument above is just more of the same politics. Why would anyone struggle to define “climate denial”? What purpose would that serve? Same with defining “climate skepticism”?

        If you argue A and I respond by arguing B and all you do is focus on the fact that I argued with you, then of course all you’re going to see is opposition. What if B happens to be the better argument and how likely are you to even hear that argument when you’re so busily engaged in railing against the opposition?

        “Opposition to adjustments
        Opposition to models
        Opposition to the consensus
        Opposition to any and all policy”

        Number 1 is arguably a scientific dispute, as is number 2, but the latter two arguments are just politics and here you are conflating politics with science.

        “I cant predict how folks will take it”

        Then read harder. Behavior is predictable.

      • charlieskeptic

        Jean Paul, you seem to be a real, thinking human being. Thank you.

      • No Steve, we reject pompous a$$es who really don’t know what they are talking about. And that’s giving you the benefit that you are clueless and really don’t understand what’s happening, as opposed to understanding while spewing disingenuous half truths.

        It is simple though.

    • Steven Mosher,

      Maybe you deny this –

      “The Oxford Dictionaries Online define the scientific method as “a method or procedure that has characterized natural science since the 17th century, consisting in systematic observation, measurement, and experiment, and the formulation, testing, and modification of hypotheses”. It is important to design a experiment so that a hypothesis can be tested. The most important part of the scientific method is the experiment.”

      You may well have another definition of “science”, as Climatological Cultists do when faced with inconvenient reality.

      You claim to have a falsifiable hypothesis – here, there, and everywhere – but you can’t actually bring yourself to state what it is. You might be worried I’d try to find fault with it – another Climate Cult reason for hiding stuff.

      Some scientist – just more claims that anyone who has the temerity to laugh at your unsubstantiated assertions, has to bow to your demands to explain how increased heat results in hotter thermometers. If you don’t understand that, no amount of explanation will help.

      So what is this simple falsifiable GHE hypothesis involving CO2? Or are you just going to continue to attempt to deny, divert and confuse? CO2 can be heated? So can any other gas. The Earth’s surface gets hotter in sunlight? Indeed. It also loses all that heat at night, plus a little more, as Fourier pointed out.

      What have you got? Zip, nada, nothing, nought – a quantity indistinguishable from zero.

      No GHE. No heating due to CO2. No unicorns or luminiferous aether, either!

      Keep believing if you wish. It shouldn’t result in severe personal injury, if you’re careful.

      Cheers.

      • Bugger. Sorry. Forgot to turn the emphasis off. Oh well.

        Cheers.

      • Mike, the difference between rocket science and climate science is that the rocket must rise off the pad or else the rocket science is deemed wrong. (that being the difference between hard science and soft science) The problem that i have with this piece by dr curry is that there seems to be a given that TCR is at least 1.33C without any real way to confirm that. Who knows just how way off (or for that matter, spot on) the luke warmers really are. Absent any real verification, luke warmism is just (borrowing the mosherism here) merely alt-science…

      • p. s., don’t let him “flynn flam” you without hitting him back with a “skull full of mosh” (☺)

      • afonzarelli,

        Thank you for your kind thoughts.

        I’m far too couth to lower myself to Steven Mosher’s level. I even refrain from challenging Steven Mosher to a duel of wits. My father taught me it was not good form to fight an unarmed opponent. Some of the Climate Cultist wouldn’t have the wits to even bring a picture of a knife to a gunfight, let alone a real knife – and that would be useless enough.

        Cheers.

      • charlieskeptic

        Mike! Shame on you!

        You are making an unsafe place for Mr. Mosher. Allusions to knifes and guns are beyond microaggressions. Please offer him huggiebears and crayons as amends.

      • Definition of mosh –

        “Dance . . . in a violent manner involving jumping up and down and deliberately colliding with other dancers . . .”

        Not relevant, but I could probably have some fun, if I had a mind to.

        But I won’t, of course.

        Cheers.

  40. I don’t remember seeing anything by Mosher that takes exception to the essence of this post, which arguably is more on the skeptic side than the alarmist side. I think all skeptics should welcome Steve with open arms to the correct side of the issue. Always better late than never, I say. I’ve seen some indications of his supreme prescience and now he has decided to get ahead of the pack so he can say “I told you so.” Smart guy.

    • How many times have I told Skeptics to DO WHAT NIC LEWIS DOES.

      Ya know.. do science..

      Not Alt science..

      • Like I said, welcome to the enlightened side. A committee is already hard at work planning a reunion for skeptics in December, 2030. A celebration of sorts. RSVP required.

      • “Not Alt science..”

        You’re a leading authority on Alt science, what with your AlGoreithms and data Mannipulation, aren’t you Mosher?

      • Steven Mosher,

        How many times do you have to keep doing the same thing, hoping for a different result? I know you don’t like what dead people with first rate minds wrote.

        “Insanity: doing the same thing over and over again and expecting different results. – Albert Einstein.”

        Maybe you’re a glutton for punishment. How many times does it take until you realise that the only people taking notice of your incessant demands are the credulous, gullible and mentally afflicted. Other Climate Cultists are too busy, trying to demand obedience to the Cult, to pay much attention.

        Endlessly reexamining historical maximum and minimum temperature records tells you nothing about temperature variations, let alone important things such as wind speed and direction, cloud amount, density, height and type. What about precipitation – even whether it was snow, sleet, hail, or rain? Maybe sunshine intensity and duration? Or the H2O content of the air – it gets very hot in the arid tropical desert! Lack of the GHG H2O?

        Too hard? You bet. Keep demanding people “do science”. Scream it at a mirror – you can’t even get yourself to obey your bizarre exhortations! Why should anybody do anything except look bewildered, and maybe laugh?

        No falsifiable GHE hypothesis involving CO2, no scientific credibility. Just stridently proclaimed Cargo Cult Scientism.

        No GHE. Apply heat, things warm up. Remove heat, things cool down. CO2 is no different. Just necessary to maintain human life.

        Cheers.

      • charlieskeptic

        No, Mr. Mosher. Scientists do science. Educated people review scientific literature. Wandering in the Weeds is a trait of pseudo-scientists, such as yourself, apparently.

        As convincingly shown by Dr. Curry, IPCC climate models are bunk: https://curryja.files.wordpress.com/2016/11/climate-models.pdf

        Throwing up dust, as is your wont, is a vain attempt to obscure the actual science. The world is not warming as the real Alt-Science practitioners would have us believe.

        Your dismissal of Bob Tisdale is pathological. His work relating to the lasting climate impacts of the ENSO phenomenon is echoed by at least one watermelon hero.

  41. From this very important paper:

    Prospects for a prolonged slowdown in global warming in the early 21st century

    …The synthetic series in Fig. 5a also show examples of greatly accelerated warming lasting a decade or more, which are evidently spring-back effects as an internal variability cooling episode is followed by a strong internal variability warming episode. The strong warming episodes are further amplified by the underlying forced warming trend. One extreme example shows a warming of almost 1 °C in 15 years—a much greater 15-year warming rate than has occurred in the observations to date (red curves). These spring-back warmings illustrate another important potential consequence of strong internal multidecadal variability as simulated in CM3, and reinforce the need to better understand whether such internal variability actually occurs in the real world. …

    So 1 ℃ over 15 years is not likely, but what would spring-back warming after a slowdown/hiatus/pause in warming look like?

    Maybe like this.

    current decadalized rates:

    15-year – .17 ℃ per decade
    13-year – .22 ℃ per decade
    11-year – .28 ℃ per decade
    9-year – .44 ℃ per decade
    7-year – .50 ℃ per decade
    5-year – .96 ℃ per decade
    3-year- 1.20 ℃ per decade

  42. The spike in global temperatures from the 2015/2016 El Nino has caused some to declare the warming slowdown to be over. This is not necessarily the case; we will need at least another 5 years of observations to determine whether the slowdown is over and warming resumes at a pace of at least 0.2°C/decade, or whether the slowdown will continue for another decade or two.

    The recent ten thousand years has had cycles that are a thousand years with a lot of variation of different length cycles. Five years of observations will not prove anything for the long term cycles. Temperatures will stay inside the same bounds as it has been in for ten thousand years and it will change inside the bounds as it responds to all of the forcing factors that have any influence. Actual data from the records of actual earth data is the best model output we can use. We have simple and complicated models and no one seems to have a model that really works. We must toss out all the models, go back to real data and try to understand it better so that a model can be developed that is better. Some climate scientists have said that all climate models are wrong, but some are useful. I suggest that any model that is based on assumptions that produce output that does not match real data is much, much worse than totally useless.

  43. Harry Twinotter

    At last, a science post to look at. My gut feel is the slowdown (taken on face value) has something to do with heat being subducted into the oceans – this would cause unforced cooling? Agreed, it will take 5 or 10 years of data to confirm and calculate any new trend in Global Mean Temperature.

  44. “could persist, due to internal variability cooling, through 2020, 2025 or 2030 with probabilities 16%, 11%, and 6%, respectively”

    In other words, rapid warming are likely to resume with probabilities of 84%, 89%, and 94% respectively.

  45. Also “tamino” recently thinks about Knutson et al. https://tamino.wordpress.com/2016/12/01/prolonging-a-non-slowdown-in-global-warming/ . He concludes:”I consider its many, and very serious, flaws to be telling evidence that the whole “slowdown” idea was misguided from the very start.”
    He thinks he makes a strong argument that there is not a real discontinuity in the warming with his own adjusted records:

    Unfortunately he investigates his own data not very close. The record goes from 1951 to 2015 and I calculated 15 years long moving trends of the ENSO, Solar and vulcano adjusted GMST of the 4 records ( not NOAA because there is used the same SST-analyses as in GISS):

    As Nic Lewis pointed out: The clearly visible slowdown also in Taminos adjusted records after 2000 is NOT the result of ENSO ( it’s removed by Tamino). It also can’t be due to a reduced radiative forcing, the only probibly possibility seems to be the internal variability, perhaps mostly due to the AMO . So IMO the data set of Tamino is a strong confirmation for one of the key sentences of Knudson et al.:
    “Nonetheless, if CMIP5 models overestimate the TCR (forced warming rate), internal climate variability may have also played a significant role in the late 20th century global warming.”

  46. My basic case for why global cooling may set if the solar criteria I have called for is attained.

    First I want to state my bottom line is each of my solar parameters if reached should effect the amount of radiation reaching the earths surface and or change the albedo, modified by the geo magnetic field of the earth.

    Remember only a very small change in albedo is needed to have major climatic impacts.

    SOLAR PARAMETERS VERSUS CLIMATIC IMPACT.

    SOLAR IRRADIANCE at times of prolonged solar minimums this value declines maybe as much as .2 to .4 % in extreme cases such as the Maunder Minimum . Even a .1 % drop would equate to a global temperature drop of .1c to 2c, due to less overall solar radiation reaching the earths surface.

    UV light – the light just below the visible light wavelength does penetrate the ocean surface to considerable depths which would suggest if there is a decline in UV light that the amounts of energy penetrating the oceans would subside which should result in oceanic cooling. In addition greater sea ice extent would come about resulting in an increase in albedo.

    EUV light – these wavelengths being very short which have been shown to effect not only the amounts of ozone which are created in the atmosphere but the distribution as well. This in turn has been shown to effect the atmospheric circulation by a weakening of the polar vortex which in turn effect both cloud coverage and snow coverage.
    A greater meridional atmospheric circulation resulting from the above causing both global cloud coverage / snow coverage to increase which if indeed is the case would increase the albedo of the earth.

    Note , once the cooling becomes established I believe the atmospheric circulation several years forward evolves to a zonal flow.

    SOLAR WIND when this becomes very low galactic cosmic rays increase and there are studies that suggest this will enable global cloud coverage to increase and if true would again cause the albedo of the earth to increase. There are also studies that suggest an increase in galactic cosmic rays combined with a very low AP index seems to correlate with an increase in major volcanic activity . If true this would cause less solar radiation to reach the surface of the earth and in turn cause periods of rapid cooling.

    AP INDEX a measure of geo magnetic activity if very low months on end if interrupted with a sudden spike studies have shown this could trigger an increase in geological activity(volcanic) which again would cause cooling. The strength of the geo magnetic field in general being a factor in that the weaker that field is the greater will be the impacts of given solar events.

    SOLAR FLUX which measures the energy being emitted from sun if below 90 has a climatic cooling effect which is in evidenced during past prolonged solar minimum periods when this value was sub 90 and general global cooling took place.

    To sum it up I came up with solar parameters which are at or just below typical solar lulls in the 11 year sunspot cycle and reasoned if they persist as is the case with a prolonged solar minimum periods of time would cause the earth to cool because this is what the historical climatic record indicates.

    The geo magnetic field being a relevant player in present times because it is in sync with the weakening solar magnetic field which should cause given solar effects to be magnified to some degree.

    I gave the reasons as to why my solar parameters would effect the climate. You may not agree with what I have suggested but that does not take away from the fact that I gave the reasons and in addition I have come up with specific solar parameter values which I think are needed to accomplish this , along with a duration of time.

    As of today the sun has already started to reached much of my criteria with the exception of the solar wind speed/ AP index which should come about once coronal hole activity subsides which it will if sunspots stay very low going forward which is likely.

    • Salvatore del Prete: My basic case for why global cooling may set if the solar criteria I have called for is attained.

      I am all in favor of “may”, “maybe”, “might” and such, but as written your hypothesis is not disconfirmable. Is this an actual quantitative model that will be or won’t be found to have been accurate 15 years from now? Or is it a maybe that you’ll crow over if it is accurate but modify and restate your confidence in if it is not accurate? I know that sounds harsh, but the IPCC became even more confident in their predictions even as the Earth climate evolved away from the prediction they made for it.

  47. Pingback: New Study: Global Warming ‘Slowdown’ Could Last Another 15 Years | The Global Warming Policy Forum (GWPF)

  48. Probabilities in conclusions require probabilities in assumptions, and the probabilities in assumptions don’t have a strong support in evidence. Written as it has been, the probabilistic “forecast” can’t be confirmed or disconfirmed.

  49. Update from Knutson:

    Tom Knutson has commented by email in relation to this post, as follows :

    “One minor clarification is that we are not yet convinced that the mid-twentieth century hiatus and the accelerated warming of the late 20th century are simply forced responses.

    Referring to the mid 20th century hiatus (p. 7) we note that “an important role for internal climate variability is suggested, as the CMIP5 forced response deviates from the observed spatial pattern.” As to the late 20the century accelerated warming period, we note (p. 7) that “…if models overestimate the TCR (forced warming rate), internal variability may have also played a significant role in the late 20th century global warming.”

    We don’t explicitly say that the mid-20th century hiatus of global mean temperature may have been largely due to internal climate variability cooling offsetting anthropogenic warming (as opposed to aerosols and volcanic forcing providing the offsetting cooling). We probably should have said this explicitly as a possibility.

    However, once one accepts that internal variability likely played a substantial role in early 20th century warming, it follows that internal variability may have well contributed similar negative or positive trends to global mean temperature for the mid 20th century hiatus and the late 20th century accelerated warming periods.”

    • Much as some people would like to ignore what is said in Knutson’s last paragraph, it is very difficult to move past the logic of it.

      • charlieskeptic

        Jeez, people! Years ago, the first time I saw a graph of 20th Century global temperatures, I noticed that temperatures in the late period mimicked those of about 1915-1945. I wondered if similar process could be acting during both periods.

        Then I saw that IPCC climate models did not reproduce the earlier warming. After a little more investigation, I came to the conclusion I was being lied to about the impacts of CO2 on the climate.

        The wheels are beginning to fall off the IPCC’s CAGW bandwagon. Huge institutional efforts are being made to obfuscate the increasing divergence between facts and theory. The more strident the alarmists’ cries, the angrier I get.

        Knutson’s revelations are too little, too late. The scientific fraudsters in government and academia have inflicted too much damage to ever regain credibility.

      • I have fully accepted the logic of it for years. I doubt that you have.

        We are in a period of rapidly accelerated warming, and there is not a single hint in current observations that it is going to relent any time soon.

      • JCH: “We are in a period of rapidly accelerated warming, and there is not a single hint in current observations that it is going to relent any time soon.”

        Heh!

        Yeah, of course we are!

        Just keep right on taking the tablets!

      • Miracle of miracles. First Mosher jumps on the wagon to be part of Team Skeptics and then JCH has a moment of clarity and catches on to the role that natural variability has played in warming. Good on Ya’ JCH, we had every confidence you would catch on eventually.

    • Thanks to Tom Knutsons clarifications we have an explicit confirmation of the thoughts in the post from the lead author of the study IMO. This leads inevitable to a TCR of 1.3 … 1.4. It seems to me that it’s very hard to overlook the evidence for a lower TCR as suggested in the CMIP5 models.

  50. The authors say that CMIP5 and earlier generation climate models’ historical forcing simulations, in the ensemble mean, reproduce well observed late 20th century warming and a hiatus during around 1940–1970, in response to anthropogenic and natural forcings.

    The unadulterated station temperature records show a very deep swing below the 20th century average “during around 1940-1970.” At its global average extreme in 1976 it rivaled in magnitude the 1998 extreme. This was not any “hiatus,” but a sharp natural downswing preceding the subsequent rapid upward swing, whose occurrence is misattributed to AGW. Much of the discussion here is predicated upon accepting UHI-corrupted urban data as a global indicator.

    • Are these scientists or used cars salesmen talking? Are we to believe it was a mere hiatus that caused global cooling scare in the 1970s. This is pure Orwellian doublespeak / historical negationism aka denialism.

  51. UAH November is posted on Roy Spencer’s blog. Looks like 2016 will be a UAH record hot year by a few hundredths of a degree.

    http://www.drroyspencer.com/2016/12/uah-global-temperature-update-for-november-2016-0-45-deg-c/

  52. A TCR in the range 1.33–1.38 K is well below the mean TCR of ~1.8 K for CMIP5 models. This supports Knutson et al.’s suggestion that the transient response of CMIP5 models, on average, exceeds that of the real climate system. Based on these estimates, the average TCR of CMIP5 models likely exceeds that in the real world by about 30%

    The transient climate response (TCR) estimate from both the Figure 1 and the Lewis & Curry 2014 paper are wrong because the temperature record includes the effects of both the urban warming effect and the millennium warming cycle recovery from the Little Ice Age (LIA). Nick seems to have forgotten that a paper by Bjorn Stevens shows that the negative aerosol forcing in the IPCC AR5 report is too large. This result is supported by a recent paper based on the CLOUD experiment at CERN.

    Numerous papers have shown conclusively that the HadCRUT4 and other temperature indexes are contaminated by the effects of economic development. It is unforgivable to ignore this effect. The paper by McKitrick and Michaels 2007 shows that the HadCRUT temperature index trend from 1979 to 2002 over land would decline from 0.27 °C/decade to 0.13 °C/decade if the influence from economic development were removed. A study by Laat and Maurellis 2006 came to identical conclusions. The urban heat island correction on a global basis is 0.042 °C/decade from 1980.

    The temperature history shows an obvious millennium scale temperature oscillation, indicating that natural climate change accounts for a significant portion of the 20th century warming. Using the 2000-year Ljungqvist temperature reconstruction, the global natural recovery from the LIA is estimated at 0.084 °C/century during the 20th century.

    Applying the aerosol correction to the Lewis & Curry 2014 result lowers the transient climate response (TCR) from 1.33 °C to 1.21 °C. Applying the millennium cycle and UHIE corrections reduces the TCR to 0.85 °C, [likely range 0.70-1.1 °C]. The corresponding equilibrium climate sensitivity (ECS) is 1.02 °C, likely range 0.75-1.35 °C]. The greenhouse gas forcings must be compared to only the portion of the temperature rise that was cause by the GHG forcings, not to all the temperature rise for calculating the TCR!

    The FUND integrated assessment model is the only model in the world that estimates both the costs and benefits of warming. The PAGE and DICE models used by Obama’s Interagency Working Group on the social cost of carbon (dioxide) do not include the benefits of CO2 fertilization or the enormous benefits of warming. The sea level rise estimates in these model are far higher than consensus estimates. The FUND model shows that the best estimate of the social net benefit of carbon dioxide emissions, using the corrected ECS estimate, is 16.6 US$/tCO2, which is very beneficial. The likely range is 19.3 to 11.5 US$/tCO2, and it is extremely likely to be more beneficial than 4.3 US$/tCO2.
    See http://www.friendsofscience.org/index.php?id=2205

  53. Alt-science def: The continual statistical re-analysis of the same data in order to obtain a pre-conceived politically biased speculation.

  54. “Combining the CMIP5 ensemble warming rate with internal variability episodes from a single climate model – having the strongest multidecadal variability among CMIP5 models – we estimate that the warming slowdown (< 0.1 K decade-1 trend beginning in 1998) could persist, due to internal variability cooling, through 2020, 2025 or 2030 with probabilities 16%, 11%, and 6%, respectively."

    I think that mischaracterises the AMO. Firstly global surface warming was strongly accelerated from 1995-2005 with the AMO warming. The warming slowdown is the AMO leveling off as it does in its warm mode typically for a number of decades. With a 65-69 year envelope, the actual cooling phase of the AMO won't kick in until the 2030's.

  55. I have judged for some time now that observed and climate model temperature series should be compared on individual model bases and compared for indices of more than temperature trends. I have made such comparisons in the past and here I have linked below to an Excel file in my Dropbox the results of an update of that comparison including the last 2 most recent years of data (with 2016 through November) where the observed temperatures have shown a rapid increase. The comparisons used the RCP4.5 series from CMIP5 and the observed series of HadCRUT4 Cowtan Way Infilled, HadCRUT4, GISS 250 km and GHCN with all data obtained from the KNMI Climate Explorer. The time period was from 1880 to 2016 (through November).

    I have also made the comparison based on the usual combined observed measurements of SAT for land and SST for oceans, and, instead of the commonly used SAT for the models, I have used the combined land SAT and ocean SST. The models definitely show greater trends using ocean SAT than ocean SST and since the observed ocean SAT data is too sparse to make a comparison an overall apples to apples global comparison necessarily requires the method I have used here.

    The comparison made here are for the ar1 of the ar1 model residuals of the residuals from breakpoint linear regressions, the standard deviations of these same residuals and the temperature trends for the time periods 1907-1945, 1992-2016 and 1880-2016. I would normally not restrict the trend estimates to linear regressions but do so here for the convenience and keeping in line with what others have used in these comparisons. The comparisons were made for trends and ar1 using the standard deviations of the trends/ar1 for the climate models with multiple runs and determining how many standard deviations the observed trends/ar1 were from the climate model mean. Where the model only had a single run I used the single run and observed difference divided by the average of the standard deviations from all the models with multiple runs. These comparisons are not formalistic but can be used to compare model to model differences with the observed series. For comparisons of the standard deviations of the AR1 model of the breakpoint regression residuals for the models and observed, the p value of an F-Test was used.

    I also did breakpoint analysis of the these global series. I was particularly interested in whether breakpoints would be reasonably reproduced close to the same breakdates for multiple runs of a climate model. Simulations were used (and not reported here) to determine how much one might expect a breakdate with a given amplitude change to move in time given the red and white noise. I was motivated in this analysis to determine whether linear breakpoint segments could be considered to more or less define deterministic trends in the series. While the results are not totally conclusive there is a strong indication the breakpoint segments might reveal deterministic trends – at least in the climate models where multiple runs are available.

    The compared results in the Excel file show that the climate models nearly all show some rather significant differences from the observed series for one or more of the indices used. The results also point to why it is not instructive to compare observed and model series using a mean of all model runs. I normally have included a Northern to Southern hemisphere warming ratio in my model to observed comparison and will probably include that in the future. Needless to say when it becomes apparent that a model has emulated the observed global temperature trends closely but with very different hemispheric trends that model skill should be downgraded dramatically.

    https://www.dropbox.com/s/rogdoig8d7z8sqw/RCP45_Vs_Observed_CETC.xlsx?dl=0

    • I have updated the Excel file in Dropbox by making a correction to the F.Test p.values in the table. I original used the wrong degrees of freedom. The p.values are increased in value, but not by much.

      I also plan to add the NH/SH warming ratio comparisons to that Excel file today.

    • I have updated the Excel file in my Dropbox to include a RCP 4.5 composite to Observed temperature series comparison for Northern – Southern hemisphere trends for the time periods 1992-2016 and 1880-2016. I used differences instead of ratios to avoid working with a Cauchy distribution – which does have a mean. Again for comparisons where the model had multiple runs I used the mean of those runs minus the observed value divided by the standard deviation of the model runs. Where models had a single run I used that run minus the observed value and divided by the mean standard deviation of all the models with multiple runs.

      While obvious from a statistical point of view, it bears noting, that since there can only be a single realization of the observed values used here, the only reasonable comparison can be to the distribution of individual models that have multiple runs. That comparison, if it were to be rigorous, has to assume that the model runs and the single observed realization belong to the same distribution. It is used here merely as a reasonable comparison to differentiate the model skills. Model runs with only a single run are rather worthless in this comparison in that we cannot estimate from a single run the mean of those runs. If we can assume that the distribution of those runs are narrow and we use larger standard deviation separation of the singlet model and observed values to denote model skill in replicating the observed values, there is some value derived in making the comparison using the average standard deviations of the models with multiple runs – as was done is these comparisons shown in the Excel file.

      I am certain that there are more sophisticated and robust comparison methods for evaluating the skill of the climate models in replicating observed values, but it appears to me in my reading of the literature that there is some hesitation in doing these evaluations and demanding replicate runs of a model in order to make the evaluations more meaningful. Contrarily it appears that many workers in these fields

      • but it appears to me in my reading of the literature that there is some hesitation in doing these evaluations and demanding replicate runs of a model in order to make the evaluations more meaningful. Contrarily it appears that many workers in these fields

        My opinion is that it’s all bad for them, and this (correct me if I’m wrong) only comparing global average temp, and not how the energy is distributed across the globe. After thinking about this, rerunning will likely give the same answer as before. It’s all dependent on the input setting.
        As for different initialization values, the number of unique surface patterns is probably very very large, GAT’s would be a smaller set of ranges out of all of those different surface patterns. in other words a million runs with different surface distributes, might have 400,000 that all average a GAT of 54.0F. I just fear 20 runs isn’t going to tell you that.
        Plus, if you add in the number with sequential years having the same GAT. Someone needs more cpu time, probably why they don’t do more(they use all they can get when they get it, and then wait for the next chance to run all they can again).

      • I have updated the Excel file in my Dropbox, noted above, by converting the p values for the white noise comparisons to standard deviations so that the results are consistent across the table and further added a method with results for attempting to compare skill levels of the individual models in emulating the observed series for 4 different metrics.

        That method involved taking the average distance in standard deviations units the observed series results were from the model estimated mean over the 27 model to observation comparisons listed in the table. This average standard deviation was compared with average standard deviations that were estimated from simulations using 5000 realizations where the probability of the standard deviations listed from the simulations being equal to or greater than that value by chance are determined for 5%, 2.5%, 1.0% and 0.5%.

        Using these probabilities it is rather easy to see that none of the models are very close to emulating the observed results over the metrics used in the comparison and those metrics being red noise, white noise, trends at 3 time periods and NH versus SH warming trends.

        A model skill rating method was devised where the ratio of average of the standard deviations from the tabled model to observed comparisons to the simulated 5% average standard deviation is used. The closer this ratio approaches 1 the better is the model emulation of the observed metric. It appears the that the model with the greater number of multiple runs tends to have the worse skill rating which could indicate either the method is somewhat flawed and biased towards the lower multiple runs or that the result is an indication of selection of best runs from a larger number of runs for the lower multiple run models.

        While the analysis done here for the multiple run models is relatively straight forward with regards to simulations, that for the singlet run models is less clear. While the handling of the difference between the single run result and the observed result with the singlets is valid, the choice of a standard deviation to use, or even whether it is valid to attempt to use a pseudo standard deviation here, is not clear. The standard deviation used as noted previously was from an average of the standard deviations of the multiple run models.

  56. Last sentence in my last post should have read as:

    Contrarily it appears that many workers in these fields are content to either arbitrarily pick and chose from model results for comparisons or to be content that an array of model results encompasses an observed result.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s