The Eruption Over the IPCC AR5

Guest Post by Willis Eschenbach

In the leaked version of the upcoming United Nations Intergovernmental Panel on Climate Change (UN IPCC) Fifth Assessment Report (AR5) Chapter 1, we find the following claims regarding volcanoes.

The forcing from stratospheric volcanic aerosols can have a large impact on the climate for some years after volcanic eruptions. Several small eruptions have caused an RF for the years 2008−2011 of −0.10  [–0.13 to –0.07] W m–2, approximately double the 1999−2002 volcanic aerosol RF.

and

The observed reduction in warming trend over the period 1998–2012 as compared to the period 1951–2012, is due in roughly equal measure to a cooling contribution from internal variability and a reduced 2 trend in radiative forcing (medium confidence). The reduced trend in radiative forcing is primarily due 3 to volcanic eruptions and the downward phase of the current solar cycle.

Now, before I discuss these claims about volcanoes, let me remind folks that regarding the climate, I’m neither a skeptic nor am I a warmist.

I am a climate heretic. I say that the current climate paradigm, that forcing determines temperature, is incorrect. I hold that changes in forcing only marginally and briefly affect the temperature. Instead, I say that a host of emergent thermostatic phenomena act
quickly to cool the planet when it is too warm, and to warm it when it is too cool.

One of the corollaries of this position is that the effects of volcanic eruptions on global climate will be very, very small. Although I’ve demonstrated this before, Anthony recently pointed me to an updated volcanic forcing database, by Sato et al. Figure 1 shows the amount of forcing from the historical volcanoes.

volcanic forcing 1850 2012 Sato

Figure 1. Monthly changes in radiative forcing (downwelling radiation) resulting from historical volcanic eruptions. The two large recent spikes are from El Chichon (1983) and Pinatubo (1992) eruptions. You can see the average forcing of -0.1 W/m2 from 2008-2011 mentioned by the IPCC above. These are the equilibrium forcings Fe, and not the instantaneous forcing Fi.

Note that the forcings are negative, because the eruptions inject reflective aerosols into the stratosphere. These aerosols reflect the sunlight, and the forcing is reduced. So the question is … do these fairly large known volcanic forcings actually have any effect on the global surface air temperature, and if so how much?

To answer the question, we can use linear regression to calculate the actual effect of the changes in forcing on the temperature. Figure 2 shows the HadCRUT4 monthly global surface average air temperature.

hadCRUT4 1850-2012 and gaussianFigure 2. Monthly surface air temperatures anomalies, from the HadCRUT4 dataset. The purple line shows a centered Gaussian average with a full width at half maximum (FWHM) of 8 years.

One problem with doing this particular linear regression is that the volcanic forcing is approximately trendless, while the temperature has risen overall. We are interested in the short-term (within four years or so) changes in temperature due to the volcanoes. So what we can do to get rid of the long-term trend is to only consider the temperature variations around the average for that historical time. To do that, we subtract the Gaussian average from the actual data, leaving what are called the “residuals”:

residual hadcrut4 monthly anomaliesFigure 3. Residual anomalies, after subtracting out the centered 8-year FWHM gaussian average.

As you can see, these residuals still contain all of the short-term variations, including whatever the volcanoes might or might not have done to the temperature. And as you can also see, there is little sign of the claimed cooling from the eruptions. There is certainly no obvious sign of even the largest eruptions. To verify that, here is the same temperature data overlaid on the volcanic forcing. Note the different scales on the two sides.

residual hadcrut4 monthly anomalies plus forcingFigure 4. Volcanic forcing (red), with the HadCRUT4 temperature residual overlaid.

While some volcanoes line up with temperature changes, some show increases after the eruptions. In addition, the largest eruptions don’t seem correlated with proportionately large drops in temperatures.

So now we can start looking at how much the volcanic forcing is actually affecting the temperature. The raw linear regression yields the following results.

R^2 = 0.01 (a measure from zero to one of how much effect the volcanoes have on temperature)

"p" value of R^2 = 0.03 (a measure from zero to one how likely it is that the results occurred by chance) (adjusted for autocorrelation).

Trend = 0.04°C per W/m2, OR 0.13°C per doubling of CO2 (how much the temperature varies with the volcanic forcing)

"p" value of the TREND = 0.02 (a measure from zero to one how likely it is that the results occurred by chance) (adjusted for autocorrelation).

So … what does that mean? Well, it’s a most interesting and unusual result. It strongly confirms a very tiny effect. I don’t encounter that very often in climate science. It simultaneously says that yes, volcanoes do affect the temperature … and yet, the effect is vanishingly small—only about a tenth of a degree per doubling of CO2.

Can we improve on that result? Yes, although not a whole lot. As our estimate improves, we’d expect a better R^2 and a larger trend. To do this, we note that we wouldn’t expect to find an instantaneous effect from the eruptions. It takes time for the land and ocean to heat and cool. So we’d expect a lagged effect. To investigate that, we can calculate the R^2 for a variety of time lags. I usually include negative lags as well to make sure I’m looking at a real phenomenon. Here’s the result:

rsquared forcing and temperatureFigure 5. Analysis of the effects of lagging the results of the volcanic forcing. 

That’s a lovely result, sharply peaked. It shows that as expected, after a volcano, it takes about seven-eight months for the maximum effects to be felt.

Including the lag, of course, gives us new results for the linear regress, viz:

R^2 = 0.03 [previously 0.01]

"p" value of R^2 = 0.02 (adjusted for autocorrelation) [previously 0.03]

Trend  = 0.05°C per W/m2, OR 0.18 ± 0.02°C per doubling of CO2 [previously 0.13°C/doubling]

"p" value of the Trend = 0.001 (adjusted for autocorrelation). [previously 0.02]

As expected, both the R^2 and the trend have increased. In addition the p-values have improved, particularly for the trend. At the end of the day, what we have is a calculated climate sensitivity (change in temperature with forcing) which is only about two-tenths of a degree per doubling of CO2.

Here are the conclusions that I can draw from this analysis.

1) The effect of volcanic eruptions is far smaller than generally assumed. Even the largest volcanoes make only a small difference in the temperature. This agrees with my eight previous analyses (see list in the Notes). For those who have questions about this current analysis, let me suggest that you read through all of my previous analyses, as this is far from my only evidence that volcanoes have very little effect on temperature.

2) As Figure 5 shows, the delay in the effects of the temperature is on the order of seven or eight months from the eruption. This is verified by a complete lagged analysis (see the Notes below). That analysis also gives the same value for the climate sensitivity, about two tenths of a degree per doubling.

3) However, this is not the whole story. The reason that the temperature change after an eruption is so small is that the effect is quickly neutralized by the homeostatic nature of the climate.

Finally, to return to the question of the IPCC Fifth Assessment Report, it says:

There is very high confidence that models reproduce the more rapid warming in the second half of the 20th century, and the cooling immediately following large volcanic eruptions.

Since there is almost no cooling that follows large volcanic eruptions … whatever the models are doing, they’re doing it wrong. You can clearly see the volcanic eruptions in the model results … but you can’t see them at all in the actual data.

The amazing thing to me is that this urban legend about volcanoes having some big effect on the global average temperature is so hard to kill. I’ve analyzed it from a host of directions, and I can’t find any substance there at all … but it is widely believed.

I ascribe this to an oddity of the climate control system … it’s invisible. For example, I’ve shown that the time of onset of tropical clouds has a huge effect on incoming solar radiation, with a change of about ten minutes in onset time being enough to counteract a doubling of CO2. But no one would ever notice such a small change.

So we can see the cooling effect of the volcanoes where it is occurring … but what we can’t see is the response of the rest of the climate system to that cooling. And so, the myth of the volcanic fingerprints stays alive, despite lots of evidence that while they have large local effects, their global effect is trivially small.

Best to all,

w.

PS—The IPCC claims that the explanation for the “pause” in warming is half due to “natural variations”, a quarter is solar, and a quarter is from volcanoes. Here’s the truly bizarre part. In the last couple decades, using round numbers, the IPCC predicted about 0.4°C of warming … which hasn’t happened. So if a quarter of that (0.1°C) is volcanoes, and the recent volcanic forcing is (by their own numbers) about 0.1 W/m2, they’re saying that the climate sensitivity is 3.7° per doubling of CO2.

Of course, if that were the case we’d have seen a drop of about 3°C from Pinatubo … and I fear that I don’t see that in the records.

They just throw out these claims … but they don’t run the numbers, and they don’t think them through to the end.

Notes and Data

For the value of the forcing, I have not used the instantaneous value of the volcanic forcing, which is called “Fi“. Instead, I’ve used the effective forcing “Fe“, which is the value of the forcing after the system has completely adjusted to the changes. As you might expect, Fi is larger than Fe. See the spreadsheet containing the data for the details.

As a result, what I have calculated here is NOT the transient climate response (TCR). It is the equilibrium climate sensitivity (ECS).

For confirmation, the same result is obtained by first using the instantaneous forcing Fi to calculate the TCR, and then using the TCR to calculate the ECS.

Further confirmation comes from doing a full interative lagged analysis (not shown), using the formula for a lagged linear relationship, viz:

T2 = T1 + lambda (F2 – F1) (1 – exp(-1/tau)) + exp(-1/tau) (T1 – T0)

where T is temperature, F is forcing, lambda is the proportionality coefficient, and tau is the time constant.

That analysis gives the same result for the trend, 0.18°C/doubling of CO2. The time constant tau was also quite similar, with the best fit at 6.4 months lag between forcing and response.

<strong>

In this case it’s the Sato paper, which provides a dataset of optical thicknesses “tau”, and says:

The relation between the optical thickness and the forcings are roughly (See “Efficacy …” below):

instantaneous forcing Fi (W/m2) = -27 τ
adjusted forcing Fa (W/m2) = -25 τ
SST-fixed forcing Fs (W/m2) = -26 τ
effective forcing Fe (W/m2) = -23 τ

And “Efficacy” refers to

Hansen, J., M. Sato, R. Ruedy, L. Nazarenko, A. Lacis, G.A. Schmidt, G. Russell, et al. 2005. Efficacy of climate forcings. J. Geophys. Res., 110, D18104, doi:10.1029/2005/JD005776.

Forcing Data

For details on the volcanic forcings used, see the Sato paper, which provides a dataset of optical thicknesses “tau”, and says:

The relation between the optical thickness and the forcings are roughly (See “Efficacy …” below):

instantaneous forcing Fi (W/m2) = -27 τ
adjusted forcing Fa (W/m2) = -25 τ
SST-fixed forcing Fs (W/m2) = -26 τ
effective forcing Fe (W/m2) = -23 τ

And “Efficacy” refers to

Hansen, J., M. Sato, R. Ruedy, L. Nazarenko, A. Lacis, G.A. Schmidt, G. Russell, et al. 2005. Efficacy of climate forcings. J. Geophys. Res., 110, D18104, doi:10.1029/2005/JD005776.

(Again, remember I’m using their methods, but I’m not claiming that their methods are correct.)

Future Analyses

My next scheme is that I want to gin up some kind of prototype governing system that mimics what it seems the climate system is doing. The issue is that to keep a lagged system on course, you need to have “overshoot”. This means that when the temperature goes below average, it then goes above average, and then finally returns to the prior value. Will I ever do the analysis? Depends on whether something shinier shows up before I get to it … I would love to have about a dozen bright enthusiastic graduate students to hand out this kind of analysis to.

I also want to repeat my analysis using “stacking” of the volcanoes, but using this new data, along with some mathematical method to choose the starting points for the stacking … which turns out to be a bit more difficult than I expected.

Previous posts on the effects of the volcano.

Prediction is hard, especially of the future. 

Pinatubo and the Albedo Thermostat

Missing the Missing Summer

Volcanic Disruptions

Dronning Maud Meets the Little Ice Age 

New Data, Old Claims about Volcanoes 

Volcanoes: Active, Inactive and Interactive

Stacked Volcanoes Falsify Models

463 thoughts on “The Eruption Over the IPCC AR5

  1. I thought the plan was to hide the warming in ocean depths and under ice sheets where no one could monitor it with existing taxpayer funded systems like satellites and ocean buoys.

  2. Is it possible that the “Year Without a Summer” in 1816 was not really caused by the eruption of Mount Tambora in Indonesia after all?

  3. Willis says:

    let me remind folks that regarding the climate, I’m neither a skeptic nor am I a warmist.
    I am a climate heretic.

    I agree that “skeptic” is too mild a term. We aren’t doubters, we’re disbelievers / dissenters / deviationists / contrarians (the term I prefer). Or “hot-air heretics.”

  4. Interesting,

    Just be sure you set the correct boundary conditions when looking at your thermodynamic balance. I noticed in your previous work you listed “rain” as a potential cooling mechanism. I have heard from an uneducated caller on a radio show that more lakes means global cooling (because it is cooler near the lakes)

    This kind of thinking isn’t really helpful since the correct boundary conditions are set about 1/4 of a mile above the top of the atmosphere. If the energy going into that sphere containing the whole earth and all of it’s functions is more than the energy leaving that sphere then the planet is warming.

    There is no other scientific reality. You cannot put energy into an object without it warming, you cannot take energy from that object without it cooling.

  5. With 2/3 of the planet covered by water, perhaps volcanoes’ greatest impact is the result of injecting heat into the oceans. Question here is whether this would result in net warming, or cooling (depending on location) due to positive cloud cover/precip? Any studies on this?

  6. Willis – please can you explain how you translate W/m^2 into “per doubling of CO2″. The IPCC report would presumably have two separate measures, direct and indirect (ie. without and with feedbacks). Also, why the figures you quote relate to ECS not transient sensitivity (I didn’t get that bit).
    TIA

  7. Willis, I’m a little troubled by the units you’ve chosen: 0.18°C/doubling of CO2. It’s not clear what this means or where it comes from.

    I presume it is converting W/m2 into some ‘carbon equivalent’ but what value is used (ie assumed) to be the effect of CO2 doubling.

    If you want to use such a unit it would be good to explain what it means.

    It would be interesting to do your processing for tropical and extra-tropical regions separately. I found significantly different responses and since your hypothesis is based largely on the climate response in the tropics, this is where you need to look for confirmation or falsification.

    My volcano stack processing showed exactly the kind of overshoot you are referring to in the tropics but not in extra-tropical regions. The small ex-tropical effect is also temporary. Due to units I don’t know how this compares quantitatively. But in form, it is in agreement.

    You may well find better correlations to the ‘small’ effect in the split analysis since the tropics would tend to reduce the correlation.

    It would presumably be very little effort to use regional temp time series and re-run the scripts.

  8. Jim S says:Don’t volcanoes also emit large quantities of CO2 into the atmosphere? Is this taken into account?

    Largeness is relative. In view of what we chuck out and the natural annual carbon cycle volcanoes are a fart in the wind.

  9. Jim S, the emissions from individual eruptions is pretty much negligible. Overall volcanoes emit around 1% of the amount from fossil fuels.

    Maybe Eschenbach has written about it before, but I’m a bit confused on how he can reconcile “I hold that changes in forcing only marginally and briefly affect the temperature. Instead, I say that a host of emergent thermostatic phenomena act quickly to cool the planet when it is too warm, and to warm it when it is too cool” with the existence of ice age cycles. Whatever thermostat the Earth has doesn’t seem all that good.

  10. theyouk says:With 2/3 of the planet covered by water, perhaps volcanoes’ greatest impact is the result of injecting heat into the oceans.

    Look at my volcano stack graphs. They are interlinked and have some commentary. I worked on SST rather than land+sea indices but I did comment of the possible impact of land/sea ratios and the way land rate of change is twice ocean rate of change (dT/dt) and how this affects NH / SH differences.

  11. theyouk:

    At September 22, 2013 at 10:16 am you ask

    With 2/3 of the planet covered by water, perhaps volcanoes’ greatest impact is the result of injecting heat into the oceans. Question here is whether this would result in net warming, or cooling (depending on location) due to positive cloud cover/precip? Any studies on this?

    What mechanism would cause volcanos to “inject heat into the ocean”?
    There can be no studies of something which is not known to exist.

    And the “injection” could not happen in the warm tropics because there is a maximum sea surface temperature of 305K. When the sea surface is at that maximum temperature then additional heating (from any cause) results in the ocean COOLING.

    The effect was first reported by Ramanathan & Collins in 1991.
    Their paper is Ramanathan v & Collins W, ‘Thermodynamic regulation of ocean warming by cirrus clouds deduced from observations of the 1987 El Niño’, Nature 351, 27 – 32 (02 May 1991) doi:10.1038/351027a0
    Its Abstract says
    Observations made during the 1987 El Niño show that in the upper range of sea surface temperatures, the greenhouse effect increases with surface temperature at a rate which exceeds the rate at which radiation is being emitted from the surface. In response to this ‘super greenhouse effect’, highly reflective cirrus clouds are produced which act like a thermostat shielding the ocean from solar radiation. The regulatory effect of these cirrus clouds may limit sea surface temperatures to less than 305 K.

    In other words, the effect they found is that increased heating of tropical ocean increases evapouration to increase cover by cirrus clouds which shield the surface from solar heating. This shielding sets a limit of 305K to maximum surface temperature.

    But clouds don’t stay in one place. Therefore, when a region of the surface has temperature of 305K, then additional heating (from any cause) increases cirrus clouds which spread to shield surrounding regions. Thus, the effect does not merely set a maximum temperature: it induces a drop in surface temperature of the surrounding ocean regions when surface heating is increased.

    The R&C Effect can induce a fall in surface temperature when surface heating is increased. And the Eschenbach Effect does that, too.

    As Willis says, these effects – and any similar effects – would provide a homeostatic control to global temperature.

    Richard

  12. The large climate sensitivities claimed by IPCC not only give large negative temperature excursions but also very long times to return to normal. (See for example Lindzen http://link.springer.com/article/10.1140/epjp/i2012-12052-8#page-1). This must be true because climate senitivity is basically the climatic relaxation time divided by the effective specific heat of the climate system. So if IPCC were right, we still have not recovered from Krakatoa (1883) and Katmai (1912), not to mention Pinatubo (1992). In fact the IPCC climate sensitivity is so large that a normal century’s eruptions would keep the earth about a deg C cooler than otherwise and we would never be far from a volcanic “winter” from a few major eruptions.

    This alone should raise eyebrows about the plausibility of the IPCC values of climate sensitivity.

  13. Thomas: ” Whatever thermostat the Earth has doesn’t seem all that good.”

    what happens at glaciation and deglaciation is clearly different from what happens in between. There is apparently two stable states ( attractors ) for the climate system. A positive feedback seems to make it snap form one state to the other. We don’t really know what triggers the change-over.

    Assuming Willis is basically correct there are limits to the tropical storms range as a feedback mechanism. It cannot go beyond totally clear skies or fully cloud covered tropics. May be when it hits the rails the climate state flips?

    I don’t see glaciation as being a major argument against what Willis is proposing.

  14. ‘I say that the current climate paradigm, that forcing determines temperature, is incorrect. I hold that changes in forcing only marginally and briefly affect the temperature. Instead, I say that a host of emergent thermostatic phenomena act quickly to cool the planet when it is too warm, and to warm it when it is too cool.’

    I think you will find plenty of agreement there. The statement above seems to mirror a similar point made by Dr. Richard Lindzen. I can’t recall his exact wording but he seemed to say, that in the end, Global Warming distilled down to a philosophical question: Does one believe that a long running natural system amplify perturbations, or would it minimize perturbations?

    I think the answer is obvious simply from the description, ‘long running natural system.’ How could the Earth possibly have been a long running natural system if it stampedes away into nightmare land with every little deviation away from what nobody knows is really the norm? I think the planet has a lot more things to worry about then us little humans. And I think we have a lot more things to worry about too. And to take joy in. Welcome back from your trip Willis Eschenbach. I took joy in it.

  15. While volcanos cause cooling initially, some may actually cause warming afterwards.
    Have a look at the Temperature of the lower statosphere ftp://ftp.ssmi.com/msu/graphics/tls/plots/rss_ts_channel_tls_global_land_and_sea_v03_3.png
    You can see a clear boom and bust signal, probably due to the SO2 stripping H2O out of the stratosphere from where it cannot be easily replenished.
    Of course multidecadal volcanic cooling of the stratosphere may have nothing to do with tropospheric temp, but its hard to know when nobody has ever looked.

  16. Tom McCord says:
    September 22, 2013 at 9:58 am

    IMO evidence supports the contention that The Year Without a Summer was indeed caused by Tambora, with the effects amplified by having occurred during the already cold-stressed Dalton Sunspot Minimum.

    Earlier in the Little Ice Age, the Pacific volcano Kuwae had similar effects after its c. 1452 eruption, shortly before the Spoerer Minimum. Also, an eruption of Huaynaputina in Peru is blamed for the severe Russian famine of 1601–03. The 1783 eruption of Laki in Iceland caused thousands of deaths in Europe. The latter two events fell before & after the Maunder Minimum.

  17. richardscourtney says: … these effects – and any similar effects – would provide a homeostatic control to global temperature.

    Thanks, Richard, I was not aware of that paper. Thunderstorms have more to do with convection and evaporation the “super greenhouse effect” but I suppose they had to boost AGW keyword count in order to get published.

    the article is available through ReadCube (which is only just above unusable in my browser, I’ll perisist)

    http://www.readcube.com/articles/10.1038/351027a0?locale=en

  18. Thomas says:
    September 22, 2013 at 10:30 am

    During glacial epochs, climate seems to have two (or three) phases, ie long glacial & shorter interglacial, with possibly an even colder phase within the glacial, associated with Heinrich Events. The Last Glacial Maximum is an instance of the possible third “steady state”, shorter than interglacials & perhaps just extra-cold D-O events.

  19. In all probability, the really huge volcanic eruptions of the past 2,000 years – in 542 and 1816 – did have a significant effect on temperature, but these were eruptions an order of magnitude greater than the Pinatubo one in 1992

  20. “I ascribe this to an oddity of the climate control system … it’s invisible. For example, I’ve shown that the time of onset of tropical clouds has a huge effect on incoming solar radiation, with a change of about ten minutes in onset time being enough to counteract a doubling of CO2. But no one would ever notice such a small change.”

    Well, of course it is invisible and it will remain so until some climate scientists go into the natural environment, maybe the Virgin Islands, and start measuring the thermostatic control phenomena. Because climate scientists are so very averse to empirical research, they are not going to do the necessary work, at least not this generation of climate scientists.

  21. Maybe the 17 yr pause in global warming is due to a time-warp teleconnection to past volcanic eruptions. Their cooling was just in the pipeline until recently.

  22. Willis, once a mere skeptic myself I became a climate heretic by simply reading your simple and elegant articles on homeostatic mechanisms for resisting changes up or down in earth temperature. Your demonstrating that there is an upper limit to SST of ~31C from the data clinched it for me. I came at the stability of climate from the long geologist’s view in which global temperature fluctuations over hundreds of millions of years has an amplitude of only 3-5K! If I now understand your ideas, the upper limit for SST is a sharp 31C – the sea surface won’t surpass this because of the governor mechanism. The lower limit is less firm in that with all the thunder clouds gone with cooling, heating depends essentially on insolation which may be insufficient to bring it back up toward 31C (Milankovitch cycles, and other). This would mean a slowing of the tropical – polar thermal exchange causing greater ice accumulation in a cooling system. It is my contention that a string of thermometers along the ITCZ is all we need to tell us the direction of significant changes in climate.

    I think the groundwork you have laid will lead to the slam dunk physics of the climate system. An answer to the question why specifically 31C is the upper limit of SST will open a floodgate of understanding. Yes, to be a skeptic is good for science and good fun where you meet the interesting proportion of fellow humans. But to blow past it all and put forth an entirely new paradigm is even more exciting and interesting. There will be a lot of people who won’t like you but that is a good way to help winnow out those on whom you shouldn’t waste too much time.

  23. Willis, I too, am a heretic. The fact that our planet continues to be water based within a relatively narrow temperature range and its history of excursions towards an ice ball and back again seems to bear this out.

    The current global warming mechanisms may be valid (though I doubt some of it) but the models completely underestimate the way our planet can restore climate stability. The climate scientists are too busy trying to defend their models to understand this.

  24. Greg Goodman:

    In your reply to me at September 22, 2013 at 11:04 am you say

    Thunderstorms have more to do with convection and evaporation the “super greenhouse effect” but I suppose they had to boost AGW keyword count in order to get published.

    For sake of clarity, I point out that the Ramanathan & Collins (R&C) effect induces cirrus not thunderstorms. They argued – initially against much opposition which their finding withstood – that when sea surface temperature reaches 305K the induced evapouration rate is so great that warm air rises to lift evapourated moisture so high that cirrus formation occurs. This cirrus sets the maximum surface temperature by reflecting sunlight so it cannot reach the surface.

    The Eschenbach effect raises heat from the surface to high tropospheric altitude where it radiates to space. It starts to operate at temperatures below 305K.

    They are very different – and complimentary – mechanisms.

    Richard

  25. Once again, the IPCC has an hypothesis, backed up by simple lab experiments, that the addition of certain chemicals to the atmosphere should have an effect on global temperatures. Their response when it is pointed out that the effect is not as pronounced as predicted is to say that the
    science is basic and settled.
    Good to see Willis that your time spent in the Old Country has not blunted your analytical skills.

  26. “Instead, I say that a host of emergent thermostatic phenomena act
    quickly to cool the planet when it is too warm, and to warm it when it is too cool.”

    But what sets the temperature (or rather system energy content) other than the strength of the gravitational field, the mass of the atmosphere and top of atmosphere insolation ?

    AGW theory relies on compositional changes altering that baseline system energy content.

    Compositional changes only result in circulation adjustments which prevent a change in the baseline level hence your emergent thermostatic phenomena but globally rather than just in the tropics.

  27. jai mitchell says:… If the energy going into that sphere containing the whole earth and all of it’s functions is more than the energy leaving that sphere then the planet is warming.

    Proven counter-example of a closed system for which that is not true:
    A steam engine with more heat provided doesn’t necessarily get any hotter or ‘emit more heat’. It instead goes faster.

    That is: The energy doesn’t need to come out – it can be converted into work. Just because you didn’t make the heat cycle with steel tubing doesn’t mean it isn’t a Carnot heat engine.

  28. Thanks Willis,

    I’m not a scientist, so please pardon my ignorance, but in figure 4 are the two scales used equivalent? Does forcing of 4 w/m2 equal 0.8C in temperature anomaly? If so, what are the assumptions behind this analysis? Also, doesn’t figure 2 illustrate the temperature anomalies, rather than the actual surface temperatures? This seems like an important distinction which is often missed, but maybe I’m wrong.

  29. ” It is my contention that a string of thermometers along the ITCZ is all we need to tell us the direction of significant changes in climate. ”

    Not necessary.

    The net latitudinal position after stripping out seasonal variation is all one needs.

    Towards the north pole = warming.

    Towards the equator or the south pole = cooling.

    It would be helpful to ascertain the neutral position first though.

  30. Volcanic and Solar Forcing of Climate Change during the …

    http://www.meteo.psu.edu/holocene/public…/Shindelletal-jclim03-preprint.pdf‎

    This study concludes volcanic eruptions do have an effect on the climate.
    In regards to the IPCC saying small volcanic eruptions caused the temperature rise to slow , they are wrong if one looks at a volcanic aerosol optical thickness graph which shows very low values since the Mt. Pinatubo eruption.

    If one goes to Dr. Spencer’s website one will see a very clearly define temperature drop in global temperatures following the Mt. Pinatubo eruption in the early 1990’s. This temperature drop was despite an associated El Nino at the time, which clearly shows a large volcanic eruption will cause a substancial drop in global temperatures for a short time following the eruption.

  31. A very nice piece of work. I guess I can go back and add in the data I lopped off the beginning of the SST record to eliminate the effects of Krakatoa.
    All the best

  32. 3. Impact of Volcanic Eruptions
    The global annual average surface temperature response to volcanic
    eruptions is cooling, resulting from increased absorption and reflection of
    incoming shortwave radiation by stratospheric aerosols. Averaging all years of the
    simulations together, the mean annual average cooling was -0.35 C for the
    periodic Pinatubo eruption, -0.77 C for the periodic Tambora 2P eruption, -1.09
    C for the periodic Tambora 3P eruption, and -0.44 C for the observed 1959-1999

    The above is from the study I sent inmy previous post.

  33. Volcanic and Solar Forcing of Climate Change during the Preindustrial Era
    Drew T. Shindell1,2, Gavin A. Schmidt1,2, Ron L. Miller1,3, and Michael E. Mann4
    1NASA Goddard Institute for Space Studies, New York, NY 10025, USA.
    2Center for Climate Systems Research, Columbia University, New York, NY
    10025, USA.
    3Department of Applied Physics and Applied Mathematics, Columbia University,
    New York, NY 10025, USA.
    4Department of Environmental Sciences, University of Virginia, Charlottesville,
    VA 22902, USA.

    If one wants to google the study.

  34. If Willis is correct (and this seems very plausible based on his data) does this not mean it likely that man-made aerosols have a lower than generally assumed impact on temperature?

    I also wonder about the impact of volcanic ash on ice extent and temperatures in the polar regions. There is significant evidence that man-made deposits of black carbon will increase the thermal absorption of snow and ice accelerating the melt. Would volcanic ash not have an impact there?

    There might be some mileage in correlating those volcanic events with polar temperatures specifically.

  35. “The issue is that to keep a lagged system on course, you need to have “overshoot”. This means that when the temperature goes below average, it then goes above average, and then finally returns to the prior value”

    I have already proposed such a scenario.

    When the Earth gets warm enough for outward longwave to exceed solar incoming then there is an excess of energy going out and the system cools.

    When the Earth cools so that outward longwave is less than solar incoming then there is an excess of energy coming in and the system warms.

    The mediating mechanism in the atmosphere is the global convective air circulation which adjusts as necessary to maintain ToA energy balance..

    The importance of the water cycle is that its heat shifting efficiency is so great that it does most of the work that would otherwise need to be done by changes in the speed of the convective circulation.

    The existence of the water cycle means that the necessary adjustments need not be as violent as would otherwise be necessary.

    It is atmospheric pressure which sets that top limit for ocean surface temperatures.

  36. The evidence is that increased volcanic activity during past prolonged solar minimums enhanced the temperature declines the globe experienced at those times periods

    Further the case can be made of a solar/volcanic correlation. Many studies showing increased volcanic activity being associated around solar minimum periods.

  37. richardscourtney says:
    September 22, 2013 at 11:48 am

    Hi Richard.

    Hans Jelbring is correct in general terms as regards his ‘Wind Driven Climate’ but I have topped and tailed the whole climate story so as to incorporate a scenario that also accommodates the ideas of Willis, the data of Bob Tisdale, the latest upper atmosphere data and many other contributions from many other commentators.

  38. Peter Miller says:
    September 22, 2013 at 11:06 am

    Volcanologists recognize or suspect three or four VEI7 eruptions in the past 2000 years: Tambora (1815), Rinjani (13th century, but unconfirmed), Baekdu (969) & Taupo (180-230). The candidates for the c. 540 event, Rabaul & Ilopango, are currently rated as VEI6. I might have missed some.

  39. “But what sets the temperature (or rather system energy content) other than the strength of the gravitational field, the mass of the atmosphere and top of atmosphere insolation ?”

    That is important along with atmosphere composition. I realize that your website concludes that atmospheric composition is not important, but simulations of atmospheric columns differ from that conclusion. See hitran results in table 2: http://web.archive.org/web/20121226202653/http://www.john-daly.com/forcing/hug-barrett.htm

    The important point that we can all agree on for this thread is that there are various thermostatic effects both local to the tropics and global that limit the planet’s warmth. Short term that includes convection and clouds. Long term thermostats include heat transport to the poles both by meridional flow in the atmosphere and by oceanic currents. For the latter, a good example is ice-free waters that allow more cold bottom water to form in the Arctic and flow into the Atlantic.

    Cloud thermostats also limit cooling. Ice increases limit cooling as well by preventing heat loss and limiting cold bottom water formation. It’s not as if the planet is finely tuned regarding energy balance, but the thermostatic mechanisms are nonlinear and kick in more as the planet deviates from the baseline. That also determines the significance or insignificance of raised CO2 levels.

  40. John L. Casey1

    Released for world wide web (www) distribution on Monday, March 1, 2010.

    [1] An independent review of historical records was performed for 350 years of global volcanic activity

    Link

    [Salvatore, please do not post such a huge chunk of text, I’ve replaced it with a link. We can all read, so make your point and post a link. In this, you’ve not even commented on why you posted the link. -w.]

  41. Tom McCord says:
    September 22, 2013 at 9:58 am

    Is it possible that the “Year Without a Summer” in 1816 was not really caused by the eruption of Mount Tambora in Indonesia after all?

    I suppose it’s possible. The Sato dataset apparently doesn’t go that far back. Given that Tambora lifted 25 cubic miles of stuff into the atmosphere and Krakatau in 1883 only lifted 4.5 cubic miles, Tambora could have blocked 20 W/m^2. OTOH, the SO2 output was only a third greater, so maybe only about 5 W/m^2. The biggest impact of the Year Without a Summer was at latitudes where summer freezes killed crops, so perhaps the distance to the tropical convection governor allowed for the cooling and southerly shift in the storm track that made for an “interesting” year.

    Willis shoots most of that down in http://wattsupwiththat.com/2012/04/15/missing-the-missing-summer/ so your mileage may vary.

  42. Stephen Wilde:

    Thankyou for your reply to me at September 22, 2013 at 12:00 pm. However, it seems I was inadequately clear.

    I was not mentioning Jelbring’s PhD thesis on wind driven climate. I was mentioning his hypothesis that any planet has a surface temperature defined by gravity and atmospheric mass (assuming an atmosphere with sufficient atmosphere which e.g. Mars lacks).

    Richard

  43. Thanks Richard. I’m struggling with Readcube so I have not been able to get a good understanding of that paper yet.

    It’s a shame that most of the interesting and objective climate science seems to have stopped being published around 1990.

  44. This has an uncanny similarity, in my mind at least, to the way weight gain and obesity is treated by most people, including those in medicine. That is, only the forcings are considered (food intake and exercise output as positive and negative forcings) and the homeostatic system is ignored. The governor (feedback) system actively affects the forcings in both cases.

    Our friend W. M. Briggs might not be too pleased with your use of p-values. ;)

  45. Would somebody on this blog (the most &c…) please highlight the sleight-of-hand (a.k.a. the movement of the pea …. watch it!) which the IPCC uses.
    All their previous (woe is me, doom & gloom) prognostications have used 1971 (± whatever) to 1998 (± whatever) which has given them ) 0.2°C/decade. Continuation of that they used to scare us (to 2100! – catastrophe!).
    Now, when we have the ‘pause’ (plateau … whatever) they start it at 1951 – that gives 0.13°C (or whatever). Comparatively that makes the current ‘pause’ (whatever …) look less destructive to their narrative (catastrophe … &c.).
    w, I’ve followed you for a long time. Can’t you pick this up?
    Help, anybody.

  46. Steven Wilde: re ITCZ
    Towards the north pole = warming.
    Towards the equator or the south pole = cooling.
    It would be helpful to ascertain the neutral position first though.

    Can you point me to data that supports that? This is something ( one of many things ) I’ve been meaning to look into.

  47. Eric1sceptic said:

    ” I realize that your website concludes that atmospheric composition is not important, but simulations of atmospheric columns differ from that conclusion. ”

    Not quite right.

    I accept that atmospheric composition has a role in climate change but not as regards total system energy content.

    I am aware of simulations of the atmospheric column but they do not give enough weight to solar effects and give too much to GHG effects. Recent observations are suggesting that the solar effects are overwhelming but not from TSI variation alone. It is rather the effect of solar particles and wavelengths on the vertical temperature profile that matters and in particular as to how those variations affect the equator to pole gradient of tropopause height. That is what allows the jets and climate zones to slide to and fro latitudinally beneath the tropopause thereby adjusting the global energy budget.

    Our emissions would affect the global air circulation such that the circulation must change to negate their net thermal effect.

    However, observations and historical records suggest that solar and oceanic influences shift the circulation latitudinally by up to 1000 miles in certain regions.

    Since mass determines the greenhouse effect and composition only the circulation the obvious conclusion must be that our emissions have a miniscule effect.

    I would be surprised if we had shifted the circulation by as much as a mile. In reality we could never measure it because of the weather ‘noise’ in the climate system.

  48. Greg Goodman says:
    September 22, 2013 at 12:38 pm

    Check out the behaviour of the jets and climate zones in historical documents during the LIA, MWP and Modern Warm Period.

    I also saw something about the Marshall Islands which are near the ITCZ I’m sure it moved northward during the recent warming spell and was nearer the equator in the LIA.

  49. .Richard Courtney said:

    “I was mentioning his hypothesis that any planet has a surface temperature defined by gravity and atmospheric mass ”

    Plus ToA insolation. If there is no insolation the atmosphere stays frozen on the ground (excluding geothermal energy)

    I recall that being accepted science back in the 1960s. My contribution is to incorporate the principle into a plausible climate hypothesis.

    It all went awry when the radiative only concept took over.

    It is a pity that the old text books seem to have been destroyed and predated the internet. Some may still exist so there is a research project for someone

  50. jai mitchell says:
    September 22, 2013 at 10:04 am

    Interesting,

    Just be sure you set the correct boundary conditions when looking at your thermodynamic balance. I noticed in your previous work you listed “rain” as a potential cooling mechanism. I have heard from an uneducated caller on a radio show that more lakes means global cooling (because it is cooler near the lakes)

    This kind of thinking isn’t really helpful since the correct boundary conditions are set about 1/4 of a mile above the top of the atmosphere. If the energy going into that sphere containing the whole earth and all of it’s functions is more than the energy leaving that sphere then the planet is warming.

    There is no other scientific reality. You cannot put energy into an object without it warming, you cannot take energy from that object without it cooling.

    Thanks, Jai. While what you say about the TOA is correct, it is immaterial. Why? Because we are not talking about the temperature of the system as a whole. We are talking about the surface temperature … and as the poorly named “greenhouse effect” clearly shows, we can have identical conditions 1/4 mile above the TOA, and very very different conditions at the surface.

    That is why rain is a cooling mechanism … because we’re talking, not about the heat content of the planet, but the temperature of the surface, and rain definitely cools the surface.

    w.

    PS—I’d suggest that you just put your claims out there, without claiming that they represent “scientific reality”. It’s bad enough to be wrong as you are in your claims above.

    But being wrong after you’ve claimed that your result is “scientific reality”? Not good at all. In fact, whenever a man makes that claim, that he can distinguish “scientific reality”, I examine his work triple-hard … just sayin’.

  51. Mike Jonas says:
    September 22, 2013 at 10:25 am

    Willis – please can you explain how you translate W/m^2 into “per doubling of CO2″. The IPCC report would presumably have two separate measures, direct and indirect (ie. without and with feedbacks).

    Since my calculations involve observations, they include all feedbacks. The trend is not in W/m2, but in degrees per W/m2. Since the IPCC puts the forcing from a doubling of CO2 at 3.7 W/m2, I multiply the trend in °C W/m2 by 3.7 to give °C per doubling of CO2.

    Also, why the figures you quote relate to ECS not transient sensitivity (I didn’t get that bit).
    TIA

    Because I didn’t use the transient (instantaneous) forcing Fi. If I had, I’d get a number for the TCR, which I’d then have to increase (by about 20% according to the results of OTTO) … but that just gives me the figures I got using the smaller equilibrium forcing Fe.

    w.

  52. Ric Werme says:
    September 22, 2013 at 12:05 pm

    Willis shoots most of that down in http://wattsupwiththat.com/2012/04/15/missing-the-missing-summer/ so your mileage may vary.
    —————————————————

    Thanks for the link to Willis’ post skeptical of the Year Without a Summer. I had missed it.

    A few comments. Crop prices fell after 1814 because of the end of the Napoleonic Wars (except for the 100 days). They’d have fallen even more in Britain but for the Corn Laws.

    Also, Willis’ closing comment, “That was the point I was trying to make above, that if the weather really had been all that bad in 1992 the crop yield would have reflected it, and it didn’t. Not for any type of produce, not for tubers, not for legumes, not for vegetables, not for fruits, not for grains” is simply wrong, in the case of wheat, an important grain crop.

    Global wheat production fell dramatically in 1992 & didn’t recover to 1991 levels until 1998, prices took a big jump that year (as I well recall) & at least in the US, yield fell (from 39.5 bu/A to 34.3):

    http://www.agmrc.org/media/cms/ccpwheat_47A4CABBA76E0.pdf

    Table1. World Wheat Production, Consumption, Trade, and Ending Stocks (1986-
    1999a)
    Year Production Consumption Tradeb Ending stocks Stocks-to- Traded
    (Million metric tons) (%) (%)
    1986 494.9 490.4 84.7 170.6 34.8 17.11
    1987 524.1 515.6 90.7 179.1 34.7 17.31
    1988 496.0 527.2 115.6 147.8 28.0 23.31
    1989 495.0 524.5 104.3 118.4 22.6 21.07
    1990 533.2 532.7 103.8 118.9 22.3 19.47
    1991 588.0 561.9 101.1 145.1 25.8 17.19
    1992 542.9 555.5 111.2 132.5 23.8 20.48
    1993 562.4 550.3 113.0 144.5 26.3 20.09
    1994 559.0 561.9 101.4 141.5 25.2 18.14
    1995 524.8 547.6 100.8 118.7 21.7 19.21
    1996 538.6 550.6 98.8 106.7 19.4 18.34
    1997 582.8 576.7 101.3 112.8 19.6 17.38
    1998 610.0 584.9 100.6 137.9 23.6 16.49
    1999 586.6 597.1 95.6 127.4 21.3 16.30
    aJuly-June Marketing Year
    bExclued intra-European Union Trade
    cStocks-to-consumption ratio
    dTrade-to-production ratio
    ePreliminary estimate

  53. Thomas says:
    September 22, 2013 at 10:30 am

    Maybe Eschenbach has written about it before, but I’m a bit confused on how he can reconcile “I hold that changes in forcing only marginally and briefly affect the temperature. Instead, I say that a host of emergent thermostatic phenomena act quickly to cool the planet when it is too warm, and to warm it when it is too cool” with the existence of ice age cycles. Whatever thermostat the Earth has doesn’t seem all that good.

    The planet’s temperature varied by ± 0.3°C over the last century. This is a regulation to within about ± 0.1% … on a free-running system which is regulated by nothing more substantial than wind and water.

    If you know anything about heat engines, you’ll agree that that is a fantastic governor …

    w.

  54. ‘The forcing from stratospheric volcanic aerosols can have a large impact on the climate for some years after volcanic eruptions. ‘ like a small child , when there caught out lying they can never admit to their mistakes , but keep making up ‘reasons’ when they weren’t wrong despite all the evidence .

  55. richardscourtney says:

    September 22, 2013 at 12:06 pm
    I was not mentioning Jelbring’s PhD thesis on wind driven climate. I was mentioning his hypothesis that any planet has a surface temperature defined by gravity and atmospheric mass (assuming an atmosphere with sufficient atmosphere which e.g. Mars lacks).

    Amazing how often I have heard this explanation posited by people on this site as an alternative reason for climate change

    what is even more amazing is that this theory has been thoroughly debunked ON THIS VERY WEBSITE:

    http://wattsupwiththat.com/2012/01/24/refutation-of-stable-thermal-equilibrium-lapse-rates/

    As we can see, it is an introductory physics textbook exercise to demonstrate that an adiabatically isolated column of gas in a gravitational field cannot have a thermal gradient maintained by gravity. The same can readily be demonstrated by correctly using thermodynamics at a higher level or by using statistical mechanics, but it is not really necessary. The elementary argument already suffices to show violation of both the zeroth and second laws of thermodynamics by the assertion itself.

    In nature, the dry adiabatic lapse rate of air in the atmosphere is maintained because the system is differentially heated from below causing parcels of air to constantly move up and down. Reverse that to a cooling, like those observed during the winter in the air above Antarctica, and the lapse rate readily inverts. Follow the air column up above the troposphere and the lapse rate fails to be observed in the stratosphere, precisely where vertical convection stops dominating heat transport. The EEJ assertion, that the dry adiabatic lapse rate alone explains the bulk of so-called “greenhouse warming” of the atmosphere as a stable feature of a bulk equilibrium gas, is incorrect.

  56. Noblesse Oblige says:
    September 22, 2013 at 10:43 am

    The large climate sensitivities claimed by IPCC not only give large negative temperature excursions but also very long times to return to normal. (See for example Lindzen http://link.springer.com/article/10.1140/epjp/i2012-12052-8#page-1). This must be true because climate senitivity is basically the climatic relaxation time divided by the effective specific heat of the climate system. So if IPCC were right, we still have not recovered from Krakatoa (1883) and Katmai (1912), not to mention Pinatubo (1992). In fact the IPCC climate sensitivity is so large that a normal century’s eruptions would keep the earth about a deg C cooler than otherwise and we would never be far from a volcanic “winter” from a few major eruptions.

    This alone should raise eyebrows about the plausibility of the IPCC values of climate sensitivity.

    Thanks, Noblesse. That’s what I was trying to say in my “future analyses” note after the end of the post. To maintain temperature, if we lose energy we have to gain it back … and the IPCC has no mechanism for that at all.

    w.

  57. As a complete amateur who has followed these discussions… The impact of volcanoes will be greatly affected by the latitude of the volcano, of course.

    But, my unproven thought is that the impact will also be greatly affected by the physical geography too. If the volcano feeds into the jet stream then the impact would be amplified.

    On the other hand, if the volcano is in a location next to a dustbowl then there will already be particulates in that place and only the cooling effect of the sulphates will have an effect. That would reduce the perceived impact.

    And the effect of a volcano in a forest would have a longer time lag than in an arid area. There must be more complications too.

    My apologies if this is self-evident and already considered.

  58. Seth says:
    September 22, 2013 at 11:44 am

    Thanks Willis,

    I’m not a scientist, so please pardon my ignorance, but in figure 4 are the two scales used equivalent? Does forcing of 4 w/m2 equal 0.8C in temperature anomaly? If so, what are the assumptions behind this analysis? Also, doesn’t figure 2 illustrate the temperature anomalies, rather than the actual surface temperatures? This seems like an important distinction which is often missed, but maybe I’m wrong.

    First, the two scales are not equivalent. As is usually the case when a graph has two scales, they are used because the two variables are not in the same numeric range.

    And yes, Figure 2 is anomalies, as is spelled out in the caption.

    w.

  59. Salvatore Del Prete says:
    September 22, 2013 at 11:50 am

    3. Impact of Volcanic Eruptions
    The global annual average surface temperature response to volcanic eruptions is cooling, resulting from increased absorption and reflection of incoming shortwave radiation by stratospheric aerosols. Averaging all years of the simulations together, the mean annual average cooling was -0.35 C for the periodic Pinatubo eruption, -0.77 C for the periodic Tambora 2P eruption, -1.09 C for the periodic Tambora 3P eruption, and -0.44 C for the observed 1959-1999

    The above is from the study I sent inmy previous post.

    Please point out to us in Figure 3 where the temperature dropped by ~ four-tenths of a degree in 1992 as your citation claims … USE YOUR EYES AND YOUR BRAIN, Salvatore, don’t blindly believe something because Gavin Schmidt claims it is true … in fact, if you see his name on the paper, triple your skepticism.

    w.

  60. OBSERVATIONS OF THE ATMOSPHERIC EFFECTS OF THE 1991 PLINIAN ERUPTION OF MOUNT PINATUBO

    Whether or not the data seem to become lost in long time series, Plinian volcanic eruptions can indeed influence climate and other parameters for a year or more. The major eruption of Mount Pinatubo on 15 June 1991 injected some 20 megatons of SO2 into the stratosphere. This evolved into a layer of aerosols above the tropopause that eventually blanketed most of the planet. At my observing station in South Central Texas, the arrival of the aerosol cloud in July 1991 was visually obvious during the day and especially at dawn and dusk.

    As Willis observes, the aerosol blanket reflects sunlight back into space. The aerosols also absorb sunlight. In the case of the Pinatubo cloud, I used a calibrated unfiltered silicon solar cell and a variety of calibrated, filtered sun photometers to measure both direct and full sky solar irradiance. The aerosols caused an increase in the aerosol optical depth at 1000 nm of about 0.04 during the latter half of 1991 and most of 1992. This was a reduction of about 4 percent at 1000 nm. The aerosols reduced the photocurrent from the solar cell at noon by about 5 percent during this time. These observations were associated with a reduction in temperature of about 2 degrees F, which is similar to other reports elsewhere.

    In contrast with the global time series presented by Willis, my time series from 1990 to the present clearly shows the Pinatubo eruption’s association with anomalies in temperature, aerosol optical depth, total solar irradiance and, later, a reduced total ozone column and an increase in solar UV-B.

    During the early months the volcanic aerosols sometimes formed alto cirri clouds that were visible in full daylight and, especially, during twilight. The aerosols also formed a Bishop’s ring on a number of occasions. Brilliant, colorful sunsets were visible for more than 2 years, photographs of which I have published online and in print.

    The Pinatubo aerosol cloud first arrived over the Gulf of California when I was aboard a cruise ship chartered for observers of the total solar eclipse of 11 July 1991. The aerosol cloud arrived the evening of 12 July with an extraordinarily red sky. Had it arrived around noon during the eclipse the previous day, my measurements of the ozone layer before and after the eclipse would have been compromised. Fortunately, the eclipse occurred a day earlier, and measurements were made of several waves in the ozone layer along the path or totality over the Gulf of California. My son Eric simultaneously measured waves just outside the path of totality in Texas, and we published a joint paper on our findings. (F. M. Mims III, and E. R. Mims, Fluctuations in Column Ozone During the Total Solar Eclipse of July 11, 1991, Geophysical Research Letters, 20, 5, 367-370, 1993. Also a poster paper at the Quadrennial Ozone Symposium, University of Virginia, June 1992.)

  61. Salvatore Del Prete says:
    September 22, 2013 at 11:58 am

    The evidence is that increased volcanic activity during past prolonged solar minimums enhanced the temperature declines the globe experienced at those times periods

    When a man starts out with “the evidence is” and doesn’t cite the evidence, sorry, Salvatore, but I ignore the comment. As far as I know, we have no evidence for your claim at all.

    In fact, I recall this BS claim now, and I discussed it specifically in “Dronning Maud Meets The Little Ice Age” as cited above … so in addition to not providing evidence, you also haven’t done your homework. If you want to discuss my work, READ IT FIRST!

    w.

  62. “The reduced trend in radiative forcing is primarily due 3 to volcanic eruptions and the downward phase of the current solar cycle.”

    The second point is of interest to me as it seems that the IPCC finally recognise that solar activity operates inversely with global temperature. Doubtless this will now be puit into their models and the late C20th warming explained as being as result of increased solar activity during the Grand Maximum with (any) contribution due to CO2 being adjusted downwards in line with an accurate quantification of this effect. Oh! And an appology for getting things all wrong yet again.

  63. “In response to this ‘super greenhouse effect’, highly reflective cirrus clouds are produced which act like a thermostat shielding the ocean from solar radiation.”

    I find it a bit odd that we are referring to this as a “greenhouse effect”

    More like a “”shadecloth” effect. It is not stopping heat escaping, it is stopping extra heat coming in.

    Pretty much the opposite of a greenhouse.

  64. Willis –

    It would be very helpful to see your lag linear-predicted lagged temperature anomaly overlaid on the actual temperature residual, to get a feel for how much volcanoes do or do not explain. Some of the forcing values (-4W/m2) are very large, so would explain anomalies of several tenths of a degree by your analysis, which are comparable to that in the residuals.

    R.

  65. Stephen Wilde says:
    September 22, 2013 at 12:00 pm

    richardscourtney says:
    September 22, 2013 at 11:48 am

    Hi Richard.

    Hans Jelbring is correct in general terms as regards his ‘Wind Driven Climate’

    Oh, dear God, not the Jelbring Hypothesis again. It has been stomped into the ground here, and is not even worth discussing. Richard proposed that you do a search on WUWT for the discussion … which you clearly have not done.

    Jelbring’s hypothesis violates the Second Law of Thermodynamics. I say so. Professor Robert Brown wrote a whole post saying so. Joel Shore, who is a warmer but whose science-fu is good says so. Elementary thermodynamics texts say so.

    And in my post “A Matter Of Some Gravity“, I put forward an interesting post that showed that not only the proposed Jelbring mechanism, but NO MECHANISM, can warm the surface as Jelbring claims.

    If you wish to discuss it, take it to Tallblokes. Do not try discussing it here. At least at Tallblokes, people won’t point and laugh and throw things. Here, you’re just damaging your reputation.

    w.

  66. I’m not a scientist, just a citizen fascinated by the global warming debate. The AGW hypothesis seems to be an groupthink IPCC assumption. Richard S Courtney says on September 22, 2013 at 10:16 am Willis asks:

    “With 2/3 of the planet covered by water, perhaps volcanoes’ greatest impact is the result of injecting heat into the oceans.”

    Apparently there are hundreds or even thousands of unknown submarine volcanoes. My questions to Willis (or anybody) are:

    1. Does the IPCC process estimate the GHG emissions, particularly CO2, that come from all these uncharted submarine volcanoes?

    2. What about emissions from hydrothermal vents?

    3. What about CO2 that must be bubbling up from much of the ocean floor?

  67. Jim S says: Don’t volcanoes also emit large quantities of CO2 into the atmosphere? Is this taken into account?

    Greg Goodman says: Largeness is relative. In view of what we chuck out and the natural annual carbon cycle volcanoes are a fart in the wind.

    The annual human contribution to the atmosphere is 9 gigatons of CO2 measured as Carbon. There is only one place where the entire output of a volcanic seep is (or was) sequestered–Lake Nyos. Estimates of that source alone range as high as 700 million SCF of CO2 annually. Using a more conservative yearly figure of 187 million SCF and multiplying by three million subsea volcanoes, I get 5.61 X 10¹⁴ SCF/year, globally. Some fart, Greg!

    Converting to metric tons of carbon per year, that’s about 8 gigatons versus the human contribution of 9 gigatons. No, we don’t know the sizes of those subsea volcanoes nor their emission rates, nor the emission of land-based volcanoes. But these figures hint that the volcanic CO2 release rate may have been grossly understated. [Note that I don’t say “underestimated,” because AGW-activist scientists don’t always state what they estimate.]

  68. Willis Eschenbach says:
    September 22, 2013 at 1:40 pm

    I don’t accept every aspect of the Jelbring hypothesis, merely the part that emphasises the function of winds in redistributing energy. I don’t see the mass/gravity issue as his since to my recollection it was once the consensus view.

    I am aware of your antipathy to the gravity/atmospheric mass issue but respectfully consider you and all those who support you to be wrong.

    It is interesting to note that your own hypothesis proposes some baseline sea surface temperature that cannot be exceeded even if the atmospheric composition changes.

    You have provided no suggestion as to how that might be achieved.

    The only way I can see it being achieved is via atmospheric pressure on the water surface and that involves mass not composition.

    Your own thermostat hypothesis needs the gravity/mass relationship to work in the first place unless you can come up with a more likely explanation.

  69. Gerald Wilhite:

    Your post at September 22, 2013 at 1:46 pm begins saying

    I’m not a scientist, just a citizen fascinated by the global warming debate. The AGW hypothesis seems to be an groupthink IPCC assumption. Richard S Courtney says on September 22, 2013 at 10:16 am Willis asks:
    “With 2/3 of the planet covered by water, perhaps volcanoes’ greatest impact is the result of injecting heat into the oceans.”

    I am confused because neither Wiilis nor I said what you quote and – as far as I can see – neither of us replied to it.

    However, I did respond to theyouk:suggesting volcanic heat could have an effect in his post at September 22, 2013 at 10:16 am.
    My reply is at September 22, 2013 at 10:40 am

    http://wattsupwiththat.com/2013/09/22/the-eruption-over-the-ipcc-ar5/#comment-1423700

    and reports the Ramanathan&Collins effect.

    Your post then goes on to discuss CO2 emissions from submarine volcanos. This could not affect atmospheric CO2 because of the carbonate buffer: almost all the CO2 in the carbon cycle is already in the oceans. However, sulphate emissions from submarine volcanos could significantly alter atmospheric CO2 concentration by altering ocean surface layer pH. I have explained this in WUWT threads where it is pertinent. It is NOT pertinent here and would confuse this thread which concerns atmospheric (n.b. NOT oceanic) SOx.

    Richard

  70. John Daly used 6 different methods to calculate 2XCO2 sensitivity and all of them were in line with your result, around 0.2C.

    http://www.john-daly.com/miniwarm.htm

    Looking at the 7 major eruptions from Krakatoa onwards, 5 of them occurred in spring to mid-summer.

    Mount Pinatubo Luzon Volcanic Arc Jun 15, 1991
    5 Mount St. Helens Cascade Volcanic Arc May 18, 1980
    6 Novarupta Aleutian Range Jun 6, 1912
    6 Santa María Central America Volcanic Arc Oct 24, 1902
    5 Mount Tarawera Taupo Volcanic Zone Jun 10, 1886
    6 Krakatoa Sunda Arc Aug 26–27, 1883
    5 Cosigüina Central America Volcanic Arc January 20, 1835

    Plus El Chichon, March/April 1982

    If I am correct in my thinking that monsoon intensity (especially NH monsoon) is a major climate feedback (negative) on an annual scale then the lag you show in Fig 5 would be significantly affected by the time of year the eruptions occurred.

  71. Stephen Wilde:

    In your post addressed to Willis at September 22, 2013 at 1:59 pm you say

    It is interesting to note that your own hypothesis proposes some baseline sea surface temperature that cannot be exceeded even if the atmospheric composition changes.

    You have provided no suggestion as to how that might be achieved.

    He does not need to because that has been known since 1991. Please read my post in this thread at September 22, 2013 at 10:40 am

    http://wattsupwiththat.com/2013/09/22/the-eruption-over-the-ipcc-ar5/#comment-1423700

    which reports the Ramanathan&Collins effect.

    Richard

  72. Forrest M. Mims III says:
    September 22, 2013 at 1:33 pm

    OBSERVATIONS OF THE ATMOSPHERIC EFFECTS OF THE 1991 PLINIAN ERUPTION OF MOUNT PINATUBO

    Whether or not the data seem to become lost in long time series, Plinian volcanic eruptions can indeed influence climate and other parameters for a year or more. The major eruption of Mount Pinatubo on 15 June 1991 injected some 20 megatons of SO2 into the stratosphere. This evolved into a layer of aerosols above the tropopause that eventually blanketed most of the planet. At my observing station in South Central Texas, the arrival of the aerosol cloud in July 1991 was visually obvious during the day and especially at dawn and dusk.

    As Willis observes, the aerosol blanket reflects sunlight back into space. The aerosols also absorb sunlight. In the case of the Pinatubo cloud, I used a calibrated unfiltered silicon solar cell and a variety of calibrated, filtered sun photometers to measure both direct and full sky solar irradiance. The aerosols caused an increase in the aerosol optical depth at 1000 nm of about 0.04 during the latter half of 1991 and most of 1992. This was a reduction of about 4 percent at 1000 nm. The aerosols reduced the photocurrent from the solar cell at noon by about 5 percent during this time. These observations were associated with a reduction in temperature of about 2 degrees F, which is similar to other reports elsewhere.

    Forrest, it’s always great to hear from you. For those unfamiliar with the name, Forrest used to write the “Amateur Scientist” column in Scientific American … back when both were actually about science.

    Forrest, I assume that the measurements that you refer to were taken by you at a single point … and the effect of Pinatubo or other “Plinian” eruptions in local areas is well known. And indeed the stratospheric effects are global … but you’re not seeing the other side of the coin.

    When a volcano cuts down the incoming solar in the tropics, they cool. When they are cool (as I have shown) the clouds form later, and as a result, you get less sunlight, but for a longer time. So the effect you measured is indeed real … it just doesn’t do what people think it does.

    However, it does not become “lost in long time series” as you say … if it were there on a global basis we’d see it regardless of the length of the time series. It is counteracted by emergent phenomena.

    When the globe cools, the tropical clouds form a few minutes later, the thunderstorms form a few minutes later … and that brings the global temperature back up. This is because the thermostatic mechanisms which maintain the temperature, such as the time of tropical cloud formation, are TEMPERATURE based, and not FORCING based. The clouds don’t know or care WHY it is cooler—if it’s cooler they form later or don’t form at all, and the full heat of the sun quickly warms the planet back up.

    So when a volcanic eruption cools the world, or when a change in solar strength warms the world, it is offset by changes in, inter alia, cumulus formation time and amount, thunderstorm formation time and amount, and El Nino/La Nina formation time and amount. A trivially small, almost unmeasurable change in these homeostatic mechanisms is more than enough to offset the effects of the volcanoes.

    Best regards, and my thanks for your work with Sci Am … which was back before they morphed into the Sc Am.

    The “Amateur Scientist” column that your predecessor Martin Gardner had and that you continued was my inspiration as a child on the ranch and then as younger man. It was one reason I’m an amateur scientist today. I waited for it every month. Can’t tell you how many things I’ve built from that column, including a Wilson cloud chamber (which kinda worked), water-based logical switches and flip-flops using eyedroppers, and a Hilsch Vortex Tube. You have my eternal gratitude.

    w.

  73. milodonharlani says:
    September 22, 2013 at 1:08 pm
    Global wheat production fell dramatically in 1992 & didn’t recover to 1991 levels until 1998, prices took a big jump that year (as I well recall) & at least in the US, yield fell (from 39.5 bu/A to 34.3):
    ====================================================================
    You’re applying a micro to a macro. Yes, for a very short time global wheat production declined, but at the same time corn and rice increased. There are many, many reasons for price fluctuation other than the ability to produce. If you wish, you can go here to see the near constant increase of crop global production. https://suyts.wordpress.com/2013/09/08/oh-puh-leeese-do-you-see-a-twelve-year-cycle-here-maybe-but-it-doesnt-make-any-difference/

  74. About six years ago my colleagues and I published a series of papers on Pinatubo. The first paper provoked two comment papers by prominent climate scientists disagreeing with our results. We published replies to both showing that their analysis was wrong. All five of these papers were published in Geophysical Research Letters (GRL).

    Douglass, Knox, Pearson and Clark (DKPC) published a sixth summary paper “Thermocline flux exchange during the Pinatubo event” also in GRL.
    (http://www.pas.rochester.edu/~douglass/papers/Douglss_Knox_pearson_clark2006GL026355.pdf). We found that the delay was 4.4 months and more importantly that the total heat flux integrates during this event integrates to zero — i.e. The temperature of the Earth after the event returns to what it was before.
    ——————————————-
    ABSTRACT
    “We analyze the temperature anomaly of the Pinatubo eruption using an exact mathematical solution of a standard energy balance model that includes coupling between the mixed layer and the thermocline. Our solution yields a short response time t = 4.4 months and a small climate sensitivity l = 0.22 C/(W/m2), implying short-term negative feedback. Also, our analysis determines a value of the effective eddy diffusion constant k = 2 x 10 (-6) m2/s that is much smaller than that assumed in many climate models. We find for this model that the heat flux to the thermocline reverses sign and integrates to zero for any forcing of finite duration. This effect should be observable in any future Pinatubo- type event.”
    ——————————————
    The authors of the last IPCC report knew of our papers and the comment papers by the two prominent climate scientists who disagreed with our results. However, they deliberately failed to reference our replies showing that they were wrong. When that report came out, it was obviously why — our results disagreed with their conclusions.

    I suspect that IPCC AR5 will again ignore our results which after six years is still the last word on Pinatubo.

    David Douglass
    Dept of Physics and Astronmy
    University of Rochester

  75. Willis, I am a bit perplexed by the volcanic forcing dataset you graphed. I am not doubting the veracity of the dataset, it being from GISS and all…sorry. No, seriously, it is likely accurate, but why then is the forcing from Novarupta in 1912 so small? I wrote about it and the beheading of Katmai and other “extreme weather” events of 1912 recently so it is fresh in my mind. It was supposedly the “most powerful volcanic eruption of the 20th Century”:

    By the time the eruption ended the surrounding land was devastated and about 30 cubic kilometers of ejecta blanketed the entire region. This is more ejecta than all of the other historic Alaska eruptions combined. It was also thirty times more than the 1980 eruption of Mount St. Helens and three times more than the 1991 eruption of Mount Pinatubo, the second largest in the 20th Century.

    Then why would it be but a blip in your graph? Novarupta’s volcanic forcing appears to rank 7th in your graph and about half as powerful a forcing as Mount Pelée in 1902. The other spikes all seem to correlate with the other known large eruptions of the last century-and-a-half, but Novarupta seems the anomaly. Any idea why it’s forcing would have been so weak? Different type of eruption perhaps? I know very little about volcanology so I figured I’d ask in case you or anyone else already had a likely explanation before I go digging any further.

  76. Welcome back from your travels, Willis, and thanks for a great thought provoking post to start off. The Earth has survived long enough to show that its natural system does not crash into wild fluctuations caused by minor variations of CO2 whatever their cause. Earth has a long term agenda. CO2 should not be on our agenda except to be thankful for its benefits to plants.

  77. Hi Willis, can you explain how your thermostat relates to Svensmark’s theory? I think I’ve got it – Svensmark’s GCR represents an adjustment to the thermostat itself – is that right?

  78. richardscourtney says:
    September 22, 2013 at 2:14 pm

    I found this in your link:

    “the effect they found is that increased heating of tropical ocean increases evapouration to increase cover by cirrus clouds which shield the surface from solar heating”

    The cirrus clouds are a secondary effect resulting from the initial increase in evaporation. My New Climate Model does in fact include cloud feedback effects.

    By shading the surface the cirrus clouds reduce the amount of extra evaporation required to restore equilibrium.

    Therefore that is not an adequate explanation for the process of increased evaporation in the first place, merely an example of the negative system response to whatever caused the initial rise in evaporation.

    It is the increased rate of evaporation that is the primary effect arising from the addition of extra energy to the ocean surface. Willis does not explain how that increase in evaporation is always sufficient to put a firm lid on maximum sea surface temperature.

    If the cirrus clouds did not form there would still be the same maximum temperature, just faster evaporation which would not decline if no cirrus clouds formed.

    There comes a point where ALL the added energy is ejected by increased evaporation over water or increased convection over land.

    That point is determined by atmospheric mass via pressure on the surface and not composition though composition does change the circulation required to achieve stabilisation of the system.

    This issue is highly relevant to the volcanic aspect too.

    Up thread it was pointed out that the effects of eruptions were neutralised eventually though there is some debate about how long it takes.

    That neutralisation is achieved by the global air circulation changing to adjust the ToA energy budget and the changes in circulation required are a product of atmospheric mass and surface pressure.

    The main point Willis made in his old thread was this:

    “Once the system is at equilibrium, therefore, there is no net flow between the surface and the atmosphere. ”

    But that begs the question because the system never is at equilibrium and every divergence from equilibrium causes an equal and opposite circulation response in order to adjust ToA energy balance.

    There is always a net flow between surface and atmosphere and it is always a negative system response.

    That brings us back inevitably to the question as to what sets the baseline temperature at the surface for any given level of ToA insolation.

    That is surface pressure.

    Even with a non GHG atmosphere the surface must be more than S-B because the mass (and thus pressure) of the non radiative atmosphere determines the amount of non radiative energy transfer between surface and atmosphere.

    The rate of non radiative transfer is infinitely variable so that if anything other than mass, gravity or ToA insolation seeks to disturb system equilibrium (as is always happening in reality) then the appropriate negative system response is supplied via circulation changes which give rise to NON-NET energy exchanges between surface and atmosphere until ToA radiative balance is corrected.

    It is wrong to simply ignore non radiative energy transfers between surface and atmosphere just because they net out over time.

    They are constantly supplying the necessary negative system response by becoming non net as necessary which whichever sign of response is required.

    No thermostat hypothesis can work without that mass/gravity relationship.

  79. Jai says
    “There is no other scientific reality. You cannot put energy into an object without it warming, you cannot take energy from that object without it cooling.”

    Really, did you ever hear of photosynthesis:

    “Photosynthesis is a process used by plants and other organisms to convert light energy, normally from the sun, into chemical energy that can be used to fuel the organisms’ activities. “

  80. suyts says:
    September 22, 2013 at 2:25 pm

    Wheat production, as I noted, didn’t recover for seven years, not a short time.

    I wasn’t arguing for a major effect from Pinatubo, but merely pointing out that Willis overstated the case. There was at least one major crop that did show a negative response. Moreover, at the time, the agricultural literature I read attributed at least some of the drop in wheat production & lowering in yield to Pinatubo, based upon what to me seemed valid argument & compelling evidence.

    Of course many factors go into total production & yield numbers for each crop. For instance, wheat acreage in the US also fell from 1991 to ’92, for a variety of reasons. Longer term, volcanoes fertilize soils locally & regionally.

    I wasn’t arguing for or against the overall importance of Pinatubo in global agriculture. But the fact is that wheat yield & production fell after 1991 & stayed down for most of the rest of the ’90s.

  81. “It is interesting to note that your own hypothesis proposes some baseline sea surface temperature that cannot be exceeded even if the atmospheric composition changes.
    You have provided no suggestion as to how that might be achieved.
    The only way I can see it being achieved is via atmospheric pressure on the water surface and that involves mass not composition.”

    1- phase change of water happens with no change of temperature – much heat is absorbed by the conversion of water from liquid to gas.
    2- the density of water gas is less than that of all other gases of an significant quantity in our atmosphere – therefore water gas rises without convection
    3- once it’s risen, phase change radiates the heat with no change in temperature and the gas condenses to liquid – changing its volume from a bottleful to a spoonful.

    (degrees ain’t watts when you have phase change, willis)

  82. Stephen Wilde:

    I am replying to your long post at September 22, 2013 at 3:02 pm.

    You say of the Ramanathan&Collins (R&C) effect

    Therefore that is not an adequate explanation for the process of increased evaporation in the first place, merely an example of the negative system response to whatever caused the initial rise in evaporation.

    That is a change of subject!
    You had complained that Willis did not explain the limit of 305K to maximum sea surface temperature. I replied that he does not need to because that has been known since 1991 and I cited the R&C effect. It explains the limit as I had reported earlier in this thread (with quotation and reference) in my post at at September 22, 2013 at 10:40 am

    http://wattsupwiththat.com/2013/09/22/the-eruption-over-the-ipcc-ar5/#comment-1423700

    The “the process of increased evaporation” is an increase in sea surface temperature induced by any cause.

    You then promote the Jelbring Hypothesis which – unless you can provide a reference – was not around in the 1960s because Jelbring did not provide it until the late 1990s. Discussion of it is not relevant to this thread and I repeat my suggestion that you do a Search for discussions of it.

    Richard

  83. Tom: “Is it possible that the “Year Without a Summer” in 1816 was not really caused by the eruption of Mount Tambora in Indonesia after all?”

    Possibly. But it was cold from 1809 to 1816.

    Consider HADCET June/July/Aug. Yes 1816 was the 3rd coldest JJA ever at 13.4. But the long term mean for JJA HADCET was 15.3

    All of the JJA’s were colder than the long term mean from 1809 to 1816. 1812 was almost as cold as 1816.

    1808 16.6
    1809 14.5
    1810 14.8
    1811 14.9
    1812 13.8
    1813 14.4
    1814 14.3
    1815 14.8
    1816 13.4

    http://wp.me/a1ASzZ-Iv

    http://www.metoffice.gov.uk/hadobs/hadcet/ssn_HadCET_mean.txt

  84. milodonharlani says:
    September 22, 2013 at 1:08 pm
    Global wheat production fell dramatically in 1992 & didn’t recover to 1991 levels until 1998, prices took a big jump that year (as I well recall) & at least in the US, yield fell (from 39.5 bu/A to 34.3):
    ====================================================================
    It may have fallen “dramatically from the abnormally high 1991 production. But it is right on the trendline for the period 1986 – 1999. Graph those figures in Excel and add a linear trendline*, then tell us you can see a “dramatic” drop.

    (*A trivial exercise, unless your real name is Phil Jones).

  85. I had a quick look at the data. Maybe i was looking at too many numbers for today but i think i can see the drop and the lag response. GIStemp 3 months avg. plus Sato’s forcing.

    DJF MAM JJA SON
    1990 34 54 33 34 1990
    1991 40 37 44 32 1991
    1992 36 31 12 0 1992
    1993 29 26 18 10 1993
    1994 14 29 27 37 1994
    1995 53 37 45 39 1995

    1991.458 0.0179 0.0205 0.0153
    1991.542 0.0377 0.0402 0.0353
    1991.625 0.0710 0.0668 0.0751
    1991.708 0.0964 0.0958 0.0970
    1991.792 0.1197 0.1187 0.1208
    1991.875 0.1380 0.1281 0.1480
    1991.958 0.1385 0.1468 0.1301
    1992.042 0.1483 0.1627 0.1338
    1992.125 0.1494 0.1691 0.1298
    1992.208 0.1428 0.1585 0.1270
    1992.292 0.1386 0.1456 0.1317
    1992.375 0.1379 0.1424 0.1333
    1992.458 0.1265 0.1199 0.1332
    1992.542 0.1228 0.1155 0.1300
    1992.625 0.1141 0.1068 0.1214
    1992.708 0.1044 0.1021 0.1068
    1992.792 0.0978 0.0975 0.0982
    1992.875 0.0918 0.0914 0.0921
    1992.958 0.0793 0.0823 0.0763

    1992 was a moderate El Nino year. Sorry about the format but it’s all about raw data, isn’t it.
    Am i seeing things?

  86. That was unexpected. Before i posted my comment there were at least spaces between the data.
    What a mess.
    Here 1992 again – 3 months avg.: DJF 0.36 MAM 0.31 JJA 0.12 SON 0.0

  87. Richard said:

    “You had complained that Willis did not explain the limit of 305K to maximum sea surface temperature. I replied that he does not need to because that has been known since 1991 and I cited the R&C effect”

    I didn’t ‘complain’, I just asked if he had a better idea.

    The suggestion from R&C is that the capping of sea surface temperature is achieved by the formation of certain types of cloud and not simply by an acceleration of the rate of evaporation sufficient to cancel out the added energy. Evaporation is a net cooling process so the faster it runs the more it cools a water surface.

    Clouds are fickle entities and vary greatly in timing location density and depth. If it were clouds that achieved the capping effect we would see a much wider range of sea surface temperatures with no firm cap because conditions would regularly allow the cap to be exceeded when conditions were unfavourable for the right sort of clouds at the right time and in the right place.

    Therefore I do not accept the R&C effect as an adequate explanation for such a firm cap. The observations are clear that 305C is the most we can get whatever the clouds do.

    We are left with the rate of evaporation and the energy cost of the latent heat of vaporisation.

    That is related to surface pressure.

    I well remember being taught in the 60s and having read in books back then that the maximum surface temperature of a planet with an atmosphere is limited by atmospheric mass and gravity at any given level of ToA insolation.

    The obvious reason for that is that any atmosphere whether radiative or not will acquire energy from and exchange energy with the surface and the amount exchanged will be related to mass rather than composition.

    Furthermore the rate of energy exchange is not net zero at any given moment. It constantly changes as a negative system response to any disruptive forcing element other than more mass, more gravity or more ToA insolation.

    You said I said:

    “The “the process of increased evaporation” is an increase in sea surface temperature induced by any cause. ”

    Where?
    Doesn’t sound like me at all.
    I might have said somewhere that an increased rate of evaporation arises from the addition of more energy from any cause but was it in this thread?

  88. Mike Smith says:
    September 22, 2013 at 11:54 am
    If Willis is correct (and this seems very plausible based on his data) does this not mean it likely that man-made aerosols have a lower than generally assumed impact on temperature?

    ‘Assumed’ is a good word to describe the published aerosol forcings from NASA/GISS, because they don’t have much empirical basis. There are important differences between anthropogenic aerosols and volcanic aerosols, including that the former are overwhelmingly in the lower troposphere and have a short residence time so their effects are primarily local to regional scale, while volcanic aerosols from major eruptions have a global effect and a residence time of months to perhaps several years. I wouldn’t draw any conclusions about anthropogenic aerosols from Willis’ conclusions about volcanic aerosols.

  89. Right off the bat I am wondering where (and exactly who) obtained the radiometers back in the late 1800’s to supply the aerosol data to us today that I.P.C.C. is then using to supply to politicians to make new laws and regulations?

    ( Willis, gee, I’m a climate heretic too (two years now), glad to have you onboard )

    Now I’ll read the rest.

  90. milodonharlani says:
    September 22, 2013 at 3:12 pm

    suyts says:
    September 22, 2013 at 2:25 pm

    Wheat production, as I noted, didn’t recover for seven years, not a short time.

    I wasn’t arguing for a major effect from Pinatubo, but merely pointing out that Willis overstated the case. There was at least one major crop that did show a negative response. Moreover, at the time, the agricultural literature I read attributed at least some of the drop in wheat production & lowering in yield to Pinatubo, based upon what to me seemed valid argument & compelling evidence.
    ==============================================================
    No worries. I understand how it is. Crops are of particular interest to me relative to the greater climate discussion. So, when an opportunity presents itself to demonstrate global production has increased in spite of all of the wailing, I usually try to demonstrate it.

  91. sadly you cant calcuate sensitivity that easily from the response to volcanoes.
    Imagine if I tried to calculate the sensitivity from the drop in termperature when the sun goes down.

    If you want to extract sensitivity information from the relaxation response you have to do a bit more work

    http://www.gfdl.noaa.gov/cms-filesystem-action?file=research/weather-atmos-dynamics/wallace_held.pdf

    Here is some more background. with references to papers one should read

    http://ceres.larc.nasa.gov/documents/STM/2007-11/ce0711151415Boer.pdf

    you need more than a back of the envelope.

  92. Willis Eschenbach: “T2 = T1 + lambda (F2 – F1) (1 – exp(-1/tau)) + exp(-1/tau) (T1 – T0)”

    As my excuse for choosing this time to pick that particular nit, all I can say is that you’ve given this formula more than once before, and each time I’ve failed to understand how it applies to the usual situation with which you’re dealing, i.e., to matching data representing averages over intervals that are potentially significant fractions of the system’s major time constant.

    Rather than attempt to set out my misgivings verbally, I’ll just set forth the following code, which draws a graph that compares how I understand you do it with the way I would have thought it should be done.

    I do this with some trepidation, since as I recall your formula received Paul_K’s imprimatur, but my frustration over failing to comprehend a one-line formula has overcome my embarrassment at that failure.

    simple = function(x, tau, lambda, delta_t, y0= 0){
    # For the simple case of a single-pole, scalar (also called “one-box”,
    # i.e., dy/dt = lambda / tau * x – y / tau), linear model, boils
    # linearSystemResponse() down to a two-line iteration

    n = length(x);
    xl = lambda * x;
    y = numeric(n);
    alpha = 1 – exp(-delta_t / tau);
    beta = 1 – tau / delta_t * (1 – exp(-delta_t / tau));
    ys = y0;
    for(i in 1:n){
    y[i] = ys + beta * (xl[i] – ys);
    ys = ys + alpha * (xl[i] – ys);
    }
    y;
    }

    eschenbach = function(x, tau, lambda, delta_t = 1, y0 = 0){
    # Implements my understanding of the implementation used at wattsupwiththat.com
    # by Willis Eschenbach after discussionwith Paul_K
    x = c(0, 0, x);
    n = length(x);
    y = numeric(n);
    y[1] = 0;
    y[2] = y0;
    for(i in 3:n){
    y[i] = y[i – 1] + lambda * (x[i] – x[i – 1]) * (1 – exp(-delta_t/tau)) +
    exp(-delta_t/tau) * (y[i – 1] – y[i – 2]);
    }
    y[-(1:2)];
    }

    modelImplementationQuery = function(){
    # Illustrates my difficulty with applying the Paul_K / Willis Eschenbach
    # approach to characterizing a system in accordance with the “one-box” model
    # when the system data take the form of averages over intervals whose
    # durations may be a significant fraction of the system time constant.

    x = rep(1, 5); # The stimulus’s averages over each of five successive intervals
    tau = 2; # The (single) time constant of the system to be simulated
    delta_t = 1; # The duration of each time interval over which x represents stimulus averages
    lambda = 1; # The equilibrium ratio of the system’s response to its stimulus
    t = (0:(length(x) – 1)) * delta_t; # Each interval’s start time

    # Now compute the system’s response
    # averages over those intervals in accordance with . . .
    y1 = eschenbach(x, tau, lambda, delta_t); # the Paul_K/Eschenbach approach
    #y2 = linearSystemResponse(x, -1 / tau, lambda / tau, delta_t); # my general-purpose routine
    y3 = simple(x, tau, lambda, delta_t); # that routine boiled down to the first-order-scalar (“one-box”) case
    # Since the data are intended to represent interval averages, plot them at
    # respective intervals’ midpoints:
    plot(t + delta_t / 2, y3, xlim = c(0, range(t)[2] + delta_t),
    ylim = c(0, lambda), xlab = “time”, ylab = “response”);
    #points(t + delta_t / 2, y2, pch = 3);
    points(t + delta_t / 2, y1, col = “red”);

    # Now plot the analytically determined step response:
    tFine = seq(t[1], t[length(t)] + delta_t, by = diff(range(t)) / 100);
    lines(tFine, lambda * (1 – exp(-tFine / tau)), lty = 3);

    }
    modelImplementationQuery();

  93. Stephen Wilde:

    I am writing to say that as a result of your post at September 22, 2013 at 3:56 pm I shall not be replying to anything else from you.

    At September 22, 2013 at 3:02 pm

    It is the increased rate of evaporation that is the primary effect arising from the addition of extra energy to the ocean surface. Willis does not explain how that increase in evaporation is always sufficient to put a firm lid on maximum sea surface temperature.

    In your post I am answering you quote my saying of that

    You had complained that Willis did not explain the limit of 305K to maximum sea surface temperature. I replied that he does not need to because that has been known since 1991 and I cited the R&C effect

    And you assert

    I didn’t ‘complain’, I just asked if he had a better idea.

    Not true! You did NOT ask if he had a better idea. You complained that he “does not explain”.

    You assert without evidence

    Therefore I do not accept the R&C effect as an adequate explanation for such a firm cap. The observations are clear that 305C is the most we can get whatever the clouds do.

    That is arm-waving. How do you know “whatever the clouds do”?

    And you say to me

    You said I said:
    “The “the process of increased evaporation” is an increase in sea surface temperature induced by any cause. ”
    Where?
    Doesn’t sound like me at all.

    I did NOT say you said that! I said that.
    I said it in my post at September 22, 2013 at 3:21 pm where I objected to your changing the subject when shown to be wrong

    http://wattsupwiththat.com/2013/09/22/the-eruption-over-the-ipcc-ar5/#comment-1423928

    So, you changed the subject when shown to be wrong, you claim to have not said what you did say, you arm-waved to refute when presented with evidence you didn’t like, and you falsely accused me of misrepresentation. Hence, I see no purpose in further discussion with you.

    Richard

  94. milodonharlani says:
    September 22, 2013 at 3:12 pm
    But the fact is that wheat yield & production fell after 1991 & stayed down for most of the rest of the ’90s.
    ========
    Scab and Vomitoxin….

    November 1996
    Wheat diseases becoming a national priority

    TCK smut and Karnal bunt barely register a bite on U.S. wheat acreage and production, but are bad dogs that bark loudly in the export market. It is the perceptibility of these two fungal diseases in U.S. wheat, not actual infections, that may be blamed for economic losses, largely in lost sales opportunities and testing, containment, and processing costs.

    On the other hand, Fusarium head blight, or scab, has been a more tangible disease. Yield and quality losses from scab, and its toxic byproduct, vomitoxin, have devastated half of the six major U.S. wheat classes

    It’s estimated that wheat growers in N.D. have lost well over $1 billion to scab in the 1990s, and wheat and barley growers in Minnesota, about $1.2 billion.

    http://www.smallgrains.org/springwh/November96/Sayler.htm

  95. @Willis Eschenbach

    I have real issues with your idea I am afraid Willis and I do not care what relationship you can show on a graph … CORRELATION DOES NOT EQUAL CAUSATION.

    Instead, I say that a host of emergent thermostatic phenomena act
    quickly to cool the planet when it is too warm, and to warm it when it is too cool.

    So for that to be remotely believed you need to provide mechanisms by which it would do it and correlating a few lines on a graph is far short of that, I think I would rather believe that the decrease in the number of pirates is causing global warming because that is a near perfect fit.

  96. The effect of the eruptions is easily seen in the moving average global temperature data at this climate4you site.
    The regular 7.5 year cycle of plateau and dip, plateau and dip is interrupted twice once by El Chichon and once by Pinatubo and manifests itself by the loss of shoulders on the transition between plateau and dip. The temperatures soon bounce back pretty much in line with the atmospheric transition data.

  97. suyts says:
    September 22, 2013 at 4:10 pm

    Crops are of interest to me, too, as a former wheat rancher whose family still farm wheat.

    But I’m more interested in reality even than making a point.

  98. MouruanH says:
    September 22, 2013 at 3:53 pm

    That was unexpected. Before i posted my comment there were at least spaces between the data.
    What a mess.
    Here 1992 again – 3 months avg.: DJF 0.36 MAM 0.31 JJA 0.12 SON 0.0

    Try putting <pre> before and </pre> after the block of text that is “preformatted,” e.g.:

    mysql> select dt, out_temp, dew_pt, rain, rain_rate, bar from raw where dt > '2013-9-21 7:0:0' and rain > 0.0;
    +---------------------+----------+--------+------+-----------+--------+
    | dt                  | out_temp | dew_pt | rain | rain_rate | bar    |
    +---------------------+----------+--------+------+-----------+--------+
    | 2013-09-22 04:00:00 |     68.3 |   65.3 | 0.01 |      0.00 | 29.543 |
    | 2013-09-22 04:10:00 |     67.4 |   64.7 | 0.25 |      7.20 | 29.543 |
    | 2013-09-22 04:20:00 |     66.1 |   63.7 | 0.05 |      0.46 | 29.545 |

    See my Guide to WUWT at http://home.comcast.net/~ewerme/wuwt/index.html for more HTML formatting information and experiment at http://wattsupwiththat.com/test-2/

  99. Latitude says:
    September 22, 2013 at 4:44 pm

    Scab didn’t hit MN wheat until 1993, & the effect on national production was negligible.

    Maybe the infestation was a delayed effect of Pinatubo. Who knows?

  100. RERT says:
    September 22, 2013 at 1:39 pm

    Willis –

    It would be very helpful to see your lag linear-predicted lagged temperature anomaly overlaid on the actual temperature residual, to get a feel for how much volcanoes do or do not explain. Some of the forcing values (-4W/m2) are very large, so would explain anomalies of several tenths of a degree by your analysis, which are comparable to that in the residuals.

    R.

    I think I’ll leave that as an exercise for someone else, I’m kinda under the weather today. You can estimate it by multplying the forcing by 0.05. The maximum forcing is -3.8 W/m2, so you’d expect a temperature swing of -0.2°C … for that largest swing.

    w.

  101. Richard.

    The simplest way to deal with the R&C issue is as follows:

    I) If surface atmospheric pressure were higher then more energy would be required to break the bonds between water molecules and evaporation could only occur at a higher temperature than at present. The temperature at which the sea surface temperatures are capped would have to rise The formation of clouds would be delayed until after the evaporation had occurred. It is obvious that higher pressure would need a higher temperature to counter it.

    II) If surface atmospheric pressure were lower then of course the opposite would happen.

    The key to the temperature at which the cap is set is the amount of energy required to break the bonds between water molecules and that is pressure dependent.

    Recognition of these facts rounds off any proposition of a thermostat effect so as to make it plausible in terms of the known physics.

    Willis, in my humble opinion, needs such a process to make his hypothesis work.

  102. Gerald Wilhite says:
    September 22, 2013 at 1:46 pm

    I’m not a scientist, just a citizen fascinated by the global warming debate. The AGW hypothesis seems to be an groupthink IPCC assumption. Richard S Courtney says on September 22, 2013 at 10:16 am Willis asks:

    “With 2/3 of the planet covered by water, perhaps volcanoes’ greatest impact is the result of injecting heat into the oceans.”

    Apparently there are hundreds or even thousands of unknown submarine volcanoes. My questions to Willis (or anybody) are:

    1. Does the IPCC process estimate the GHG emissions, particularly CO2, that come from all these uncharted submarine volcanoes?

    2. What about emissions from hydrothermal vents?

    3. What about CO2 that must be bubbling up from much of the ocean floor?

    We don’t have good answers to that question, Gerald, but my take is that it is quite small. Estimates for the global geothermal heat flow are on the order of a few tenths of a watt, if that.

    The thing about both CO2 and heat coming from volcanoes, either under or over the water, is that volcanoes are not very common. The US is active tectonically on a world scale, but there are only a few active volcanoes at any time. Sure, each puts out heat … but only occasionally. Think about how hard it is, for example, to find an area suitable for geothermal heat in Kansas …

    So when you average it 24/7 over the whole US, or the whole planet, you get small numbers. It’s a really, really big world.

    As always, YMMV,

    w.

  103. @ MouruanH

    To help prevent a mess with formatting of columnar data on a blog, do this: format in a separate text editor using a mono-spaced font like Courier, !! use NO tabs !!, only spaces. When you post, put a tag, <pre> before your data, put </pre> behind, the PRE tag. Your frustration should vanish. Brush up on how to use the PRE or CODE tags on the web.

  104. A true super eruption of a super volcano might be at odds with your “self regulating” surface temperature hypothesis. Yellowstone for instance:

    “The oldest identified caldera remnant straddles the border near McDermitt, Nevada-Oregon, although there are volcaniclastic piles and arcuate faults that define caldera complexes more than 60 km (37 mi) in diameter in the Carmacks Group of southwest-central Yukon, Canada, which is interpreted to have formed 70 million years ago by the Yellowstone hotspot.[5][6] Progressively younger caldera remnants, most grouped in several overlapping volcanic fields, extend from the Nevada-Oregon border through the eastern Snake River Plain and terminate in the Yellowstone Plateau. One such caldera, the Bruneau-Jarbidge caldera in southern Idaho, was formed between 10 and 12 million years ago, and the event dropped ash to a depth of one foot (30 cm) 1,000 miles (1,600 km) away in northeastern Nebraska and killed large herds of rhinoceros, camel, and other animals at Ashfall Fossil Beds State Historical Park. Within the past 17 million years, 142 or more caldera-forming eruptions have occurred from the Yellowstone hotspot.[7]” per Wikipedia (sorry)

    Living 243 mi from this caldera, much of the soil here is imbued with bentonite, a form of ancient volcanic ash. I suspect the atmospheric effects of such an eruption might last longer and be more significant than those you have looked at from the recent past.

  105. milodonharlani says:
    September 22, 2013 at 4:58 pm
    Scab didn’t hit MN wheat until 1993, & the effect on national production was negligible.
    ====
    I thought you were talking global?….dunno, I grew cows

    From the link….
    “Scab and vomitoxin were significant problems this year in the soft red winter wheat growing area, which includes Ohio, Illinois, Indiana, Arkansas, and Kentucky. Michigan’s ag director said that “wet weather and warm temperatures have created the worst outbreak of wheat scab in that state in 100 years.” It was the second major scab outbreak for the SRW growing area in the 1990s, with the previous occurrence in 1991.”

  106. I’m going to throw this out there, because for me, ANY Global Warming/Cooling theory no matter which way you lean, has a HUGE gap in it. We didn’t “discover” tectonic plate movement etc until the 1970’s. So “baby” science and added to it the fact that ocean floor is so vast and so hard to reach in some places that we just don’t have a complete picture of what the heck goes on down there.

    NOW, that said, in recent years, ocean specialists have been shocked over and over again when they find HUGE geothermal vents in the ocean floors spewing HOT WATER and CO2 and other “greenhouse gases” 24/7. They have ALSO recently discovered that submarine volcanoes can and DO “erupt” just like surface volcanoes (which they previously thought could not happen due to temperatures and deep sea pressures-so for the most part they expected ‘pillow lava’ and mild spreading etc.

    SO….it seems to me, that according to the research I’ve read, that there could be at least TWICE as many “active” submarine volcanoes (they estimate many, MANY more-in the thousands) going off-venting-erupting, etc PLUS the tectonic activity which allows HOT WATER and CO2 etc to enter the oceans all the time. Any “consensus” crap I can find on volcanic forcing is based SOLELY on land volcanoes that are KNOWN, and old, OLD measurements/calculations that don’t even begin to touch the ocean floor activity that I suspect IS GOING ON ALL THE TIME. In some cases the CO2 is so hot and under such pressure that it’s LIQUID CO2. A “superfluid”.

    Now, over centuries-that would of course cause oceans to warm, AND cause serious CO2 outgassing as those lower, compressed waters are mixed in with the oscillations etc. HUGE amounts of increase. But I don’t know all the mechanics and formulas regarding how much it could rise or intermix etc or how “fast” the effects of it could be seen on the surface and in the atmosphere. I HIGHLY suspect that once we discover and figure it out, ANY “slight” increase in CO2 in the atmosphere could be blamed totally on submarine volcanism. AND-because it wouldn’t be putting the particulate matter into the air, it wouldn’t cause “cooling” necessarily but the CO2 and methane and sulphates etc would STILL be affecting climate.

    ANYONE? Can I get a serious “scientist” to look into this as a serious contributor to our climate? Here are just a FEW of the hundreds of links I have to get you started:

    http://volcano.oregonstate.edu/submarine

    http://www.iceagenow.com/Archived_Articles-2011.htm (this guy might not be right about all of his info, but he’s been tracking underwater volcanic activity for a long time and has links up the wazoo on his “old website”)

    http://oceantoday.noaa.gov/deepoceanvolcanoes/

    http://www.foxnews.com/scitech/2011/10/21/explosive-underwater-eruptions-are-deepest-yet-seen/

    http://www.mnn.com/earth-matters/wilderness-resources/stories/underwater-volcanos-eruption-kills-fish-offers-clues-to-c

    Anyone? Thanks :-)

  107. Stephen Wilde says:
    September 22, 2013 at 1:59 pm

    Willis Eschenbach says:
    September 22, 2013 at 1:40 pm

    I don’t accept every aspect of the Jelbring hypothesis, merely the part that emphasises the function of winds in redistributing energy. I don’t see the mass/gravity issue as his since to my recollection it was once the consensus view.

    I am aware of your antipathy to the gravity/atmospheric mass issue but respectfully consider you and all those who support you to be wrong.

    I have no problem with that. However, I have respectfully asked you to do your respectful consideration of that issue elsewhere. And I do have a problem with you bringing it up again. It’s not up for debate. It is divisive, it leads to food fights, and it attracts trolls. Jelbring and any discussion of his work are like some kind of bizarre sub-etheric signal to the lunatic fringe to start piling on. I’m not interested. I know Jelbring is wrong, I’ve proven it to my satisfaction, and the smartest and best scientists I know agree with me. Appeal to authority? I don’t mind if the person actually IS an authority, and I’d say Robert Brown is.

    So I ask you politely again …

    Take it elsewhere. This is not the thread for that. Tallbloke has lots of threads discussing that very thing. Go there or somewhere else with it, Jelbring in any guise is not welcome on my thread.

    w.

    PS—There’s backstory. I’ve been fighting this pernicious nonsense of Jelbrings for over a decade now … so when some starry-eyed newbie comes in going “Ooooo, Jelbring” I reach for the airsickness bag. Take it away, Stephen. Not interested.

  108. This does not mean I think Willis is wrong. I think Willis is uninformed-as is most of the scientific community about how MUCH volcanic activity is happening that no one is reporting AND how that can be extrapolated into how much COULD BE happening and we just can’t record/measure yet.

  109. “Tambora’s 1815 outburst was the largest volcanic eruption in recorded history”

    Yet, HADCET was barely perturbed other than the temperatures staying below normal which started years earlier.

    Year  Jan  Feb Mar  Apr  May  Jun  Jul  Aug  Sep  Oct  Nov  Dec Annual
    1806  4.2  4.3 5.1  6.8 12.1 14.9 15.4 16.2 13.4 10.6  7.8  6.8   9.80
    1807  2.8  3.7 2.9  7.7 11.8 14.2 17.1 16.9 10.5 11.4  2.9  1.9   8.65
    1808  2.6  2.8 3.2  5.8 13.7 14.8 18.4 16.7 12.7  7.2  6.0  2.2   8.84
    1809  2.0  5.7 6.0  5.2 13.1 13.7 15.1 14.8 12.7 10.2  4.6  4.1   8.93
    1810  2.2  3.5 4.9  8.2  9.2 14.6 15.2 14.6 13.9  9.8  5.4  3.6   8.76
    1811  1.2  4.6 7.1  8.9 12.8 14.1 16.1 14.4 13.7 12.3  7.7  3.1   9.67
    1812  2.6  5.3 3.5  5.5 10.9 13.0 14.2 14.3 13.2  9.3  4.9  1.7   8.20
    1813  1.9  5.8 6.8  7.6 11.6 13.6 15.0 14.5 12.5  8.1  4.3  2.8   8.71
    1814 -2.9  1.4 2.9  9.6  9.2 12.2 16.0 14.7 12.8  8.1  4.7  4.3   7.75
    1815  0.3  6.5 7.3  8.1 12.6 14.3 14.9 15.3 13.4 10.3  3.4  2.3   9.06
    1816  2.7  2.1 3.9  6.6  9.9 12.8 13.4 13.9 11.8 10.3  3.9  3.1   7.87
    
  110. Good informative post Willis. It made me wonder about equilibrium mechanisms.

    What about micro-organisms as one of the equilibrium agents? They are everywhere in abundance and react to changing circumstances on short time scales. With this picture in mind you can easily see them as an active part of the weather system:

    They did find a lot of them up there:

    http://www.pnas.org/content/110/7/2575

    ”Quantitative PCR and microscopy revealed that viable bacterial cells represented on average around 20% of the total particles in the 0.25- to 1-μm diameter range and were at least an order of magnitude more abundant than fungal cells, suggesting that bacteria represent an important and underestimated fraction of micrometer-sized atmospheric aerosols.”

    Did read a book from Lovelock years ago. But I am not thinking macro-mechanisms here, just micro-organisms and shorter timescales. Guess this is speculative stuff still, not easily studied.

    I know of one intriguing example of temperature equilibrium in living organisms, apart from the warm-blooded animals: tree-leaves, from this post on WUWT:

    http://wattsupwiththat.com/2008/06/13/surprise-leaves-maintain-temperature-new-findings-may-put-dendroclimatology-as-metric-of-past-temperature-into-question/

    So live can be very adept in maintaining an equilibrium. I would speculate the biosphere is one of the players in the game.

  111. Latitude says:
    September 22, 2013 at 5:10 pm

    It’s insignificant. Outbreak in 1991 occurred in year with then highest global production. Climate does affect fungal diseases, of course, but US scab had almost no measurable affect on world production.

    To attribute global decline in wheat production to scab in US is, excuse me, simply nuts. Affect of drought in Australia was much greater, which began in second half of 1991, coincidentally after the Pinatubo eruption.

  112. Philip Bradley says:
    September 22, 2013 at 2:14 pm

    John Daly used 6 different methods to calculate 2XCO2 sensitivity and all of them were in line with your result, around 0.2C.

    http://www.john-daly.com/miniwarm.htm

    Looking at the 7 major eruptions from Krakatoa onwards, 5 of them occurred in spring to mid-summer.

    Mount Pinatubo Luzon Volcanic Arc Jun 15, 1991
    5 Mount St. Helens Cascade Volcanic Arc May 18, 1980
    6 Novarupta Aleutian Range Jun 6, 1912
    6 Santa María Central America Volcanic Arc Oct 24, 1902
    5 Mount Tarawera Taupo Volcanic Zone Jun 10, 1886
    6 Krakatoa Sunda Arc Aug 26–27, 1883
    5 Cosigüina Central America Volcanic Arc January 20, 1835

    Plus El Chichon, March/April 1982

    This is why climate science is such a mess. People are pattern-finding machines, and in far too many cases, they find what are not patterns. That’s why we invented statistics.

    The binomial distribution says that if you flip a coin seven times, you get five or more heads A QUARTER OF THE TIME!!! And that’s exactly the odds of five of seven volcanoes occurring in half of the year.

    So you look at that result, which has NO STATISTiCAL SIGNIFICANCE AT ALL, and you build a whole theory about how volcanoes operate out of it …

    Folks, don’t come in here without a fist full of numbers to back your play. I’m tired of this kind of nonsense. We’ve been doing this for some years now, so it’s time to man up, stop the handwaving, and RUN THE NUMBERS before you uncap your electronic pen.

    Sorry to make an example out of your foolishness, Philip, it’s not personal. I just can’t tell you how tired I am of people claiming significance where none exists.

    w.

  113. Forrest M. Mims III (September 22, 2013 at 1:33 pm) “The aerosols reduced the photocurrent from the solar cell at noon by about 5 percent during this time. These observations were associated with a reduction in temperature of about 2 degrees F, which is similar to other reports elsewhere”

    What happened at night? My guess is that nights became warmer.

  114. sunshinehours1 says:
    September 22, 2013 at 5:13 pm

    There were prior volcanic eruptions, too.

    Not summer temperatures in 1816, among lowest ever recorded in the CET series.

    To quote yourself:

    HADCET: http://www.metoffice.gov.uk/hadobs/hadcet/mly_cet_mean_sort.txt

    out of 353/354

    1816
    Coldest July ever
    20th coldest August
    26th coldest June
    38th coldest May
    50th coldest April
    34th coldest September
    238th coldest October – the outlier
    23rd coldest November
    11th coldest December

    1817
    3rd coldest May
    15th coldest July
    10th coldest August
    2nd coldest October.

  115. David Douglass says:
    September 22, 2013 at 2:27 pm

    About six years ago my colleagues and I published a series of papers on Pinatubo. The first paper provoked two comment papers by prominent climate scientists disagreeing with our results. We published replies to both showing that their analysis was wrong. All five of these papers were published in Geophysical Research Letters (GRL).

    Douglass, Knox, Pearson and Clark (DKPC) published a sixth summary paper “Thermocline flux exchange during the Pinatubo event” also in GRL.
    (http://www.pas.rochester.edu/~douglass/papers/Douglss_Knox_pearson_clark2006GL026355.pdf). We found that the delay was 4.4 months and more importantly that the total heat flux integrates during this event integrates to zero — i.e. The temperature of the Earth after the event returns to what it was before.
    ——————————————-

    ABSTRACT
    “We analyze the temperature anomaly of the Pinatubo eruption using an exact mathematical solution of a standard energy balance model that includes coupling between the mixed layer and the thermocline. Our solution yields a short response time t = 4.4 months and a small climate sensitivity l = 0.22 C/(W/m2), implying short-term negative feedback. Also, our analysis determines a value of the effective eddy diffusion constant k = 2 x 10 (-6) m2/s that is much smaller than that assumed in many climate models. We find for this model that the heat flux to the thermocline reverses sign and integrates to zero for any forcing of finite duration. This effect should be observable in any future Pinatubo- type event.”

    A voice of sanity in the wilderness. Your paper was excellent, David. I independently found the same thing, that the heat flux integrates to zero, in my analysis of the stacked volcanoes, and I also showed it and discussed the mechanisms in my post on Pinatubo.

    As you point out, however, publishing in the journals as you have done hasn’t changed the IPCC reports in the slightest. Neither has my posting here, for that matter. So I see no other path than for both of us to keep plugging away. The good news is that the truth will eventually come out … the bad news is that it may take years.

    But hey, I’ve always been in it for the long run from the beginning … and although at the start I didn’t think it would be this long, no matter.

    Thanks for all your work on these questions, David. Folks, David’s stuff is solid, have a look.

    w.

    PS—our estimates of the climate sensitivity are within 0.04 °C per W/m2 … nice.

  116. galileonardo says:
    September 22, 2013 at 2:43 pm

    Willis, I am a bit perplexed by the volcanic forcing dataset you graphed. I am not doubting the veracity of the dataset, it being from GISS and all…sorry. No, seriously, it is likely accurate, but why then is the forcing from Novarupta in 1912 so small? I wrote about it and the beheading of Katmai and other “extreme weather” events of 1912 recently so it is fresh in my mind. It was supposedly the “most powerful volcanic eruption of the 20th Century”:

    It’s not just how big, it’s where. I think that in part it was because it was so far North, so the ejecta might not have stayed aloft as long and certainly wouldn’t affect the world. If you look at the dataset, the peak looks like this:

    Globe, NH, SH
    -0.87, -2.04, -0.01

    All I can do is use what I’m given. However, analyzing the NH and SH separately showed no difference in the forcing or the R^2.

    w.

  117. Steven Mosher (September 22, 2013 at 4:19 pm) “Imagine if I tried to calculate the sensitivity from the drop in termperature when the sun goes down.”

    Imagine if you tried calculating sensitivity without the correct daily rise in temperature over the ocean that creates the convection that cools the planet (i.e. one of the thermostats being referred to here). That’s what climate models do, see fig 3: http://echorock.cgd.ucar.edu/cas/adai/papers/DaiTrenberth_JC04.pdf

  118. Willis,

    try it again using RAW Data if you can find it. You may be a victim of the reduction of historic temps.

  119. Hmm, my guess might actually be correct: “it is shown that the dominant decreasing trend of mean maximum temperature and the dominant increasing trend of mean minimum temperature over periods 1992–1994 and 1985–1987 relative to that over the period 1988–1990 are consistent with the distribution of stratospheric volcanic aerosols and predictions from aerosol radiative forcing in the southeastern U.S.”

    From http://www.sciencedirect.com/science/article/pii/S1352231097002446

  120. Willis, if you look at my links you’ll find references to Douglas’ paper. Hmm arguably one of the first words, but not the last word in trying to estimate sensitivity from volcano data.

  121. milodonharlani says:
    September 22, 2013 at 3:12 pm

    suyts says:
    September 22, 2013 at 2:25 pm

    Wheat production, as I noted, didn’t recover for seven years, not a short time.

    I wasn’t arguing for a major effect from Pinatubo, but merely pointing out that Willis overstated the case. There was at least one major crop that did show a negative response.

    DO THE NUMBERS! SHOW US THE DATA.

    Look, suppose you’re right. Suppose one major crop went down …

    How many major crops are there? A dozen? In a given year what are the odds of one of them being down?

    Next, there’s nobody saying the effects of volanoes last seven years … so IF wheat production went down for seven years, that DECREASES the odds it was volcanic in origin.

    Anyhow, onwards to the data. I’m sure you’ve heard of “Data”, although you’re not using it. Here’s the FAO data:

    Look at the seven years pre-Pinatubo versus post-Pinatubo, and you’ll see that your claim is absolute bosh. It’s just your fantasy. There is no sign of Pinatubo in that at all. Your kind of BS is why people continue to believe things that aren’t true. You’re working for the forces of darkness with that unfiltered hogwash.

    I said above I won’t stand for this “Oooh, I see a pattern, I see a pattern!” nonsense, and I plan to point and laugh at people stupid enough to continue to do it.

    DO YOUR HOMEWORK!! This is not the place to try to pass your false gold for real. It won’t work here.

    W.

  122. wayne says:
    September 22, 2013 at 4:08 pm

    Right off the bat I am wondering where (and exactly who) obtained the radiometers back in the late 1800′s to supply the aerosol data to us today that I.P.C.C. is then using to supply to politicians to make new laws and regulations?

    ( Willis, gee, I’m a climate heretic too (two years now), glad to have you onboard )

    Now I’ll read the rest.

    Since I’ve been a heretic for a decade or so, your attempt to be all pally by welcoming me on board is an insult … likely unintended, but that just makes it worse.

    Now read the rest.

    w.

    PS—If you want the history of the data, how about you READ THE FREAKING PAPER of Sato et al. … for a man claiming to be a long-term skeptic you seem like a newbie, but perhaps that’s just me. In any case, the paper always explains the numbers, sometimes better, sometimes not so well, but that’s where you need to start.

  123. Steven Mosher says:
    September 22, 2013 at 4:19 pm

    sadly you cant calcuate sensitivity that easily from the response to volcanoes.
    Imagine if I tried to calculate the sensitivity from the drop in termperature when the sun goes down.

    If you want to extract sensitivity information from the relaxation response you have to do a bit more work

    http://www.gfdl.noaa.gov/cms-filesystem-action?file=research/weather-atmos-dynamics/wallace_held.pdf

    Here is some more background. with references to papers one should read

    http://ceres.larc.nasa.gov/documents/STM/2007-11/ce0711151415Boer.pdf

    you need more than a back of the envelope.

    Thanks, Steven. I’ve shown that I can duplicate the models’ global temperature output with a one-line equation which fits easily on the back of an envelope.

    So I fear that your continual claim that it’s oh-so-complex is completely contradicted by the ultimate simplicity of what the models are doing.

    And since I did the exact same analysis on this that I did on the models … I fear your protests ring hollow.

    Next, you have to remember that I think that the climate sensitivity is a non-linear function of the temperature, and meaningless in any case in a governed system such as the climate … so I’m not analyzing it my way.

    I’m analyzing it your way, using the exact same procedure that perfectly emulates the models that you seem to believe in.

    Finally, below you say the Douglass paper is on your list … but I get results that are indistinguishable from Davids … so my results agree with those from a more complex analysis.

    w.

  124. “dp says:
    September 22, 2013 at 4:24 pm
    The sun doesn’t go down – it goes somewhere else. It is not an impulse when the sun goes somewhere else.”

    The sensitivity parameter is a measure of the system response over a given time period.

    Lets take a simple example. The speed of your car is D/T

    Now, lets apply a force to your rear tires by stomping on the pedal. there will be an instaneous response, a transient response as you gain speed, and an equlibrium response when you hit top speed.

    The ECS is what we are mostly interested in, the full response after all feedbacks,
    Estimating that from a tap on the breaks requires a bit more work than Willis presents.
    It’s an active area of research. What I can say is that most of the models cant replicate the effect perfectly they tend to over estimate the cooling and overestimate the rebound. This has been known for a while.
    At first I surmised it had to do with sensitivity. But it doesnt.

  125. Joe Born says:
    September 22, 2013 at 4:35 pm

    Willis Eschenbach: “T2 = T1 + lambda (F2 – F1) (1 – exp(-1/tau)) + exp(-1/tau) (T1 – T0)”

    As my excuse for choosing this time to pick that particular nit, all I can say is that you’ve given this formula more than once before, and each time I’ve failed to understand how it applies to the usual situation with which you’re dealing, i.e., to matching data representing averages over intervals that are potentially significant fractions of the system’s major time constant.

    Thanks, Joe. And yet, I’m able to perfectly match the outputs of the models … so I don’t understand your protest.

    Also, this is monthly data … how am I “averaging over intervals” that are significant fractions of the time constant? Sure, I’d prefer daily data … but it’s hard to argue with success, and an R^2 of 0.98 is success in my book.

    w.

  126. LdB says:
    September 22, 2013 at 4:49 pm

    @Willis Eschenbach

    I have real issues with your idea I am afraid Willis and I do not care what relationship you can show on a graph … CORRELATION DOES NOT EQUAL CAUSATION.

    Instead, I say that a host of emergent thermostatic phenomena act
    quickly to cool the planet when it is too warm, and to warm it when it is too cool.

    So for that to be remotely believed you need to provide mechanisms by which it would do it and correlating a few lines on a graph is far short of that, I think I would rather believe that the decrease in the number of pirates is causing global warming because that is a near perfect fit.

    I have written extensively on WUWT about the exact mechanisms involved, maybe twenty or thirty posts in all. So you’re just revealing that you haven’t done your homework. You could start here and work forwards … and cut down on the attitude. You just walked in and you don’t seem to have a clue, so more questions and less assertions would go a long way.

    w.

  127. Jim G says:
    September 22, 2013 at 5:10 pm

    A true super eruption of a super volcano might be at odds with your “self regulating” surface temperature hypothesis.

    We’ve had supervolcanoes in the past, and the temperature has always recovered. Under the models’ view, that wouldn’t happen … with my hypothesis, it would.

    w.

  128. an ambiguous ‘central’ in the headline; the Nobel Prize gets a mention of course, plus some classic quotes:

    23 Sept: BBC: Matt McGrath: Global warming pause ‘central’ to IPCC climate report
    Scientists will underline, with greater certainty than ever, the role of human activities in rising temperatures.
    But many governments are demanding a clearer explanation of the slowdown in temperature increases since 1998.
    One participant told BBC News that this pause will be a “central piece” of the summary…
    In the latest draft summary, seen by the BBC, the level of scientific certainty has increased…
    This slowdown, or hiatus as the IPCC refers to it, has been leapt upon by climate sceptics to argue that the scientific belief that emitting carbon dioxide into the atmosphere increases the temperature of the planet, is wrong.
    Scientists have attempted to explain the pause in a number of ways, with many arguing that the Earth has continued to warm but that the heat has gone into oceans…
    But there is no certainty and little agreement among scientists on the mechanisms involved…
    Prof Arthur Petersen is the chief scientist at the Netherlands Environmental Assessment Agency and part of the Dutch delegation that will review the IPCC report.
    “Governments are demanding a clear explanation of what are the possible causes of this factor,” he told BBC News.
    “I expect that this will be a central piece of the summary.”…
    Any changes to the text will need to be approved by the scientists, who will want to make sure that they are consistent with the underlying reports. This could lead to some tense moments.
    “I wouldn’t say there is a reluctance of the authors to take up such an issue as the pause, but they want to do it in a proper way,” said Prof Petersen.
    “There will remain a tension between how much you can deliver based on the peer-reviewed science and what the governments would like to have.”***
    ***In the wake of that year’s report, a small number of embarrassing errors were detected in the underlying material. The organisation’s reputation was also questioned in the Climategate rumpus.
    “Overall, the message is, in that sense more conservative I expect, for this IPCC report compared to previous ones,” said Prof Petersen.
    “The language has become more complicated to understand, but it is more precise.
    “It is a major feat that we have been able to produce such a document which is such an adequate assessment of the science. That being said, it is virtually unreadable!”***

    http://www.bbc.co.uk/news/science-environment-24173504

  129. Stephen Wilde says:
    September 22, 2013 at 11:39 am
    One has to ask what sets the governing mechanism and in my view it must be atmospheric pressure which is a consequence of mass alone held within a gravitational field:

    http://www.newclimatemodel.com/the-setting-and-maintaining-of-earths-equilibrium-temperature/

    —————————————————-
    Excellent read Stephen but does that mean temperature/climate/weather are somehow affected or controlled by earth’s gravity?
    I believe the gravity vector is different throughout the world but it is also constantly changing.
    Does that mean as gravity shifts due to earth quakes and core movement we get “climate change”?
    Or would these changes be too small to really affect atmospheric pressure?
    cn

  130. Willis Eschenbach says:
    September 22, 2013 at 5:51 pm

    I showed the data for yield & production & linked to those figures & the price numbers. Don’t know how you could have missed them. Your own FAO chart shows exactly what I posted, ie that wheat production fell from 1991 & didn’t recover until 1998. It’s you who ignore the data.

    Your claim was that every crop of every type did not fall after Pinatubo. That is simply false. Wheat is the number two crop in the world by tonnage & number one by area planted. Yield & production went down after Pinatubo, & price went up. Sorry if you don’t like inconvenient truths, but that’s a fact.

    That you don’t like this fact hardly means that I’ve gone over to the dark side. Quite the opposite. I rely on facts & data even when I might wish them to be otherwise.

    I knew you were wrong about wheat. I didn’t check other crops, but since you were wrong about wheat, could be you’re wrong on the others as well, although my recollection is that corn went up. In science it’s best to avoid categorical statements, especially without checking.

  131. “The amazing thing to me is that this urban legend about volcanoes having some big effect on the global average temperature is so hard to kill. I’ve analyzed it from a host of directions, and I can’t find any substance there at all … but it is widely believed.”

    One problem seems that you using air temperature.
    So I don’t Earth is warmed by changes in air temperature.
    So what would expect is for the oceans to gain less energy.
    And I think people think volcano affect global average temperature because
    affects the weather- you get cooler days. As in it causes it to snow in the summer.
    Such weather may not show up in temperature records as significant.
    It also seems that dust in atmosphere will have less affect at noon compared to when
    sun is lower in the sky. So it could have less to do with daytime highs or nighttime
    lows.
    So tropical temperature could not affected as much as higher latitude regions.

    So to measure affects one needs to careful measure ocean temperature in top
    100 meters- I think it affects entire ocean, but top 100 meters would easier to measure.
    Or measure long term affects of slightly cooler ocean.
    And I think matters how big the eruption is. Pinatubo ejected about 5 cubic kilometers,
    which quite different than ejecting 100 cubic km.

  132. Tom McCord says:
    September 22, 2013 at 9:58 am

    Is it possible that the “Year Without a Summer” in 1816 was not really caused by the eruption of Mount Tambora in Indonesia after all?
    ———————————————–
    Looking at Tony Brown,s reconstruction of the CET chart 1816 appears to be the heart of the Dalton Minimum. There are 3 years together that form the low point. The first of which is about -1.2, then -1.7C, the last shows -1.5+C.

  133. It’s so easy being in climate alarmism. You can get rid of the missing meat through volcanos, the deep sea, or maybe the hot water at the bottom of the ocean is being subducted at ocean trenches (funding for deep sea missions?), or going into deep sea volcanos. Maybe the soil is hotter now?. The possibilities are just endless.

  134. Joe Born says:
    September 22, 2013 at 4:35 pm

    You may have an interesting point. But how about you post a graph produced by you code so that we can see what the effect is without every man jack of us having to do it just to see your point?

    It seems you take a deliberately extreme case of tau=2x sample interval (which is fine to make the point) and imply that tau=5x sample may also be corrupted. Paul_K agreed with the basic formula , it don’t recall him commenting on what you say here.

    One added word of caution is needed on the use of iterative response formulae like the one in question. They can take a very long time to spin up, ie converge to an accurate result from the initial conditions.

    I looked into the possibility of using iterative filters to low-pass temperature time series. Often to get within even 5% accuracy it look more that half the length of the dataset. (Depends upon specifics.)

  135. Willis said,

    The binomial distribution says that if you flip a coin seven times, you get five or more heads A QUARTER OF THE TIME!!! And that’s exactly the odds of five of seven volcanoes occurring in half of the year.

    So you look at that result, which has NO STATISTiCAL SIGNIFICANCE AT ALL, and you build a whole theory about how volcanoes operate out of it …

    Sorry to make an example out of your foolishness, Philip, it’s not personal. I just can’t tell you how tired I am of people claiming significance where none exists.

    I don’t take this stuff personally. Although, I don’t why you blew up at me. Fig 5 says to me that volcanic cooling is limited by one or more sub-annual processes, and monsoons are an obvious candidate. If so, then time of year of the eruption will affect the lag to maximum cooling.

    I wasn’t constructing a theory from 7 data points. If anything I was making a prediction based on what I know of monsoon processes. I’ll see if I can find some data that supports my prediction or not.

  136. Willis: “The amazing thing to me is that this urban legend about volcanoes having some big effect on the global average temperature is so hard to kill. ”

    This was part of the cold war government by fear strategy. Nuclear winter was based of the same idea and made reference to the already accepted idea about volcanoes. The principal has been embedded for at least two generations. It is not going to be displaced because Willis said so on WUWT. It takes more than that to change an orthodoxy.

    However, we all seem agreed that there is some effect. It’s just that the IPCC are spinning it to be many times larger in order balance their GHG forcings being many time larger than life too.

    The problem is Chichon and Mt P coincided with cooling periods that were already under way, but anyone expecting to see a volcano effect will get bias confirmation and latch onto a false attribution.

  137. HadCRUT4 monthly global surface average air temperature look just like this Historical Total Solar Irradiance Chart.

    http://lasp.colorado.edu/lisird/tsi/historical_tsi.html

    Picture of chart Hansen did in 1988, Solar Irradiance and N hem temperatures.

    Biggest puzzle piece of Global temperature variations I feel.

    The SUN

  138. Latitude says:
    September 22, 2013 at 7:27 pm

    Wheat yield went down, production went down, supply went down, so yes, as you’d expect, prices for the commodity went up. Being a wheat rancher, I was perforce also a commodity trader then.

    In 1992, farmers, brokers & traders blamed the ash & SO2 from Pinatubo at least in part for abnormal WX patterns & cool spells in the Midwest & elsewhere in the US, including the famous Father’s Day Freeze. A mild winter & warm, early spring preceded a cool, wet summer & cold, early fall in 1992. Corn was slow to dry that fall & test weights were low, but national average yield set a record, unlike wheat.

    Also, California suffered heavy rains during 1991-93.

    http://www.nytimes.com/1992/02/13/us/new-california-storm-brings-worst-floods-in-decades.html

    http://ks.water.usgs.gov/pubs/reports/wsp.2499.sumca0193.html

    There was also the Inauguration Day Windstorm of January 1993 in Washington State. None of this may have anything to do with Pinatubo, of course.

    But regardless of the causes, the fact is that Willis was wrong to assert that yields of every single crop went up after Pinatubo. Wheat yield in the US fell, as did global production, by a lot. That’s my point.

  139. Chuck Nolan says:
    September 22, 2013 at 6:19 pm

    Stephen Wilde says:
    September 22, 2013 at 11:39 am

    One has to ask what sets the governing mechanism and in my view it must be atmospheric pressure which is a consequence of mass alone held within a gravitational field:

    http://www.newclimatemodel.com/the-setting-and-maintaining-of-earths-equilibrium-temperature/

    —————————————————-

    Excellent read Stephen but does that mean temperature/climate/weather are somehow affected or controlled by earth’s gravity?
    I believe the gravity vector is different throughout the world but it is also constantly changing.
    Does that mean as gravity shifts due to earth quakes and core movement we get “climate change”?
    Or would these changes be too small to really affect atmospheric pressure?
    cn

    Both of you, go read my post entitled “A Matter Of Some Gravity“. It proves, not asserts but proves, that your mechanism won’t work to heat the surface.

    Not only that, it also proves that NO mechanism depending on gravity can heat the surface.

    This is why I don’t like what I call “pressureheads” on my threads. I’ve shown you can’t do it, but you keep showing up.

    TAKE IT ELSEWHERE! I’m not interested in speculations about impossible mechanisms that depend on pressure/gravity. I’ve shown they can’t work, so please, you’re free to discuss it … anywhere but here.

    w.

  140. Willis

    A timely article which for the main part details points raised in your earlier articles, but of course, given the impending claim by the IPCC it is relevant to consider volcanos and whether the forcings associated with such and the effect that these claimed forcings have on temperatures.

    First, it should be noted that often temperature trends pre date the vocano eruption, by which I mean there is often a short term downward temperature trend occurring shortly before a volcano eruption and this masks the effect, if any, of the eruption since it is not known whether that short term trend would or would not have continued but for the eruption.

    So for example, consider:
    (i) Krakatoa (1883). my eyeballing of Fig2 and 4 suggests that temperatures were trending downwards as from about 1875 and Krakatoa just happened to erupt in the middle of a period when temperatures were already falling. The fall in temperature around the time of Krakatoa, does not appear to be any greater, ie., the eruption does not appear to have added anything over and above natural variation in the midst of an already pre-existing downward trend. According to Wiki: “Average global temperatures fell by as much as 1.2 degrees Celsius in the year following the eruption.” but I cannot see that that claim is sound when one looks at the data you have set out. The claim appears greatly exagerated.
    (ii) Novatubo (1912) my eyeballing of Fig2 and 4 suggests that temperatures were trending downwards as from about 1910 and Novatubo just happened to erupt in the middle of a period when temperatures were already falling. The fall in temperature around the time of Novatubo, does not appear to be any greater, ie., the eruption does not appear to have added anything over and above natural variation in the midst of an already pre-existing downward trend.
    (iii) Pinatuboa (1992), my eyeballing of Fig2 and 4 suggests that temperatures were trending upwards as from about 1980s and Pinatoa just happened to erupt in the middle of a period when temperatures were already rising. One can see a possible response to Pinatubo in Fig4, but this is very short lived.According to Wiki: “Global temperatures dropped by about 0.5 °C” and that claim could conceivably be reasonable based upon the data that you have set out, but if so, the depression in temperature was very short lived.

    To properly evaluate these claims, I consider that better resolution is required and it would be useful to set out the temperatures on a monthly basis for the 7 years before and after each eruption so one can see what is going on and whether the eruption adds anything significant to what ever short lived trends were already occurring in and around the time of the eruption.

    Second, the claimed forcings associated with Pinatubo and Krakatoa are remarkably similar, and yet no one in their right mind would consider that Pinatubo was nearly on a par with Krakatoa. I do not know whether the claimed forcing with respect to Karakatoa has been under assessed or whether the claimed forcing with respect to Pinatubo over assessed, but I am highly sceptical of the claim that they are nearly similar to one another. Whilst I do not like Wiki, the comments quoted from their site (of up to 1.2degC cf up to 0.5degC) suggests that the forcings canot be similar.

    Third, it is strange that the forcing with respect to the 1912 eruption (Novarupta) is so small given that it was the largest eruption of the 20th century, ie., bigger than Pinatubo. One opines whether the forcing has been assessed to be small simply because the temperature fall (if any, my comment in (ii) above refers) was so small that it would appear ridulous and indeed contradictory for climate scientists to ascribe a large forcing to this eruption.

    Fourth, whilst i agree that there is a lag before ocean temperature responds, I do not see why there should be any significant lag to land based temperatures, ie., atmospheric temperature. The atmosphere has little latent heat capacity so responds quickly to change. One can see this on a sunny day when clouds appear; the temperature drops within minutes. Most people have experienced significant temperature changes from one day to the next; it being not uncommon for there to be changes of 10degC from one day to the next and this demonstartes how little latent heat capacity the atmosphere has. Indeed, if the atmosphere possessed significant latent heat capacity, night time temperatures would be far more similar to day time temperatures (as you know over the oceans there is very little diurnal range, about 1 degC because the ocean which has a large heat capacity continuously warms the air above it such that the air temperature reflects very closely the ocean temperature night or day). Additionally, there is in most regions a quick response between summer and autumn, and autumn and winter which once again demonstrates that there is little lag. I would therefore suggest that you might like to give further consideration to lag at least as regards land based temperatures.

    In summary, if you go through each and every eruption evaluating the claimed forcing against temperature change, there is no correlation (some small eruptions arguably appear to have large effects on temperatures and some large eruptions appear to have all but no effect on temperatures, and in the main, there is a downward temperure trend pre-existing the eruption in question), and as I have said, numerous times before, this all appears to be a fudge which has been necessitated because of too high a sensitivity being given to CO2 forcings.

    I am even more suspicious given the latest claims that druing a period when there has been no notable volcano activity (late 1990s) volcanos are being used as a reson (I would say excuse0 to help explain the lack of warming. Surely, this is a patently bad claim by the IPCC and does not stand up to even cursory scrutiny.

  141. milodonharlani says:
    September 22, 2013 at 6:25 pm

    Willis Eschenbach says:
    September 22, 2013 at 5:51 pm

    I showed the data for yield & production & linked to those figures & the price numbers. Don’t know how you could have missed them. Your own FAO chart shows exactly what I posted, ie that wheat production fell from 1991 & didn’t recover until 1998. It’s you who ignore the data.

    Since it also fell at almost exactly the same rate BEFORE Pinatubo, your claim is ignoring the data. Yes, it fell … what I said was that Pinatubo had very little effect. And the data bears that out.

    Sorry, but simply pointing to an index that has been falling for seven years, and then claiming that the decrease in the following seven years from a volcanic eruption doesn’t pass the laugh test.

    w.

  142. Philip Bradley says:
    September 22, 2013 at 7:53 pm

    Willis said,

    The binomial distribution says that if you flip a coin seven times, you get five or more heads A QUARTER OF THE TIME!!! And that’s exactly the odds of five of seven volcanoes occurring in half of the year.

    So you look at that result, which has NO STATISTiCAL SIGNIFICANCE AT ALL, and you build a whole theory about how volcanoes operate out of it …
    Sorry to make an example out of your foolishness, Philip, it’s not personal. I just can’t tell you how tired I am of people claiming significance where none exists.

    I don’t take this stuff personally.

    You should. When you make dumb claims like that, it’s you that looks foolish, not me or the rest of the folks. Run the numbers first, and folks won’t be laughing at you.

    w.

  143. David Douglass says:
    September 22, 2013 at 2:27 pm

    About six years ago my colleagues and I published a series of papers on Pinatubo. The first paper provoked two comment papers by prominent climate scientists disagreeing with our results. We published replies to both showing that their analysis was wrong. All five of these papers were published in Geophysical Research Letters (GRL).

    Douglass, Knox, Pearson and Clark (DKPC) published a sixth summary paper “Thermocline flux exchange during the Pinatubo event” also in GRL.
    (http://www.pas.rochester.edu/~douglass/papers/Douglss_Knox_pearson_clark2006GL026355.pdf). We found that the delay was 4.4 months and more importantly that the total heat flux integrates during this event integrates to zero — i.e. The temperature of the Earth after the event returns to what it was before.

    So, if you accept the evidence for very vivid, very sharp two-year transmission “brownout” for the most recent two volcanoes volcanoes (shown here from the WUWT solar page )

    does the much longer assumption of earlier “brownouts” (because of an assumed greater size for those earlier volcanoes) make sense from a transmission standpoint?

  144. milodonharlani says:
    September 22, 2013 at 8:15 pm

    But regardless of the causes, the fact is that Willis was wrong to assert that yields of every single crop went up after Pinatubo. Wheat yield in the US fell, as did global production, by a lot. That’s my point.

    Dang. I just went and looked up what I said. Foolish me, I actually believed you when you said I’d made a claim about wheat production … I did no such thing, I spoke only of yield. Nor did I say that “yields of every single crop went up after Pinatubo”. You’re just making things up. Here’s what I actually said:

    That was the point I was trying to make above, that if the weather really had been all that bad in 1992 the crop yield would have reflected it, and it didn’t.

    Note that NOWHERE DID I SAY THAT PRODUCTION OF EVERY SINGLE CROP WENT UP AFTER PINATUBO. That is your fantasy. Heck, I said nothing at all about production, which is what you are either foolishly or deliberately claiming.

    I said that we cannot see any sign of Pinatubo in the yield, which is a very, very different claim.

    For example, here’s the wheat yield …

    If you claim that you can see a drop in wheat yield due to Pinatubo, you’ll have to point it out.

    w.

  145. Forcings…never liked the word…a lot of these graphs and figures of Watts/sqm are just Pretended Precision…from Aphelion to Perihelion…sunlight reaching the earth varies by almost 7%….which is a lot….yet the Earth just keeps on truckin….inertia ?…well buffered system ?…whatever it is…we human beans have no say in it.

  146. Thank you, little CO2 molecule, for averting the catastrophe of volcano-induced ice-hockey-stick global freezing.
    Thank you, IPCC, for making it clear that it a very good idea to maintain high levels of carbon dioxide in the atmosphere.

  147. Schrodinger’s Cat says:
    September 22, 2013 at 11:29 am

    the models completely underestimate the way our planet can restore climate stability. The climate scientists are too busy [making their models generate instabilities and excursions]

    Since payday depends on finding and hyping (CO2-driven) instabilities, it’s not surprising that’s what the models are set to project/predict.

  148. Willis Eschenbach says:
    September 22, 2013 at 8:44 pm (Edit)
    milodonharlani says:
    September 22, 2013 at 8:15 pm

    Wheat yield in the US fell …

    No, it didn’t. You’re making things up again.

    The 1993 value was the same as 1992. And the drop in 1994, in addition to being well after the eruption, was smaller than the drop 1990-91. In other words, there’s no sign of either a world or a US yield drop due to Pinatubo, that’s just a fantasy.

    w.

  149. richardscourtney says:

    September 22, 2013 at 10:40 am
    /////////////////////////

    Richard

    I would appreciate your further thoughts/clarification on ocean temperatures..

    The paper you refer to suggests that there may be a cap on the temperature of the tropical ocean of about (may be a little less than) 305K (ie.,32degC). You explain (sorry if my para phrasing is not accurate) that at that temperature evapoation is such taht it causes cloud formation which clouds shield the ocean below from receiving as much solar as they would otherwise receive (ie., if the clouds had not formed).

    However many oceans have temperatures significantly higher than 32degC, in particular the Red Sea, the Gulf of Mexico, the West coast of Africa (between say Equitorial Guineee and Ivory Coast), some parts of the Indian ocean, some parts of the China sea and seas around Indonesia. Temperatures in these seas frequently reaches 34 degrees, 35 and 36degC is not uncommon and I have seen temnperatures as high as 38degC (incidentally, I have some 30 years experience of reviewing ship’s logs and have reviewed hundreds of thousands of temperature entries so I know that these are not rogue reports).

    Now if there is a direct and causal link between temperature and evaporation leading to cloud formation leading to blocking of solar, how come is this not operative over these oceans?

    I raised this point when Willis wrote one of his arcticles on ARGO and I provided him with a number of links detailing the then current sea temperature (ie., the temperature on the day of my comment) as recorded by local weather stations covering some of these oceans and these were reporting temperatures in excess of 32 degC (some were reporting 34degC, I think but I can’t recal for certain that one was reporting 35degC or a high of 35degC that year).. I suggested to Willis that whilst evaporation is one factor that seeks to limit temperature, it is not the only operative factor (I suggested some other factors) and Willis only slightly widened his view (iniatially he was suggesting that it was all down to evaporation).

    I don’t dispute that evaporation leading to cloud formation is a factor, but in my opinion, it can only be one of a number of different factors, since if it was the sole governor then one would not see significantly higher temperatures in the oceans/seas that I mention.

    Your further insight would be appreciated.

    PS. ARGO only has sparse coverage. It does not really cover the ocean/sea areas that I am referring to. This is probably due to depth considerations. These oceans/seas are not shallow but neither are they deep oceans. I suspect that ARGO is not deployed in these areas since it would be unable to perform its really deep dives down to 2,000m and this would then lead to bias in that more oceans would be sampled say in the 0-700m range.

  150. Stephen Wilde says:
    September 22, 2013 at 3:02 pm
    It is the increased rate of evaporation that is the primary effect arising from the addition of extra energy to the ocean surface. Willis does not explain how that increase in evaporation is always sufficient to put a firm lid on maximum sea surface temperature.
    ——————————————————————————-
    Your first sentence is a repeat of what Willis has already said. I thought that he also explained that the once the cloud formation starts that it tends to overcompensate, which is why it always achieves the extra cooling process? Afterwards, an opposite reaction reinstates the normal mode.

  151. Greg Goodman says:

    September 22, 2013 at 8:00 pm
    /////////////////////////

    According to Hadcrut4, there was a drop of some 0.3degC (or even more) between the start of 1944 and the end of that year. Was this due to the 2 nuclear bombs dropped on Japan? Or, did scientists consider that the temperature drop was probably due to the 2 nuclear bombs dropped on Japan thereby leading them to be fearful of a nuclear winter?

  152. milodonharlani says:
    September 22, 2013 at 1:08 pm

    Since outside the US, acreage didn’t drop by much (see below), lower production necessarily implied lower yield.

    I pointed out in my original comment that wheat yield in the US fell after Pinatubo, from 39.5 bu/A to 34.3. Clearly yield fell globally as well, as your own graph shows. I can see it if you can’t. But I wonder why you posted an imprecise graph instead of the actual data, which show a drop from 2.6 tonnes per hectare to 2.5 post-eruption.

    http://www.fas.usda.gov/grain/circular/2010/05-10/grainfull05-10.pdf

    Total Wheat and Coarse Grains
    Millions of Metric Tons/Hectares
    Year Area Harvested Yield Production
    1990/91 549.0 2.6 1,417.5
    1991/92 546.3 2.5 1,355.5

    But the situation is actually worse, because of China, close to the Pinatubo eruption, where yield went down in 1991, then up in 1992, contrary to global average (didn’t check India). Since China, world’s top wheat producer (US is usually third, after India, although Russia beat us in 2009 & 2011), is mainly a spring wheat country, the growth of its crop could have been affected by the June eruption that year. See Table 1 in:

    http://repository.cimmyt.org/xmlui/bitstream/handle/10883/1222/64613.pdf

    Average wheat yield in China was 3.225 T/ha in 1990, falling to 3.120 in ’91 & rising 3.392 in ’92
    (Data Sources: Statistic Books of Chinese Agriculture).

    Your whole closing comment was: “That was the point I was trying to make above, that if the weather really had been all that bad in 1992 the crop yield would have reflected it, and it didn’t. Not for any type of produce, not for tubers, not for legumes, not for vegetables, not for fruits, not for grains”. As I noted, this is not true for wheat, as the above figures show.

    http://www.agmrc.org/media/cms/ccpwheat_47A4CABBA76E0.pdf

  153. Willis, thank you for using your superb analytical skill to show what a minor impart volcanoes have on climate. I have never accepted volcanoes as being a significant climate force, but I lack your math skill and strong ability to analysis the records.

    For a decade I have been convinced that carbon dioxide has only a minor impact on climate. But, I have been greatly concerned that no one has come up with a theory and solid supportive scientific data to explain some other reason behind the shifts in climate. No one has ever explained to my satisfaction how variations in solar activity drive variations in Earth’s climate. Nor has anyone explained how variations in ocean currents and volcanic eruptions have enough impact to drive our climate shifts.

    Recently, however, I have been reading the papers from Henrik Svensmark that link cosmic rays with the formation of low clouds on Earth. He explains how the cosmic rays produce cloud droplet nucli particles and how those clouds drive the planets climate swings. I am wondering if he has discovered an important driving force behind climate shifts.

    Willis, it would be very helpful, perhaps, if you would do an analysis of Svensmark’s theory. Please consider this as a serious request for you to apply your skills to this concept.

    Thank You.

  154. Willis Eschenbach says:
    September 22, 2013 at 9:36 pm

    It’s you who make things up. I’ve repeatedly posted the actual numbers instead of crayon-drawn graphs. When will you bother to read the data I’ve given, with sources?

    US wheat yield (& acreage planted) fell after Pinatubo from 39.5 bushels per acre to 34.3, as above.

    Why don’t you bother to check facts before posting such sweeping falsehoods, so easily shown wrong, especially after having been shown the actual data more than once?

  155. Willis Eschenbach: “Also, this is monthly data … how am I “averaging over intervals” that are significant fractions of the time constant? Sure, I’d prefer daily data … but it’s hard to argue with success, and an R^2 of 0.98 is success in my book.”

    Well, perhaps 1/6 is not as large a fraction as to be considered “significant,” and it’s true that the formula’s inaccuracies are more apparent at larger fractions. But I would have thought the graph my previous comment’s code drew would give you pause about the formula’s correctness.

    I’m not saying that you’re getting wildly inaccurate results. But if you apply your formula to data from a low-time-constant system, I believe your results will overstate the time constant. If the time constant were on the order of only a month, for example, my tests (on synthetic data with no noise) indicate that you’d tend to arrive at 50% too long a time constant. And my guess is that your 6.4-month estimate is likely to be about 25% high.

    That’s not a discrepancy of great concern in this context, of course. But I thought you might want to reconsider the formula nonetheless.

    (I hasten to add that the formula I use tends for some reason to estimate a little low–but more on the order of 10% in cases where yours gets 50%–and achieves much closer matches on noiseless synthetic data.)

  156. Eric1skeptic says:
    September 22, 2013 at 12:05 pm

    “I realize that your website concludes that atmospheric composition is not important, but simulations of atmospheric columns differ from that conclusion. See hitran results in table 2:”

    When I gave Hitran a fling, I found that one of the first things you do is input a temperature profile. One of the options is the U.S. Standard profile, which is, of course, the lapse rate. So, you input the temperature profile and a CO2 concentration and it calculates downward and upward flux at various levels, including TOA. To get a sensitivity for doubling you double the CO2 concentration and fiddle with the temperature offset at the surface until you recover the original TOA flux. The surface temperature offset is then the sensitivity for a doubling. I checked the output files and yes, the program is simply adding the “offset” to the temperature profile you entered and calculating from there. Now, I had expected the temperature profile to emerge from the calculation and not be a required input. And I speculate that the reason for this is that the mechanism that establishes the lapse rate, convection, is unknown to the program and any attempt to develop a profile from a purely radiative calculation produces a disaster. So, you forge a temperature profile as an input, allowing it to blithely ignore reality.

  157. richard verney. says:
    September 22, 2013 at 10:03 pm (replying to)

    Greg Goodman says:

    September 22, 2013 at 8:00 pm
    /////////////////////////

    According to Hadcrut4, there was a drop of some 0.3degC (or even more) between the start of 1944 and the end of that year. Was this due to the 2 nuclear bombs dropped on Japan? Or, did scientists consider that the temperature drop was probably due to the 2 nuclear bombs dropped on Japan thereby leading them to be fearful of a nuclear winter?

    Please re-consider this question very, very carefully: The two (relatively small!) A-bombs were dropped over a two-day period the first week of August 1945. Immediately after that week, WWII ended, with all firebomb raids over Japan occurring March-July 1945.

    The fire bomb and blast raids over Germany ended much earlier in that same year: May, 1945 saw VE Day come and go. (The two A-bomb blasts in August 1945 followed months of fire bomb raids over other Japanese cities, but few other areas of the world outside of ever-smaller areas of Germany and western Poland were attacked after January 1945.)

    True, “some” of the limited area blast damage in 1944 over Europe “might” have impacted worldwide dust cover during 1944, but there really were very, very few raids over Japan in 1944 since the aircraft bases had not been built yet to support large raids. And, bombing raids were days-per-month basis, not every-hour-of-every-day.

  158. The IPCC report is meant to fuel doubts about the plateau. It’s the ocean, it’s the volcanoes, it’s whatever as long as a doubt is seeded in the mind of the reader… Not of course a doubt about these alarmists models and catastrophsim, no those are still right despite reality proving otherwise. It’s post modern science and defence of a failed theory.

  159. richard verney. says:
    September 22, 2013 at 10:03 pm

    The atomic bombs were dropped on Japan in August 1945, not 1944. The conventional fire raids on Japanese cities were also mainly in 1945. European cities were fire-bombed in 1944, but the two biggest fire storms were Hamburg in 1943 & Dresden in 1945.

    For the nuclear winter scam to have a hope of working, many cities around the world would have to suffer such storms, lofting soot high into the atmosphere, rather than locally as during the Kuwait oil well fires. It’s rather hard to ignite a firestorm, so “nuclear winter” is improbable even in an all-out nuclear war intentionally targeting cities.

  160. richard verney. says:
    September 22, 2013 at 10:03 pm

    According to Hadcrut4, there was a drop of some 0.3degC (or even more) between the start of 1944 and the end of that year.
    ————————————–
    The year 1944 was the bottom of the solar minimum. Doesn’t the minimum usually translate to a slightly cooler year?

  161. milodonharlani says:

    September 22, 2013 at 11:18 pm
    ///////////////////////
    Whoops. I must be going mad, or may be it was just because I have been up all night with too little sleep, or too much alcohol, or a combination of both..

  162. jai mitchell says:… If the energy going into that sphere containing the whole earth and all of it’s functions is more than the energy leaving that sphere then the planet is warming.

    To add to what Willis said, rain does indeed help transport heat out of the earth. Sunlight primarily heats the oceans, since they account for 3/4 of its surface. One of the primary ways for that heat to get back into outer space is through evaporation, which transports heat as water vapor into the upper atmosphere, where it forms clouds. When that water vapor turns to rain, that heat is released into the atmosphere, where it can be radiated away into space. So even for the entire earth system, rain is part of the cooling mechanism whereby incoming heat is gotten rid of. The fact that it also cools the surface is simply an additional benefit. Without it, more heat would remain trapped in both the oceans and the atmosphere for longer periods of time. So rain does indeed cool the system.

  163. richard verney. says:
    September 22, 2013 at 9:42 pm

    Thanks for that additional input regarding maximum sea surface temperatures.

    I wasn’t aware of the extent to which 305K can be exceeded locally.

    Having considered the matter further overnight I still don’t see how emergent cloudiness can provide any sort of cap on achievable temperatures because a cap has to be exceeded before the clouds form (at least the types of clouds proposed).

    It is true that once the clouds form they do provide a negative feedback but they don’t form at all unless the cap is exceeded.

    It has to be surface pressure because the weight of the atmosphere pressing down sets the amount of energy required to break the bonds between water molecules.

    If that breaking of bonds didn’t occur first the clouds would not form.

  164. richard verney. says:
    According to Hadcrut4, there was a drop of some 0.3degC (or even more) between the start of 1944 and the end of that year. Was this due to the 2 nuclear bombs dropped on Japan?

    HadCRUT=hadSST+CRUtem

    There was a 0.5 deg C “adjustment” made to the SST record in 1945 which I refer to as Folland’s folly in honour of its founder.

    Since you obviously are not aware of it I suggest you read my article of last year:
    judithcurry.com/2012/03/15/on-the-adjustments-to-the-hadsst3-data-set-2

    As for the bombs it’s a reasonable question to ask but is quickly dismissed. I have specifically tried to find some evidence of an effect , more likely from the much larger aerial explosion tests or the french test at Bikini Island that unintentionally reached the ocean surface , and found none.

  165. Dear Willis,

    Yo will recognise that I have been a critic of your signal processing methods. having gained a PhD in a world class DSP lab and used it for 30 years (on and off). If you want some help I am happy to advise you.

    Richard Saumarez

  166. Having looked at Willis’s original Thermostat Hypothesis again I see that he includes vapour pressure as a relevant variable but not absolute pressure.

    I think that is a significant omission because absolute atmospheric pressure for an entire atmosphere sets the average strength of the bond between water molecules on any given planet and thus the basic minimum amount of energy required to break those bonds (latent heat of vaporisation).

    Absolute surface pressure over a dry surface also sets the average amount of energy required to lift a molecule of any particular weight off the surface to form an atmosphere.

    This is relevant to Willis’s hypothesis and to this thread because it sets the basic energy requirement from which the thermostatic process starts and the baseline to which it inevitably returns.

    The thermostat deals not only with water in vapour form but also any molecules that are caused to lift off the surface in the first place. Hence the relevance to the gaseous and particulate products of volcanic eruptions.

    Without something to set that basic energy requirement there is no baseline to which any thermostatic mechanism can return.

    It is indeed all about absolute pressure and that is a function of atmospheric mass and not composition.

    That is an elephant in this particular room.

  167. substracting the gaussian average is far from neutral, and the demonstration would be much more effective showing harcut data,the several things appear clear to me, you can “see something” in recent time, but you can’t see much in the oldest period but to me it means that uncertainties of temperature may be very underestimated for this time.

    Hard to apply the same method t and he same filters on such an “object”.

  168. Willis.

    A couple of thoughts on the diurnal tropical temperature charts you showed, many moons ago, whilst explaining your tropical cumulus governor theory. The graphs showed a cooling in the afternoon, and then a slight uptick in temperature before sunset, which was unexplained. A couple of possibilities for you:

    The afternoon has decreasing convective activity, and so less cooler air from the upper troposphere is being brought to the surface. Thus the lower atmosphere is able to recover some warmth from the warm land surface, without any tropospheric cooling.

    The afternoon heralds the end of the daytime sea breeze. If these temperatures were from coastal recording stations, they would change in the afternoon from the cooler sea breeze, to the warmer land breeze. Again, you might get a late afternoon recovery in temperature.

    Just a thought.

  169. richard verney. says:
    September 22, 2013 at 8:20 pm

    “First, it should be noted that often temperature trends pre date the vocano eruption, by which I mean there is often a short term downward temperature trend occurring shortly before a volcano eruption and this masks the effect, if any, of the eruption since it is not known whether that short term trend would or would not have continued but for the eruption. ”
    ===

    I made the same point just a few posts above yours, 20m earlier.

    I also a whole series of graphs and comments I did on this that no one seems to read or comment on.

    http://climategrog.wordpress.com/?attachment_id=278

    http://climategrog.wordpress.com/?attachment_id=310

    http://climategrog.wordpress.com/?attachment_id=312

    http://climategrog.wordpress.com/?attachment_id=285

    These specifically show a marked difference in the response of the tropics and extra tropical regions.

    The method is not statistically rigorous but is prima facea evidence of the kind of non-linear feedback response Willis is suggesting. It also shows that degree.day integral is maintained in the tropics but not outside, where the temperature does recover but there is loss of degree days.

    I suggested Willis repeat the analysis he presented here on extra-tropical data. Oddly he has not even commented on the suggestion.

    There clearly is a detectable response in ex-tropics, looking for it directly may allow a better evaluation of its magnitude and significance.

    My impression was that is was land based rather than sea based, hence more marked in NH. Some zonal mixing due to ocean gyres probably allows the very stable tropics to diminish the impact on extra-tropical zones.

  170. Stephen Wilde:

    Thankyou for your post addressed to me at September 22, 2013 at 3:02 pm.

    http://wattsupwiththat.com/2013/09/22/the-eruption-over-the-ipcc-ar5/#comment-1423915

    Firstly, to ensure that everyone is clear about this, we are discussing the Ramanathan&Collins (R&C) effect and not my work.

    You ask me

    The paper you refer to suggests that there may be a cap on the temperature of the tropical ocean of about (may be a little less than) 305K (ie.,32degC). You explain (sorry if my para phrasing is not accurate) that at that temperature evapoation is such taht it causes cloud formation which clouds shield the ocean below from receiving as much solar as they would otherwise receive (ie., if the clouds had not formed).

    However many oceans have temperatures significantly higher than 32degC, in particular the Red Sea, the Gulf of Mexico, the West coast of Africa (between say Equitorial Guineee and Ivory Coast), some parts of the Indian ocean, some parts of the China sea and seas around Indonesia. Temperatures in these seas frequently reaches 34 degrees, 35 and 36degC is not uncommon and I have seen temnperatures as high as 38degC (incidentally, I have some 30 years experience of reviewing ship’s logs and have reviewed hundreds of thousands of temperature entries so I know that these are not rogue reports).

    Now if there is a direct and causal link between temperature and evaporation leading to cloud formation leading to blocking of solar, how come is this not operative over these oceans?

    OK. My first answer is that I don’t know. I merely reported the findings of R&C. So, I will mention the reaction in the literature and then state some suggestions I make as to why the effect is not observed everywhere.

    The R&C paper initially received much opposition in the literature. The main question was whether the maximum tropical sea surface temperature actually existed. R&C ‘stood their ground’ and others reported the same result so opposition to the R&C Effect withdrew.

    But the R&C effect is observed in the tropical ocean and not everywhere. So you ask the very reasonable question (which was raised in the literature decades ago) as to WHY doesn’t it happen everywhere.

    I suggest – and please note this is merely opinion – the difference is probably a combination of geography and atmospheric circulation patterns. I explain this suggestion as follows.

    The R&C Effect results in increased evapouration from sea surface and this has two direct effects; viz.
    (a) the ocean surface is cooled by extraction of heat of evapouration
    and
    (b) the air above the cooled ocean surface gains humidity (I,e. water).

    But. both theoretical calculation and – as you say – observations elsewhere indicate the direct cooling of the sea surface is not sufficient to keep the surface temperature below 305K. This is why R&C required an additional effect to explain the maximum temperature limit.

    The R&C Effect overcomes the difficulty posed by the inability of evapouration alone to establish the maximum sea surface temperature observed in the tropics. Other observations indicate that in other regions the R&C Effect either does not exist or is not sufficient to keep sea surface temperature below 305K.

    The high temperatures in the Red Sea and the Gulf of Mexico may be because the sea is surrounded by land, and cirrus shielding of the land will not cool the water. Cirrus forms at altitude (typically higher than 20,000 feet) and winds may carry the moist air over land before it rises to an altitude where it can form cirrus.

    In other places the R&C Effect may be reduced by winds, too. When the cooling by cirrus shielding occurs adjacent to the region of highest temperature then the shielded adjacent region will cool. But the maximum sea surface temperature may rise above the limit of 305K because it is not shielded from solar radiation by the cirrus which forms nearby. In this case, total heating of sea surface is reduced by the cirrus formation, but the limit to maximum sea surface temperature is not imposed in the hottest region.

    Nearly a quarter of a century has passed since the R&C Effect was discovered. In that time US$billions have been spent on climate modelling. Nothing has been spent on field work to investigate the R&C Effect.

    Richard

  171. I couldn’t find any direct evidence of time of year of a volcanic eruption influencing the lag to minimum temperature, only the indirect evidence that major volcanic eruptions reduce monsoon intensity and in particular reduce rainfall in Central and other parts of Asia.

    http://onlinelibrary.wiley.com/doi/10.1029/2010GL044843/abstract

    https://www.geo.umass.edu/climate/papers2/Fan_etalJClimate09.pdf

    As water vapour (latent heat) in the monsoon is a major mode of heat transport poleward, cooling the climate, the reduced monsoon is a season specific negative feedback to volcanic aerosol cooling. Although its hard to say how this will pan out for surface temperatures, as reduced monsoonal flow will increase the length of the pre-monsoon season. Always the hottest time of year in monsoonal zones. A negative feedback to a negative feedback.

  172. Willis: There is a flaw in your post: You are looking for a correlation between radiative forcing (W/m2 or POWER per unit AREA) and temperature, which is proportional to mean kinetic ENERGY. The larger the object, the more kinetic energy it contains; so temperature is really ENERGY per unit VOLUME. Dimensional analysis immediately tells you that any relationship between radiative forcing and temperature needs to take into account additional factors.

    To convert power into energy, one multiplies by time. A radiative forcing by itself can’t cause any temperature change, but a radiation forcing applied for several months or years can change temperature. You need to INTEGRATE the radiative forcing over time in order to have any hope of understanding how much the temperature will change. Looking for a lagged relationship between radiative forcing and temperature isn’t a very good substitute for integrating the forcing over time (as your poor R2 shows).

    You also need to take into account the volume of material whose temperature is being changed by a forcing. We know that every summer and winter, seasonal temperature change in the ocean can be detected down to about 100 m, so several months of seasonal radiative forcing is able to warm an average of roughly the top 50 m of the ocean – the mixed layer. So the radiative forcing from volcanos passing through every m^2 of the ocean’s surface also effects the temperature of about 50 m^3 of ocean.

    With the correct factors for converting area to volume and power to energy, you might have some chance of performing a useful calculation. This complicated subject has been discussed for Pinatubo at the Blackboard by Paul_K and in references he cites. He arrives at a climate sensitivity of about 1.4 degC for doubling CO2. It appears as if he might be trying to publish this result. http://rankexploits.com/musings/2012/pinatubo-climate-sensitivity-and-two-dogs-that-didnt-bark-in-the-night/

    To get started, how long will it take a radiative forcing of 1 W/m^2 to warm the 50 m^3 of mixed layer underneath by 1 degC? The heat capacity of water is 4.17 J/cm^3/degC or 4.18*10^6 J/m^3/degC. With 50 m^3 of mixed layer below every m^2 of ocean surface, 2.09*10^8 J of energy needs to enter every m^2 of ocean surface to warm 1 degC. It therefore takes a radiative forcing of 1 W/m^2 or 1 J/s for every m^2 of surface 6.6 years for the ocean to warm 1 degC!

    The real situation is far more complicated than this: Ocean covers only 70% of the surface, but the air and land have some heat capacity too. And a little energy escapes into the deeper ocean over a period of a few years. Even worse, as the ocean warms, it will emit more energy through radiative cooling. The Planck feedback (the increase in “blackbody” radiative cooling with temperature) is -3.2 W/m^2/degC. So we can actually can warm the ocean only about 0.3 degC before the increase in radiative cooling completely negates a 1 W/m^2 of radiative forcing! And half of the forcing has been negated once the temperature has risen 0.15 degC during the first year. So the correct answer is that a forcing of 1 W/m^2 can never raise the temperature 1 degC!

    These calculations suggest that it is absurd to expect a radiative forcing of a few W/m^2 – which is mostly gone in two years – to produce more than a fraction of a degC of cooling. And Planck feedback begins to counteract this cooling as soon as begins. With Figure 3 showing noisy spikes in global temperature of about 0.3 degC, it is going to be hard to convincingly see the cooling from every big volcanic eruption. BEST averaged the results from about 9 large eruptions to reduce noise by a factor of three and the cooling from volcanos became obvious. Paul_K reduced noise by analyzing the change in temperature over 18+ individual months.

    So it makes complete sense that you can’t find unambiguous evidence for cooling by inspecting the record by eye, but others can find evidence by performing more sophisticated analyses. ALARMISTS have increased your expectations for volcanic cooling far past what is expected from my “back of the envelop” calculation. For example, those who attribute the “year without a summer” in 1816 in New England to the 1815 eruption of Mt. Tambora are confusing unusually cold WEATHER in New England with transient global cooling of perhaps 1 degC. New England is not normally within 1 degC from getting snow in August, so Tambora didn’t cause the year without a summer, it simply enhanced it a little.

  173. In my view, there is a simpler way to show the absurdity of AR5 speculation.

    According to Wikipedia, there were 18 major volcanic eruptions in the 21st century, none over VEI 4.

    In the period 1991-2000 there were two eruptions of VEI 5 or more, Mt. Hudson VEI 5+ and Pinatubo VEI 6, both in 1991. (plus a number of lesser eruptions).

    Now remember that if the Volcanic Explosively Index increases by 1, the volume of erupted material is increased by a factor of 10. Which means that Pinatubo alone ejected over 20-times the material of *all* the eruptions in the 21st century.

    If the 18 21st century eruptions caused the “hiatus” in global warming – why was the climate warming in the 1990s, despite more then 20-times larger volume of volcanic emissions?

  174. richard verney. says: “To properly evaluate these claims, I consider that better resolution is required and it would be useful to set out the temperatures on a monthly basis for the 7 years before and after each eruption so one can see what is going on and whether the eruption adds anything significant to what ever short lived trends were already occurring in and around the time of the eruption.”

    Almost exactly what I did. Try looking at the links.

  175. Frank: ” You need to INTEGRATE the radiative forcing over time in order to have any hope of understanding how much the temperature will change.”

    A valid point.

    What Willis writes in Notes and Data appendix includes an iterative formula which is in fact a exponentially weighted integral. I would have thought that doing a similar integral on the forcing would be the correct way to do the simple regression. I suspect this is what the difference between his Fi and Fe data is , though this is not stated explicitly.

  176. milodonharlani (September 22, 2013 at 10:14 pm) “US wheat yield (& acreage planted) fell after Pinatubo from 39.5 bushels per acre to 34.3, as above.”

    Please post a link that shows wheat yield in the US falling from 39.5 bu/acre to 34.3 bu/acre.

    The only document you have posted with yield data is http://www.fas.usda.gov/grain/circular/2010/05-10/grainfull05-10.pdf which shows yield fluctuating between 2.5 and 2.6 for the years in question. The particular drop from 2.6 to 2.5 has no significance since many years show similar fluctuations (or larger).http://www.fas.usda.gov/grain/circular/2010/05-10/grainfull05-10.pdf

  177. Richard: Nearly a quarter of a century has passed since the R&C Effect was discovered. In that time US$billions have been spent on climate modelling. Nothing has been spent on field work to investigate the R&C Effect.

    Yes, it seems that before 1990 there was a much broader range of ‘normal’ scientific investigation. Since about that point the zealots took over, trying to ‘save the planet’ by distorting science. Defunding, gatekeeping and nobel ;) cause corruption seem to have replaced objectivity.

    That process is starting to reverse but it’s going to take years.

  178. “I say that the current climate paradigm, that forcing determines temperature, is incorrect. I hold that changes in forcing only marginally and briefly affect the temperature. Instead, I say that a host of emergent thermostatic phenomena act
    quickly to cool the planet when it is too warm, and to warm it when it is too cool.”

    Willis why use such convoluted and hard to understand language – Why not just say “I believe there is strong negative feedback in the climate system” because that’s what your comment boils down to.

  179. Frank: “These calculations suggest that it is absurd to expect a radiative forcing of a few W/m^2 – which is mostly gone in two years – to produce more than a fraction of a degC of cooling.”

    My estimation shows 0.2 degree.years in extra-tropical SH , that mostly happened between 6 and 18mths. ( similar in hadSST3 and hadCRUT4)

    http://climategrog.wordpress.com/?attachment_id=285

    NH hadSST3 shows about 0.4 degree.years withi similar form.

    http://climategrog.wordpress.com/?attachment_id=310

    That means as an average it was 0.2 (0.4) K cooler for a year, or 0.1 (0.2) K cooler for two years. then back to previous conditions. The tropics show no such loss.

  180. michael hammer says:
    Willis why use such convoluted and hard to understand language – Why not just say “I believe there is strong negative feedback in the climate system” because that’s what your comment boils down to.
    ===

    Willis is trying to refute the idea that tropical storms and cloud cover can be treated as a simple linear negative feedback globally. This is how climate models deal with it as well as being based on guestimated “parameters” not a modelled response based in science.

    Thus he dislikes use of the word feedback at all.

    Emergent phenomena are _positive feedbacks_ at work. They are ultimately constrained by more powerful negative feedbacks that prevent the system from being unstable. Since the overall effect is a negative feedback on temperature, the internal +ve f/b makes the storms into a NON-LINEAR negative feedback.

    It is the key word non-linear that makes the overshoot required to preserve the degree.day integral. A linear neg. f/b will not do this.

    This broader understanding of feedbacks seems to be outside Willis’ way of looking at things so he tends to reject any description using the word feedback, which he associates with the IPCC linear feedback which is inadequate and denies the possible role of tropical storms he is proposing.

    The self-sustaining nature of tropical storms is due to internal +ve feebacks. I think it would be useful if Willis saw that feedback based descriptions are the proof of what he is saying , not the opposite.

  181. RC Saumarez says: “…If you want some help I am happy to advise you.”

    Willis has been specific enough here for you to be able to critique his method. If you have actual suggestions then I’m sure he’d like to hear them but I’ve now seen a couple of posts of yours implying criticism of his method in a general sense without actually being specific about anything.

  182. michael hammer:

    Your entire post at September 23, 2013 at 2:33 am says

    “I say that the current climate paradigm, that forcing determines temperature, is incorrect. I hold that changes in forcing only marginally and briefly affect the temperature. Instead, I say that a host of emergent thermostatic phenomena act quickly to cool the planet when it is too warm, and to warm it when it is too cool.”

    Willis why use such convoluted and hard to understand language – Why not just say “I believe there is strong negative feedback in the climate system” because that’s what your comment boils down to.

    No! Your misunderstanding has induced you to misrepresent Willis, so I write to answer in hope of reducing the ear-bashing which your misrepresentation deserves.

    There is no definition of emergent effects such as the Eschenbach Effect, the R&C Effect, and any similar effects which may exist. So, for convenience, I will call them ‘Reversal Effects’.

    A negative feedback reduces the magnitude of an effect.

    A governor limits the magnitude of an effect.

    A Reversal Effect arises in response to a direct effect, and it combines with the direct effect such that the combination has opposite sign to the direct effect (i.e. when the direct effect is +ve the combination is –ve).

    Richard

  183. I think your “reversal effect” is so vague as to be unhelpful. “Governor” in the sense you define necessarily uses a negative feedback.

    A “governor” will not maintain the degree.day integral as appears to happen in the tropics and so is also inadequate.

    Someone in another of Willis’ threads suggested this was closer to an industrial PID controller. I think that description is more suitable.

    In any case there is a need for precise well defined terms here which is why I favour feedback descriptions. There is a whole branch of engineering that knows how these work and describe things in precise mathematical terms.

  184. milodonharlani, never mind. I found your numbers here http://usda01.library.cornell.edu/usda/ers/WHS//1990s/1994/WHS-11-15-1994.pdf

    In appendix table 1 it shows yield decreasing from 39.5 in 1990/91 to 34.3 in 1991/92. But it also shows an increase from 32.7 in 1989/90 to 39.5 in 1990/91. If the drop was caused by the volcano, what caused the rise, an anti-volcano?

    Clearly you have cherry-picked that fluctuation. A primary cause of the fluctuations is weather, specifically drought or too much rain. It is obvious from the data that those factors vary without volcanoes. The variation you attribute to the volcano is well within the range of variation for previous and subsequent years.

  185. richardscourtney says:

    September 23, 2013 at 1:47 am
    ///////////////////////
    Richard

    Thank you for your reply which was a response to a request raised by me, not Stephen. i was aware that the point that I raised related to the R&C paper, not to your own research, but you are one of the commentators whose comments I always seek to read and consider (because of your informed research led views) and hence the reason why I addressed my point to you.

    I accept the points that you raise. I would add to those points ocean overturning and ocean currents to the mix. These oceanic currents effectively take away from the tropical ocean much heat that is/would be generated by the solar irradiance received by the tropical oceans as the heat is being created and effectively distributing this ‘excess’ ‘energy’/’heat’ to other regions of the globe, particularly polewards. I suspect that but for these currents, one would see higher temperatures in many parts of the tropical oceans.

    Personally, I consider that we have a slightly biased view of ocean temperatures due to the sampling by ARGO buoys (and I am sceptical as to whether the free floating nature adds to the bias since these buoys float with currents/density profiles which in themselves are heat dependent/correlated) . If I recall Willis’ post correctly (appologies to Willis if I have got this wrong through being too lazy to check it out), he was arguing that for practical purposes the tropical ocean temperature is capped at 30degC. I recall pointing out to him that the process he described did not cap ocean temperature at that figure and that was clear from his own data which showed some ARGO buoys reporting 32degC. My recollection was that Willis took the view that the number of ARGO buoys reporting 32degC was very small (which is so) and even took a similar approach with the buoys reporting 31degC.

    I accept that relatively few ARGO buoys report temperatures of 32degC, and whilst Willis and I may disagree as to what consitutes a ‘few’ for the purposes of considering the number that report 31degC, the small numbers are a consequence of the chosen distribution of ARGO, and the fact that this chosen distribution excludes ARGO from sampling some of the warmest oceans on the planet..

    By way of aside, in my view, the most important element of understanding the climate and how it is driven is the full and proper understanding of the oceans, their temperature profiles, oceanic currents, the interaction of the oceans with the atmosphere immediately above the oceans, the manner in which ocean heat content is distributed around the globe including how this influences the jet streams etc. In this I would include the detailed interaction of DWLWIR in the immediate atmosphere above the oceans (which is not only high in water vapour content, but also contains water droplets/a fine mist of windswept spray and spume) and not simply its absorption in the first millimetre (bearing in mind that about 60% of all LWIR is fully absorbed in just a few microns and probably less than 10% makes it way to past 10 microns – given that DWLWIR is omni-directional even the absorbtion could be even greater in the first few microns). i guess that one should not overlook the fact that it is only oceanic temperature measurements which inform upon the energy budget given that land based temperature data sets are not measuring energy and therefore can inform little as to whether there is any ongoing change to the energy budget of planet Earth. Planet Earth is a water world, and understanding the significance of that is, in my opinion, the key to understanding the planet’s climate and how it is driven. personally,I do not consider that enough emphasis has been placed on this, but this may change given the ‘warmists’ mantra that energy is being sequested and hiding in the deep oceans. This claim will inevitably lead to greater investigation and scrutiny of the oceans and the role they play in influencing climate.

  186. Greg Goodman: “It is the key word non-linear that makes the overshoot required to preserve the degree.day integral. A linear neg. f/b will not do this.”

    While I am almost certain that non-linear effects are the key to the climate’s being limited to a relatively narrow temperature range, it is not true that only non-linear systems exhibit overshoot. Any “two-box” system will overshoot, and some will in fact oscillate.

    Example: a system whose response y to a stimulus x is given by

    d^2 y / dt^2 + 2 dy/dt + 40 y = 40 x

    will exhibit a (decaying) oscillatory response to a step in stimulus.

  187. Stephen Wilde says:

    September 23, 2013 at 12:51 am
    ////////////////////
    I accept the point that surface pressure is relevant to the temperature at which bonds are broken. The world would be a different place if atmospheric pressure was different (some consider that it was different at the time of the dinosaurs and there is much research into this and in particlur with respect to flying dinosaurs as well as long necked varieties).

    I am not sure what to make of your comment “Having considered the matter further overnight I still don’t see how emergent cloudiness can provide any sort of cap on achievable temperatures because a cap has to be exceeded before the clouds form (at least the types of clouds proposed).” Don’t clouds form over all oceans even those with cold temperatures, no doubt because evaopration begins to take place as soon water is liquid, albeit the rate of evaporation is proportional to its temperature such that the rate of evaporation and hence the volume of evaporated water is greater over the tropical oceans than over high latitude oceans. Doesn’t any cloud block solar irradiance such that cloud formation is a negative feedback, although I accept that the type of cloud (not only its areal extent, but also its volume, water content, vapour & water droplet size, height, time of formation and time of dissipation) influences the extent of that negative feedback. Clouds are infinitely complex (and chaotic in nature notwithstanding that they may have certain key drivers) and without understanding these there is no chance of modelling climate.

  188. Frank says:
    September 23, 2013 at 1:53 am
    >>>>>>>

    Great input, including a little math to help our understanding. I was trying to put it all together last night, but was too tired and gave up, so was pleased to find your helpful comment this morning.

    Thanks!

  189. Greg Goodman:

    At September 23, 2013 at 3:36 am you reply to my having said

    A Reversal Effect arises in response to a direct effect, and it combines with the direct effect such that the combination has opposite sign to the direct effect (i.e. when the direct effect is +ve the combination is –ve).

    by saying

    I think your “reversal effect” is so vague as to be unhelpful.
    {snip}
    In any case there is a need for precise well defined terms here which is why I favour feedback descriptions. There is a whole branch of engineering that knows how these work and describe things in precise mathematical terms.

    The definition I provided is precise and not “vague”. I can state it in mathematical terminology if required but see no reason to do that here.

    A negative feedback reduces the magnitude of an effect.
    A positive feedback increases the magnitude of an effect.
    A governor limits the magnitude of an effect.

    No combination of feedbacks and governors reverses the sign of an effect.

    Richard

  190. what happens at glaciation and deglaciation is clearly different from what happens in between. There is apparently two stable states ( attractors ) for the climate system. A positive feedback seems to make it snap form one state to the other. We don’t really know what triggers the change-over.

    At least two attractors. More likely, the system is highly multistable with attractors all over the place and with at least two MAJOR “attractors of attractors” as it were.

    There are even multiple distinct ways the climate system can shift. As you say, when the climate is in a major-bistable critical regime, chance fluctuations can kick the system too far from the currently dominant (say) warm-phase attractor and the system can then descend — probably via a series of transitions to transiently stable intermediate attractors — to one of the many cold-phase attractors sufficiently stable to hold the system once again, or it can start to descend — as perhaps it did during the LIA — but then can pop back up. Bobbles of this sort are clearly visible in the geological record, where even during glacial eras there are stretches of warming that doesn’t reach the critical/tipping point followed by aggressive cooling, or the Younger Dryas, where it warmed to interglacial temperatures but then “suddenly” re-descended into glaciation for close to a thousand years before re-warming into the Holocene proper.

    In addition to jumping attractors, the attractors themselves appear capable of secular moment on longer timescales — the stable point itself is no doubt a weak function of a variety of forcings plus a non-Markovian integral over the climate history into the past (of the sort Willis is exploring with his lagged response models above, where it isn’t the state of things “right now” that always matters, sometimes it is the state of things a year, ten years, fifty years past PLUS the state of things right now. The ocean has mixing/turnover phenomena with timescales of centuries on up (in addition to shorter time scales as well).

    The pattern of temperature shifts over the “reliable” temperature era (the last 33-50 years) has been periods of relative stability order of 15 years followed by a rapid shift over 2-3 years followed by relative stability. The less reliable thermometric era (HADCRUT-whatever or GISS-whatever) also suggests a pattern of stable temperatures for periods of 1-3 decades, followed by 1-3 decades of warming, a punctuated series of equilibria, with the high frequency noise Willis plots above reflecting motion AROUND the current attractor, not the motion OF the attractors or the jumps between attractors (where the latter two would be very difficult to differentiate without a knowledge of the dimensionality of the space and some feel for the nonlinear functions that describe the locally stable points to gain some insight about how they might vary, appear, disappear as underlying parameters in the climate system change).

    However, the fundamental problem with doing the analysis Willis presents so ably above is that it neglects the probable errors. HADCRUT4 actually has an estimated error bar of 0.15 C for present day data. It is at least 0.5 C, if not larger, for most of the rest of the thermometric era, in particular for those parts back in the 19th century and early 20th century. Remember that entire continents were still mostly terra incognita (as far as systematic sampling with reliable thermometers is concerned) well into the 20th century.

    The inclusion of probable error into the fits makes it even more difficult to discern the effect of volcanoes or any other secular causes in the temperature trend. If one is trying to resolve a 0.1 C effect in data that is both (possibly) trended (with high autocorrelation) and noisy at 0.5 C, you simply cannot expect to obtain a reliable causal/correlative decomposition. If you like, the p-values Willis obtains are too optimistic — it is even more likely that the almost completely invisible trend he uncovers within the data is there by random chance, because the detrended noise he fits is, in fact, uncertain within a range that is slightly larger than the range of the noise itself.

    With that said, I do like looking at fluctuation-dissipation in climate models as I think it has a lot to teach us. The response of the climate to a sudden forcing (like a volcano, like a powerful ENSO) in principle gives us a direct look at the shape of the attractor(s) that govern the climate’s current set point.

    rgb

  191. richardscourtney: “A Reversal Effect arises in response to a direct effect, and it combines with the direct effect such that the combination has opposite sign to the direct effect (i.e. when the direct effect is +ve the combination is –ve).”

    Not sure I follow that, but might the following be a quotidian example?

    A water drop hitting a skillet as the skillet only starts to heat up evaporates slowly. As the skillet heats up, subsequent drops evaporate more quickly–but only up to a point. After that, there’s a regime in which increasing skillet temperature causes the drops to evaporate more slowly. (They form little water marbles that roll around.)

  192. richard verney:

    Thankyou for your reply to me at September 23, 2013 at 5:30 am.

    I apologise to you and anybody else whom I failed to specifically name but ‘lumped as one’ in my reply to requests for explanation of the R&C Effect. This was an error but was not intended as a slight or as any other offence to anyone.

    I hope my explanation was adequate, and I read your resulting comment with interest.

    Richard

  193. What eruptions are they attributing the current non-warming to? Pinitubo and Chichon? Ridiculous. It’s long been established the effect is short term (1-2 years).

  194. Joe Born:

    re your post at September 23, 2013 at 6:30 am

    Thankyou for that example. I did not know of it. However, I do not think that is an example of what I am calling a Reversal Effect.

    In your example the formation of ‘marbles’ acts to reduce evapouration rate so acts as a negative feedback on evapouration.

    [If the formation of ‘marbles’ acted to stop evapouration and to induce condensation then it would have a Reversal Effect on evapouration.]

    The R&C and Eschenbach Effects are Reversal Effects on temperature because they induce the system to COOL (n.b. not warm) in response to increased heat input to the system.

    Richard

  195. Ouch! I stupidly wrote

    If he formation of ‘marbles’ acted to stop evapouration and to induce condensation then it would have a Reversal Effect on condensation.

    I intended to write

    If the formation of ‘marbles’ acted to stop evapouration and to induce condensation then it would have a Reversal Effect on evapouration.

    Sorry. Richard

  196. Milodon, I find that I owe you an apology … but likely not for what you think.

    I was reflecting on my statement in the final comment of my post “Missing the Missing Summer.” I had said:

    That was the point I was trying to make above, that if the weather really had been all that bad in 1992 the crop yield would have reflected it, and it didn’t. Not for any type of produce, not for tubers, not for legumes, not for vegetables, not for fruits, not for grains.

    Now, I’ve grown very cautious about what I say, and I couldn’t make sense of that. More to the point, I hadn’t been making any points about Pinatubo above, only about Krakatoa. What I realized was that in my comment I’d written 1992 (thinking wrongly of Pinatubo), and I should have said 1815, the date of Krakatoa.

    That makes sense, because in “Missing the Missing Summer” I’d posted the following graphic above my comment:

    That graphic was the reason that I made such an absolute claim, which I very rarely do.

    So my apologies, Milodon for leading you astray with my incorrect date. I trust that my claim makes more sense now.

    All the best,

    w.

  197. Willis Eschenbach says:
    September 22, 2013 at 6:15 pm
    Jim G says:
    September 22, 2013 at 5:10 pm

    A true super eruption of a super volcano might be at odds with your “self regulating” surface temperature hypothesis.

    “We’ve had supervolcanoes in the past, and the temperature has always recovered. Under the models’ view, that wouldn’t happen … with my hypothesis, it would. ”

    w.

    Willis,

    I was not implying that the models were correct. Being better than the models is damning yourself with faint praise. If one takes a long enough view of climate/temperature even the Milankovich cycles do cycle back.

  198. richardscourtney says: The R&C and Eschenbach Effects are Reversal Effects on temperature because they induce the system to COOL (n.b. not warm) in response to increased heat input to the system.

    What are you trying to suggest here? Is your “reversal” the overshoot that Willis refers to which is part of a reaction that brings the system back to it’s previous state or are you suggesting a reaction the leaves the system in a settled state, cooler than it was before the perturbation. If that is the case I want to see proof. Not just inventing a name for it.

    If it’s the former, it already has a name: non-linear negative feedback.

  199. ***
    Steven Mosher says:
    September 22, 2013 at 6:06 pm

    The ECS is what we are mostly interested in, the full response after all feedbacks,
    ***

    All forcing aren’t created equal. IR back-radiation from GHGs can’t penetrate beyond water surfaces, so the effects are immediate (changes in ocean surface temp and atmospheric latent heat from evaporation). Only short-wave solar forcing can be stored in the oceans and have a significant time-lag.

  200. Quick question:

    How long do this particles stay in the stratosphere, and do their affects change as they fall through the troposphere?

  201. RC Saumarez says:
    September 23, 2013 at 12:55 am

    Dear Willis,

    Yo will recognise that I have been a critic of your signal processing methods. having gained a PhD in a world class DSP lab and used it for 30 years (on and off). If you want some help I am happy to advise you.

    Richard Saumarez

    Yeah, and you’re also the guy who claimed that two points determine a sine wave, if the time between them is less than half the shortest period in the dataset … and you’re the guy who claimed he can discern a 170.7 year cycle when given only 110 years of data.

    Since I demonstrated that the first one of those claims was 100% wrong … and since the second claim is ludicrous on the face of it … well, Richard, I’ve kinda lost trust in your expertise.

    Finally, you wrote a whole post that you later claimed was to show me where my math was wrong … but you never mentioned me or any of my claims anywhere in the post. Didn’t stop you from boasting about how you schooled me, despite doing nothing of the sort.

    So I believe I’ll pass …

    w.

  202. Stephen Wilde says:
    September 23, 2013 at 1:13 am

    Having looked at Willis’s original Thermostat Hypothesis again I see that he includes vapour pressure as a relevant variable but not absolute pressure.

    I think that is a significant omission … [lots of the standard idiotic pressurehead drivel snipped]

    That is an elephant in this particular room.

    Stephen, YOU are the elephant in the room. Despite being asked twice to take your ignorant claims that pressure warms the earth to some other thread, here you are peddling the same bullshit and stomping your big dumb elephant feet all over a discussion which has NOTHING TO DO WITH YOUR CLAIMS. This thread is about volcanoes, not your stupid ideas.

    What is it about PLEASE DISCUSS YOUR CLAIMS ELSEWHERE that you don’t understand? I’ve shown in “A Matter Of Some Gravity” that your theories are in contradiction to the Second Law. If you don’t understand my proof, and why it shows that your ideas won’t work, then read it again. Repeat as necessary until you finally get the picture—neither gravity nor pressure can provide a constant source of heat to warm the surface as you claim, because it violates the Second Law.

    And whether you agree with me or not, and whether you understand my proof or not … please don’t come back to argue that you are right. Not interested, sorry.

    Don’t go away mad.

    Just go away.

    w.

  203. @Tintoolman.
    Here some points:
    1) The model posed by Eschenbach is a simple 1st order ARMA model. It is well known that the autocorrelation fundtion of temperature does not correspnd to this model. Therefore it is incorrect. This has been widely discussed. by McIntyre, Luck and Ludeke and others in the past. See for example:

    http://judithcurry.com/2012/02/19/autocorrelation-and-trends/

    2) There are significant non-linearities in the climate system and characterisation is made over change in climate. The degree of linearisation possible is unknown and therefore it is unknown whether linear models can be applied to the climate system as has been done here( I suspect not). I would comment that establishment of linearity or non linearity, although extremely difficult using this data is a first step. A non-linear Hurst dynamic model is practically impossible to distinguish from a linear multicompartment model in the global temperature signal. (See link above for details).

    3) The errors involved in fitting parameters to a climate model are enormous, even if the data were perfect. If one is going to propose a model, rigorous assessment of that uncertainty is essential. Furthermore,the functions being fitted are highly ill-conditioned, meaning that small changes in the input data will lead to large changes in fitted parameters.

    As I have said in past, I am happy to write a criticism of the methods employed here, particularly how one can be misled by spurious linearisation of non-linear systems. Since i was challenged to put up or shut up, I have written an essay on the difficulties in establishing system responses in non-linear systems that might appear to be linear at first sight, sent it to WUTW but it is has clearly not been thought to be suitable for posting, It is difficult to take the argument further

    I have said that I am happy to help Mr Eschenbach if he wishes to develop a more elaborate model since this is a constructive approach.

  204. MikeN says:
    Is the IPCC trying to blame the pause in global warming on a handful of small volcanoes?

    Any straw in storm … ;)

  205. In the above post to get pictures and graphs just google The real climate drivers -ocean and solar cycles amplified by levels of volcanism.

    Willis study is flawed in many ways, one way is he is trying to isolate the effects of volcanos in the climatic system by putting them in isolation against all the zillion other climatic parameters trying to obtain a climate direct cause and effect due to the volcanic eruption itself.Does not work.
    The IPCC is also flawed because the Aerosol Optical thickness for the N.H. ws only .01 tau during the time the IPCC claimed volcanic activity and or aerosols was having an impact of slowing down the warming.

    Weatherbell Inc. led by Joe Bastardi and Joe D’Aleo see it differently then what Willis is trying to convey from his study. I am in their camp 100%.

  206. Stratospheric aerosols cut back on incoming solar radiation. Values of 7% reduction or higher have been measured. This is widely accepted as being a factor in causing global cooling for 2 to 3 years after a major eruption. Some eruptions have been estimated to throw aerosols to 80,000 or even 120,000 feet high into our atmosphere. While ash and aerosols (sulfate converted from O2) typically falls out from low level eruptions in days or weeks, those that make it into the stratosphere have a lifetime of a few years. While there, they serve to cool the atmosphere and surface (although with regional variances) and when they fall out to serve as nuclei for ice and water droplet clouds and precipitation (rain and snow). Remember the big Midwest flood of 1993 and the huge snows of March 1993 to the winter of 1993/94 and 1995/96

    Dr. Richard Keen, who lives in the beautiful Colorado Rockies and is both a weather observer and astronomical expert has been using eclipses to measure this effect. He found the thickness varied and estimated the effect on temperatures

    The above is part of the article by Joe D’Aleo of Weatherbell. Again to get the whole article google The real climate drivers -ocean and solar cycles amplified by levels of volcanism.

    Many graphs will be found.

  207. I have just seen Eschenbach’s charming post.

    I have never made the claims that you say i have because they are total nonsense. Having been taught this subject formally, I hope that I do not make those sort of mistakes.

    I was asked to write a post because you wrote a post on filtering and statistics that was completely wrong. I did not go out of my way to be offensive in that post, which was written at a very elementary level,, but it was completely apparant from what I wrote that filtering, as you had done in the bandwidth of a signal, will alter the correlation functions.

    I would point out that these sort of mistakes stem from a failure to understand the theory of the subject rather than an ability to do something in “R”.

    You ask me what is wrong with your model? For a start, you apply a first order ARMA process to signal when there is subtantial evidence that the temperature signal does not have an autocorrelation function that conforms to that process, Therefore you cannot apply this model.

    Second, you have failed to consider any possibilityof non-linear effects in your model.

    Third, you have not performed any sort of proper analysis of uncertainty in your analysis, which you should, The errors in the type of an analysis you have performed are potentially large. Your claim that it accounts for 90% of the energy in the signal (I presume that you really mean this? Are you familiar with Parseval’s theorem?) is inadequate to clain that that you have identified the system in question.

    Finally you ask for help from some MSc students. If they are properly taught, I have no doubt that they form an opinion on your mathematical expertise.

  208. richardscourtney: “Thankyou for that example. I did not know of it.”

    You’re welcome. Although I encountered it in the kitchen, personnel of a utility-boiler manufacturer I once had as a client referred to the phenomenon as “departure from nucleate boiling” and were wont to discuss “departure-from-nucleate-boiling ratio.”

    “In your example the formation of ‘marbles’ acts to reduce evapouration rate so acts as a negative feedback on evapouration.” As far as “feedback” goes, I guess my taxonomy is different from yours, but I doubt that either of us would gain much from further pursuing that discussion.

  209. Willis or somebody else with good science-fu… can somebody help the liberal arts major here?

    It seems to me (but sometimes my ideas accidentally violate the laws of thermodynamics), that part of the equilibrium might have something to do with axial tilt? No, I’m not trolling, hear me out for a second, and then illuminate/slap-me-around as needed:

    You get something like a volcano: it’s going to produce a lot of localized cooling. But then it becomes winter, and the local area gets dropped even further below what the volcano would be doing…end result being more heat circulation from the rest of the globe, bang, it’s evened out, resulting in, as measured globally the forcing mechanism is destroyed at a global energy cost that is distinct, but miniscule.

    I’m assuming the 305k ocean-temp limit combined with lots and lots of summer sunlight handles the other side of said equation, but… I was a history major. You wanna know why Hugh de Beaux kidnapped Mary of Siciliy in 1350, I’m your dude…. this stuff, not so much (but interested).

    What am I not getting here? Thanks in advance.

  210. Bill says: September 22, 2013 at 1:28 pm
    Willis, how does your governor theory explain ice ages?
    _____________________________________

    I know this is slightly OT for this thread, but it is an interesting and related topic.

    It is said that Milankovitch Cycles can vary the insolation at latitudes above 60N by as much as 20%. That is quite possibly enough change in insolation to generate an Ice Age, as you sure would not get an ice sheet to grow over London with our current insolation levels. So something must be stopping the insolation.

    Here is a common Milankovitch Cycle vs Temperature graph. Plus a graph of Ice Ages. I am not sure whether the match between calculated insolation and the temperature is significant mathematically, but the layman’s ‘one-eyed squint’ method says the correlation has some merit.

    But if Milankovitch Cycles were responsible for N hemisphere Ice Ages, then I would presume that the southern hemisphere must have equal and opposite glaciation periods. Is this so? I was unable to find any info on this.

    So in answer to Bill’s question:
    The Tropical Cumulus theory will probably work well as a regulator for the temperature of the system over long periods of time. But if someone turns the Sun off in the N hemisphere (a negative Milankovitch Cycle), its going to get cold.

  211. Willis, excellent pose with good scientific analysis unlike AR 5

    “…period 1951–2012, is due in roughly equal measure to a cooling contribution from internal variability and a reduced 2 trend in radiative forcing (medium confidence). The reduced trend in radiative forcing is primarily due 3 to volcanic eruptions and the downward phase of the current solar cycle.”
    Strange the same suspects denied these factors before when there was an increase in temperature and now admit to these when temperatures are flat or downward.
    Furthermore these are not new considerations (they have been mentioned before by skeptics and ignored) and if the computer models did not consider them it further confirms the flaws in the models. Let’s face it, the models are a total fabrication using preconceived, erroneous notions and bear no resemblance to science or reality.

  212. eric1skeptic says:
    September 23, 2013 at 2:27 am

    I posted the link in my first comment on the topic, & subsequently, as would have been easy to check.

  213. Greg Goodman:

    I am replying to your post at September 23, 2013 at 7:19 am which says in total

    richardscourtney says:

    The R&C and Eschenbach Effects are Reversal Effects on temperature because they induce the system to COOL (n.b. not warm) in response to increased heat input to the system.

    What are you trying to suggest here? Is your “reversal” the overshoot that Willis refers to which is part of a reaction that brings the system back to it’s previous state or are you suggesting a reaction the leaves the system in a settled state, cooler than it was before the perturbation. If that is the case I want to see proof. Not just inventing a name for it.

    If it’s the former, it already has a name: non-linear negative feedback.

    NO!
    I do not know how I could be more clear than I have been in my series of attempts to explain this matter in this thread.

    A negative feedback – be it linear or otherwise – reduces the magnitude of an effect.
    So, for simplistic illustration, if a heat input causes warming of 2.0K per hour then a negative feedback of 0.5 will reduce that rate of rise to 1.0K per hour.

    A Reversal Effect is induced when the system passes a threshold. It is larger and has opposite result to the original effect. So, when the Reversal Effect initiates the combination of the original effect and the reversal Effect provides the opposite of the original effect.

    In the simplistic illustration, the heat input causes warming of 2.0K per hour but when the Reversal Effect initiates the system starts to COOL although the heat input is continued. This has nothing to do with overshoot because it has nothing to do with feedbacks on the original effect .
    It happens because an ADDITIONAL effect initiates in the system.

    Thunder clouds and storms are the additional initiated effect of the Eschenbach Effect.
    Cirrus clouds are the additional initiated effect of the R&C Effect.

    In other words, the system becomes a different system when a Reversal Effect initiates.

    Richard

  214. RC Saumarez:

    In your less than “charming” post at September 23, 2013 at 8:24 am you say

    I hope that I do not make those sort of mistakes.

    I can accept that because we all have similar hopes. I hope to win the lottery.

    Richard

  215. Willis is trying to show ever so hard that volcanic forcing is much less, then all the past evidence shows to the contrary. He is entitled to his opinion and methold of trying to prove it, although it is not convincing to me.
    Look at what happened to the temperatures following the Mt. Pinatubo eruption, they went down in the face of an El Nino.

    Each volcanic eruption is different in regards to it’s location, composition of what it ejects into the air, height/amounts of material it ejects into the air making a volcano /temperature correlation amost impossible to obtain, not to mention other climatic items acting in concert or against the volcanic climatic effects.

    Willis says and it it correct in a sense( but not really) the following: a host of emergent thermostatic phenomena act quickly to cool the planet when it is to warm, and to warm it when it is to cool.

    If true 100% of the time as you convey more or less; reconcile that statement with the fact the earth has had many ABRUPT climatic changes in the past, and have shifted from a glacial state to an inter glacial state many times in the past. Apparently that statement does not hold up under all circumstances for if it did the climate would never have abrupt climate swings in temperature both up and down, or vary from a glacial to an inter glacial state..

    Something else has to be at work here.

    The three dryas periods(oldest,old and younger) and the 8200 years ago abrupt cooling period are some examples of rapid climatic temperature drops, which run counter to your statement.

  216. Is your analysis in conflict with the following study?
    Global surface-temperature responses to major volcanic eruptions
    SEAR, KELLY, JONES & GOODESS
    Nature 330, 365 – 367 (26 November 1987)
    …In a previous study5 it was shown that significant surface cooling occurs over the landmasses of the Northern Hemisphere in the first few months after a major eruption in that hemisphere. Here we extend that work using new surface-air temperature compilations based on land and marine data6 for both the Northern and Southern Hemispheres. Our results indicate that major Northern Hemisphere eruptions have an immediate effect on the Northern Hemisphere average surface temperature but little or no effect on the Southern Hemisphere average. Southern Hemisphere eruptions affect both Southern and Northern Hemisphere temperatures after a lag of between six months and a year.

  217. http://books.google.com/books?id=3me16ll_TRkC&pg=PP152&lpg=PP152&dq=39.5+34.3+bushels+acre&source=bl&ots=kl-4vVGj5c&sig=87fK59G59caccvIgP5J6GlrtRUA&hl=en&sa=X&ei=f2RAUsL5HYiwigKN8oGIDQ&ved=0CE4Q6AEwBQ#v=onepage&q=39.5%2034.3%20bushels%20acre&f=false

    Checking I see that it was 39.5 bu/A in 1990 & 34.3 in 1991, then back up in ’92, so I was off by a year. My mistake; yet price did go way up in ’92. USDA marketing year differs from international, but the error appears to have been mine.

    However the fact remains that globally yield fell in 1992, as shown above, from 2.6 T/ha in ’91 to 2.5 T/ha. It’s noticeable, to me at least, even in Willis’ graph. I believe it went up for corn, but didn’t check other crops.

  218. WIllis if volcanos made no contributions to the Maunder Minimum and Dalton Minimum ,you are then adding more evidence that is was caused by direct changes in various solar parameters rather then perhaps a secondary factor associated with the direct solar changes in solar parameters.

    In this case volcanism being an associated secondary factor asociated with prolonged solar minimum periods,(which many studies show to be the case) which enhances any change in the climate due to direct solar vairiations, which you also do not feel is real ,despite once again past history showing very clearly that is not the case.

    Willis something has caused the climate to change in abrupt ways and from glacial to inter glacial states. I say, it is changes in the sun and all the associated secondary effects which phase into either a warming /or cooling of the climate, if not you tell me what it might be?

    I will be most interested in any explanations you may have. Seriously.

  219. PIPPEN I agree 100%. Evidence is so very clear,look at the Mt. Pinatubo eruption(very recent evidence) and what temperatures did for a short time following that eruption.Look at the graph Dr. Spencer has, which clearly illustrates this fact. Google Dr. Spencer to go to this web-site.

    Yes the effects are short lived but they are real and again each and every volcanic eruption is different, and goes off under different climatic scenarios..

  220. I would like to ask Willis Eschenbach a direct question.

    Have you formed the auto-correlationfunction of your model? If so, does it conform to the calculated ACF from the data? I think you may find determination of the latter quite difficult and not immediately accessible in “R”. I would be extremely careful about applying a filter if I were you as this can give rise to wildely incorrect results in calculating correlation functions.

    You will find that that your model does not does not conform to the correlation structure of the temperature signal – this is well known.

    In this case your model is wrong, either because the structure of the system is linear but more complex than the single thermal capacitance you assume, or it is a non-linear process with possibly, as has been speculated, Hurst process dynamics.

    I would be interested to know how you resolve this problem.

  221. ” ALL-CLEAR IN THE STRATOSPHERE”

    http://spaceweather.com/archive.php?view=1&day=19&month=12&year=2010

    Earth’s stratosphere is as clear as it’s been in more than 50 years. University of Colorado climate scientist Richard Keen knows this because he’s been watching lunar eclipses. “Since 1996, lunar eclipses have been bright, which means the stratosphere is relatively clear of volcanic aerosols. This is the longest period with a clear stratosphere since before 1960.”

    “The lunar eclipse record indicates a clear stratosphere over the past decade, and that this has contributed about 0.2 degrees to recent warming.”

  222. I wonder about the IR blocking effects of a major equatorial volcanic eruption with enough ash thrown into the upper atmospheric layers to prevent rain-out AND that occurs in equatorial regions.

    What would be the effects during El Nino, neutral, and La Nina events if all this ash was in the upper atmosphere circling the equatorial belt? Recharging events (neutral to La Nina) serve to keep our oceans warm, thus keeping US warm when El Nino’s cough up heat onto land. What if ash prevented that recharge from occurring to the degree it would normally occur under clear sky conditions?

    Given the above, I am wondering if explosive power, timing and location are important factors in whether or not volcanic eruptions can have long term affects that go beyond the rain-out period immediately following explosions.

  223. Steven Mosher September 22, 2013 at 4:19 pm

    [..] sadly you cant calculate sensitivity […] (from)…

    Which bit of THERE IS NO SUCH THING AS SOME Y=(F)x ‘climate sensitivity’ calculation are you not following?

    Reading through the thread we have everything from ‘my new climate model suggests’, ‘the price of Wheat would indicate’ and ‘the year without a Summer’ (Scotland only?) all held up as evidence that Willis is wrong. All of it it BS.

    Simple – tending toward a ‘Snowball Earth’ – his ‘mechanism’ can’t help. No clouds/thunderstorms ever form and so that control mechanism has reached its lower limit.

    Working the other way (eg Solar output increases noticeably) the ‘emergent thermostatic phenomena’ can spread north and south and increase in intensity and I have no idea where the upper limit lies, many thousands of W/m2, I suspect.

    Bottom line (for those worried about ‘anomalies’) – grab a base temp (15C or whatever you chose) and superimpose those anomalies. Tell me then just how much you would expect to be paid for designing a Planet and a control system that, with no unexpected changes, could hold surface temps to +/- tenths of a degree over significant periods. I would be a very rich man.

    Something also occurred regarding ‘Sun reflecting particles high up in the atmosphere’ – more (badly named) ‘greenhouse effect’ – once reflected from the surface – does that ‘layer’ not also prevent ‘visible light’ from leaving? 15um or 500nm shouldn’t they both bounce back and forth increasing their time at the surface with suitable interference?

    Anyway Willis … welcome back. Hope you enjoyed our historical landmass.

    Having read some of the responses to this post – bet you wish you could have extended your holiday though. Some people just don’t read. You are just too polite sometimes

  224. Willis: “Since I’ve been a heretic for a decade or so, your attempt to be all pally by welcoming me on board is an insult … likely unintended, but that just makes it worse.”

    Willis, easy, I sure didn’t mean that derogatory, yes, unintended, didn’t realize you held such a viewpoint of yourself for so long, thought you were making a movement to total disbelief. That’s all. If it’s been a decade, fine.

    I, as others, do greatly appreciate your work and insight into these areas though I hope you have the flexibility to possibly be incorrect in a few of your viewpoints where we do differ, one of those being that you do not feel the mass of atmospheres matter at all, I do. Keep an open mind, stay a real scientist. I’ll always give you that same courtesy. Even here on this post.

    Now after reading the post and comments, give a close look at what Frank commented, his stood out, it adds a thick layer of support for what you have found on volcanoes even though I think his comment of 1.4°C sensitivity is not correct. He is correct on the problem of using power instead of energy, mass, and time. Hope you give me the leeway to disagree with you every now and then without you always taking such offense.

  225. Thomas says: September 22, 2013 at 10:30 am
    Maybe Eschenbach has written about it before, but I’m a bit confused on how he can reconcile “I hold that changes in forcing only marginally and briefly affect the temperature. Instead, I say that a host of emergent thermostatic phenomena act quickly to cool the planet when it is too warm, and to warm it when it is too cool” with the existence of ice age cycles. Whatever thermostat the Earth has doesn’t seem all that good.

    Willis Eschenbach says: September 22, 2013 at 1:10 pm
    The planet’s temperature varied by ± 0.3°C over the last century. This is a regulation to within about ± 0.1% … on a free-running system which is regulated by nothing more substantial than wind and water. …. If you know anything about heat engines, you’ll agree that that is a fantastic governor … w.

    What manner of observational methods, data collection methods, experimental methods, and analytical methods would be necessary either to demonstrate or to disprove Willis’ theory that the earth’s climate system operates mostly as a self-regulating, self-stabilizing heat engine?

  226. Greg and Geran: Thanks for the replies to my comment. I usually feel like most readers will cheer anything opposing the consensus (no matter how flawed) and will ignore or jeer anyone that criticizes. Unfortunately, in this case Willis is trying the do the equivalent of understanding the relationship between acceleration and velocity by looking for a correlation between the two. The two phenomena are related (by integration or differentiation), but correlation isn’t the right method for understanding their relationship. One must integrate power over time and take heat capacity into account, before you can understand how much temperature change a forcing should produce. And since radiative forcing represents change in the power flux entering or leaving the earth, Willis is likely to remain confused as long as he continues to reject this concept. Recognizing the relevance of radiative forcing to temperature change doesn’t mean that temperature change can’t be suppressed by negative cloud feedback (or amplified by positive water vapor feedback). Accepting the concept of radiative forcing doesn’t mean one must believe in high climate sensitivity. Observations are most consistent with an ECS around 1.5 degC. (High climate sensitivity comes from climate models – which rely on 1-2 dozen parameters whose precise values aren’t known (in addition to physics). Change those parameters and you change climate sensitivity.

    If you want to know more about Pinatubo, see the discussion at the Blackboard. It’s excellent. http://rankexploits.com/musings/2012/pinatubo-climate-sensitivity-and-two-dogs-that-didnt-bark-in-the-night/

  227. Beta Blocker says:
    September 23, 2013 at 11:33 am
    ……. to demonstrate or to disprove Willis’ theory that the earth’s climate system operates mostly as a self-regulating, self-stabilizing heat engine?
    =======
    I’ve been thinking about that one…..long and hard….didn’t want to bring it up and derail this volcano thread
    Maybe Willis will do another thread on it….

  228. Latitude says:
    September 23, 2013 at 12:29 pm

    IMO that’s not a controversial assertion, if its brunt is to argue that earth is homeostatic most of the time, or all of the time, given long enough. Its temperature range for billions of years has been roughly from about 50 degrees below zero C during Snowball Earths to perhaps +40 degrees C during the melt-off phase of these global glacial interludes.

    But for the past 550 million years, that range has narrowed to around +7 degrees C during Icehouse phases & +23 degrees C during Hothouses. As we’re in an interglacial cycle in a glacial epoch, we’re presently 14-15 degrees C, depending upon whose special pleading seems more convincing to you.

  229. Depends on scale I would suggest. At fine scale micro climates, local address would come into play regarding self-adjusting thermostats. Which could be why we have fairly well-defined ranges in each of our climate zones, allowing thermostats free range inside those boundaries. On a larger longer scale (IE large areas of extreme heat, cold, etc over an extended period of time), something happens (IE rare oscillation convergence, equatorial event, etc) that fouls up the thermostats and we are impacted by powerful regional, hemispheric, and global forcings that overcome (for a while anyway) local thermostats resulting in extensive and longer term climate range excursions outside the norm. Again just pondering.

  230. Wayne: 1.4 degC was the most likely climate sensitivity derived by Paul_K from monthly observed changes in: albedo due to aerosols (radiative forcing), LWR emitted through the TOA, and surface temperature. He performed a best fit of these observations to climate sensitivity, the depth of the mixed layer, and heat diffusion below the mixed layer. Values of ECS below 0.9 degC or above 1.7 degC couldn’t be made to match observations by any changes in the size of the mixed layer and diffusion of heat below it.

    Interestingly, 1.4 degC is similar to Nic Lewis’s value of 1.6 degC (1.3-2.3), Otto, et al and other work based on recent observations. All of these number arise from calculations, but the require equations that properly describe all of the important heat flows. The observations also contain uncertainty. These numbers are results that depend on the quality of the analysis, not opinions.

  231. milodonharlani says:
    September 23, 2013 at 12:46 pm

    I should specify global mean T, to the extent such a figure can be measured or reconstructed.

    There appears to be a downside limit to runaway cooling & at least until the sun goes red giant, an upside as well. That may change as the sun becomes more radiant & as our planet loses surface water.

  232. “I hold that changes in forcing only marginally and briefly affect the temperature. Instead, I say that a host of emergent thermostatic phenomena act quickly to cool the planet when it is too warm, and to warm it when it is too cool.”

    A glance at any global temperature data, on any timescale, proves this idea wrong.

  233. Iben Browning did a lot of work on this before he died. He showed something like a Kondratieff Wave for tidal pressure on the earth and volcanic activity. Increased volcanic activity in the northern hemisphere when coupled with increased solar activity warming the equatorial region might appear to offset but results in the jet stream whipping back and forth. Even if the overall global temperature doesn’t drop, crops don’t do well with the instability.

  234. Pamela Gray says:
    September 23, 2013 at 12:55 pm
    Depends on scale I would suggest.
    ====
    Since we did it anyway……
    That’s what I think too Pamela….looks like the planets default setting is colder…a lot colder
    In my field we try to work with extremes…..because the extremes are very unstable…either the extreme of high or lowb
    All I see is that we’re in an extreme and it’s unstable…
    Not that the planet is regulating it back to some “normal”…..but the default setting is normal and there’s something else limiting that’s making it bounce around…sorta the opposite of the way Willis sees it

  235. Frank, I understand what you are saying, I have followed all of the information having to do with ‘climate sensitivity’, it’s just that I have come across some personal analysis that says via the Galileo probe into Jupiter and Venus’s atmosphere along with what we know of our atmosphere says the climate sensitivity for carbon dioxide is fleetingly tiny. Has to do with optical depth of the pertinent lines of the infrared absorbing gases at the planetary scale but this is not a good place to get into that, it’s off topic and would drag this right back into a property of matter called mass that Willis disagrees with its importance, but I do understand it is all just calculations of observations that may very well be just natural variability mistaken as unrelated signals.

    That is, better leave this for a future post.

    But I did perfectly agree with most of what you were saying on the lines of the physics involved having to do with this post, must take the masses and time into account.

  236. Upthread, someone asked me to show the lagged forcing versus the temperature. I said I’d leave it as an exercise, because I was feeling lousy. I still am, but I can’t let that stop me. Here is the requested graph

    The values are lambda (sensitivity) = 0.05 °C per W/m2 (0.18°C per doubling of CO2), and tau (the time constant of the lag) = 6.5 months.

    w.

  237. Frank says:
    September 23, 2013 at 1:53 am

    Willis: There is a flaw in your post: You are looking for a correlation between radiative forcing (W/m2 or POWER per unit AREA) and temperature, which is proportional to mean kinetic ENERGY.

    Frank, that’s because I’m using the canonical relationship of the current (and in my opinion incorrect) climate paradigm. This is that ∆T = lambda ∆F, where F is forcing, T is temperature, and lambda is climate sensitivity.

    As I tried to say up top, I don’t believe that’s how the world works … I’m just trying to show that by their terms the so-called “sensitivity” is very small.

    w.

  238. michael hammer says:
    September 23, 2013 at 2:33 am

    “I say that the current climate paradigm, that forcing determines temperature, is incorrect. I hold that changes in forcing only marginally and briefly affect the temperature. Instead, I say that a host of emergent thermostatic phenomena act
    quickly to cool the planet when it is too warm, and to warm it when it is too cool.”

    Willis why use such convoluted and hard to understand language – Why not just say “I believe there is strong negative feedback in the climate system” because that’s what your comment boils down to.

    Thanks, Michael, but actually I’m not talking about feedback. Instead, I’m talking about a governed system, which is a very different critter. See my post called “It’s Not About Feedback” for a discussion of the difference.

    w.

  239. RC Saumarez says:
    September 23, 2013 at 7:39 am

    @Tintoolman.
    Here some points:
    1) The model posed by Eschenbach is a simple 1st order ARMA model. It is well known that the autocorrelation fundtion of temperature does not correspnd to this model. Therefore it is incorrect. This has been widely discussed. by McIntyre, Luck and Ludeke and others in the past. See for example:

    http://judithcurry.com/2012/02/19/autocorrelation-and-trends/

    2) There are significant non-linearities in the climate system and characterisation is made over change in climate. The degree of linearisation possible is unknown and therefore it is unknown whether linear models can be applied to the climate system as has been done here( I suspect not). I would comment that establishment of linearity or non linearity, although extremely difficult using this data is a first step. A non-linear Hurst dynamic model is practically impossible to distinguish from a linear multicompartment model in the global temperature signal. (See link above for details).

    3) The errors involved in fitting parameters to a climate model are enormous, even if the data were perfect. If one is going to propose a model, rigorous assessment of that uncertainty is essential. Furthermore,the functions being fitted are highly ill-conditioned, meaning that small changes in the input data will lead to large changes in fitted parameters.

    As I have said in past, I am happy to write a criticism of the methods employed here, particularly how one can be misled by spurious linearisation of non-linear systems. Since i was challenged to put up or shut up, I have written an essay on the difficulties in establishing system responses in non-linear systems that might appear to be linear at first sight, sent it to WUTW but it is has clearly not been thought to be suitable for posting, It is difficult to take the argument further

    I have said that I am happy to help Mr Eschenbach if he wishes to develop a more elaborate model since this is a constructive approach.

    RC, I’m using the same method (lagging the canonical equation of ∆T = lamba ∆F) that I used to analyze the models. I got an R^2 of about 0.98 between my simple emulation and the models. And despite that actually quite amazing result, you think I need help?? … are you shooting for an R^2 of 1.5 or something?

    Next, as I said above, I’m using their claims and their methods to do the analysis, so I can compare apples to apples. I have specifically said I don’t think the actual temperature works that way … so you coming along to proudly announce that the actual temperature doesn’t work that way just shows that you haven’t understood what I said … but instead of saying “I don’t understand”, you want to tell us how brilliant you are, with your PhD from a leading LSD lab or whatever it was, and offer to help me. No thanks.

    Finally, you say that:

    Furthermore,the functions being fitted are highly ill-conditioned, meaning that small changes in the input data will lead to large changes in fitted parameters.

    Actually, that’s not true at all. Small changes in the input data hardly affect the results at all. How do I know that? Because unlike you, I actually did the exercise, instead of just making claims … so once again, you’re operating out of your own fantasies of what’s going on, instead of actually doing what I did to find out the truth of the matter.

    So you’ll excuse me if I find you pompous, insulting, and more to the point … wrong.

    w.

  240. Willis your arguments do not hold water, when it comes to volcanic eruptions and there impact or lack of impact on the climate. Read my earlier pos. In addition hundreds of studies do not agree with your study.

    I guess this is what makes climate what it is, everyone having a different take on things. Time will tell in all of this.

  241. MikeN says:
    September 23, 2013 at 7:57 am

    Is the IPCC trying to blame the pause in global warming on a handful of small volcanoes?

    No, they say it’s half “natural variations”, a quarter solar, and a quarter volcanoes.

    Here’s the bizarre part. In the last couple decades, using round numbers, the IPCC predicted about 0.4°C of warming … which hasn’t happened. So if a quarter of that (0.1°C) is volcanoes, and the recent volcanic forcing is (by their own numbers) about 0.1 W/m2, they’re saying that the climate sensitivity is 3.7° per doubling of CO2.

    Of course, if that were the case we’d have seen a drop of about 3°C from Pinatubo … and I fear that I don’t see that in the records.

    They just make these claims … but they don’t think them through to the end.

    w.

  242. Willis can not explain and will not explain why the earth has gone from glacial to inter glacial periods from time to time and why the climate has undergone several epsiodes of abrupt climate change in the past.

  243. Willis is correct to point out that the IPCC is trying to say volcanic action has caused the warming to slow down, and in reality it has had nothing to do with it. That is easily seen by looking at a chart of the aerosal optical thickness for the N.H from 2005-2011. The thickness is at very low values.

  244. Craig says:
    September 23, 2013 at 1:36 pm
    “Iben Browning did a lot of work on this before he died. ”

    I attended two of Iben Brownings lectures where in he hypothesized colder climate due to his predicted increase in volcanic activity which I understood to be based upon planetary alignments in his theory. This was late 1980’s I believe. Very interesting guy. He said we had the Soviets fooled in the nuclear showdown since our “big cities were false targets” as we would be better off without them.

  245. Beta Blocker says:
    September 23, 2013 at 11:33 am

    Thomas says: September 22, 2013 at 10:30 am
    Maybe Eschenbach has written about it before, but I’m a bit confused on how he can reconcile

    “I hold that changes in forcing only marginally and briefly affect the temperature. Instead, I say that a host of emergent thermostatic phenomena act quickly to cool the planet when it is too warm, and to warm it when it is too cool”

    with the existence of ice age cycles. Whatever thermostat the Earth has doesn’t seem all that good.

    Willis Eschenbach says: September 22, 2013 at 1:10 pm

    The planet’s temperature varied by ± 0.3°C over the last century. This is a regulation to within about ± 0.1% … on a free-running system which is regulated by nothing more substantial than wind and water. …. If you know anything about heat engines, you’ll agree that that is a fantastic governor … w.

    What manner of observational methods, data collection methods, experimental methods, and analytical methods would be necessary either to demonstrate or to disprove Willis’ theory that the earth’s climate system operates mostly as a self-regulating, self-stabilizing heat engine?

    Beta, that’s an excellent question. I’ve been collecting evidence wherever I can, from the TAO buoys and the Argo floats and the CERES dataset, and I’ve published it all on WUWT. I’m always looking to add to that.

    Regarding Thomas’s questions, the earth appears to be “bi-stable”, that is it has two different states, and it can shift from one to the other. I hold that the shift is totally explained by the Milankovic changes in the planet’s orbit, leading to a long-lasting shift in the planetary albedo. Each of the two states, however, appear to be stable, indicating that my mechanism is operating during both parts of the ice age cycles.

    w.

  246. David says:
    September 23, 2013 at 1:23 pm

    “I hold that changes in forcing only marginally and briefly affect the temperature. Instead, I say that a host of emergent thermostatic phenomena act quickly to cool the planet when it is too warm, and to warm it when it is too cool.”

    A glance at any global temperature data, on any timescale, proves this idea wrong.

    David, take the timescale of 100 years. Over that time, the global average surface air temperature has not varied by more than ± 0.1%. As I said above, any engineer will tell you that that is very, very good regulation.

    w.

  247. Salvatore Del Prete says:
    September 23, 2013 at 2:59 pm

    Willis your arguments do not hold water, when it comes to volcanic eruptions and there impact or lack of impact on the climate. Read my earlier pos. In addition hundreds of studies do not agree with your study.

    Salvatore, you don’t seem to understand the process of falsification. If you think that I’m wrong, then you need to point out the flaws in my logic or my math or my data.

    Because simply claiming that my arguments “don’t hold water” doesn’t do it. Neither does your risible claim that there are “hundreds of studies” that disagree with me. If you think I’m wrong, first QUOTE MY WORDS that you think are wrong, and then show us, not simply claim but show us, why my words are wrong.

    w.

  248. Willis Eschenbach says:
    September 23, 2013 at 3:15 pm
    “David, take the timescale of 100 years. Over that time, the global average surface air temperature has not varied by more than ± 0.1%. As I said above, any engineer will tell you that that is very, very good regulation.”

    w.
    Short term and minor events, yes. Longer term and major events, not so much. Toba 74,000 ybp is still hypthesized by many to have cooled the climate for 1000 years and possibly started the glaciers growing due to 5 to 10 years of winter and the resultant increase in earth’s albedo. Yea, I know others disagree with this. Generally, though, I have to agree with your theory re regulation. It is one of those myriad of issues which somehow always work out to the benefit of life. Now how can you figure that? Energy levels in the atom, expansion rate of the universe, and so on, all just right. It’s the Goldilocks syndrome.

  249. Willis you are trying to show ever so hard that volcanic forcing is much less, when past evidence shows the contrary.

    For example look at the temperature data following the Mt. PINATUBO eruption from Dr. Spencer’s web-site where he shows month by mothn temp. data goiing back to 1979.
    Look at what happened to the temperatures following the Mt. Pinatubo eruption, they went down in the face of an El Nino.

    Why it is hard to prove what Willis wants to prove.

    Each volcanic eruption is different in regards to it’s location, composition of what it ejects into the air, height/amounts of material it ejects into the air making a volcano /temperature correlation amost impossible to obtain, not to mention other climatic items acting in concert or against the volcanic climatic effects.

    Willis says and it it correct in a sense( but not really) the following: a host of emergent thermostatic phenomena act quickly to cool the planet when it is to warm, and to warm it when it is to cool.

    If true 100% of the time as you convey more or less; reconcile that statement with the fact the earth has had many ABRUPT climatic changes in the past, and have shifted from a glacial state to an inter glacial state many times in the past. Apparently that statement does not hold up under all circumstances for if it did the climate would never have abrupt climate swings in temperature both up and down, or vary from a glacial to an inter glacial period.

    Willis you make it work out okay with the data you use, the problem is the data you are using ,like global climatic models can’t account for all the variables that are going on in the climatic system of earth, which can translate into the very percise data you suggest will happen if a volcanic eruption takes place. You don’t have all the pieces of the puzzle to make such a sure fire conclusion in my opinion.

    Joe D’Aleo had a very interesting arcticle about volcanism and climate effects. Came out July 19th on icecap.com. A very different take.

    Still Willis,you put in great efforts in what you do right or wrong, that I must admit.

  250. Willis, I have been interested in ocean fertilization by dust and its effect on marine CO2 uptake, and ultimately, carbon mineralization.
    Plot the 1980-2002 with Log(Dust) & Keeling and have a look at the drop in the rate that atmospheric CO2 rises. Don’t think this is a SST cooling as the oceans do not respond that quickly.
    Thanks for the aerosol link, never had the monthly data before.

  251. I would say the temperature data following the Mt.Pinatubo eruption showing a drop in world average temerature. despite an El Nino is very convincing counter evidence to what you conclude.

    Again there are many studies, and to try to get an accurate picture one must look at many studies and weigh the pros and cons.
    .

  252. That data shows around a -.5c drop in global temp. following the Mt. Pinatubo eruption , for a short period of time even though an El Nino is going on. This is based on satellite temperature data from Dr. Spencer’s web-site.

  253. WIllis wrote: “I’m using the canonical relationship of the current (and in my opinion incorrect) climate paradigm. This is that ∆T = lambda ∆F, where F is forcing, T is temperature, and lambda is climate sensitivity. As I tried to say up top, I don’t believe that’s how the world works … I’m just trying to show that by their terms the so-called “sensitivity” is very small.

    ∆T = lambda ∆F only applies to equilibrium situations where the temperature has had plenty of time to come into equilibrium with the change in radiation. As you can see from my calculations, the effects of volcanos don’t last for long enough for equilibrium to be reached. You need to take the heat capacity of the mixed layer into account to properly model for the temperature change after an eruption. A forcing of 1 W/m2 can’t change the temperature of the mixed layer more than 0.3 degC and it takes more than a year for half the change to occur (0.15 dgC). A transient forcing several times bigger won’t produce easily detected spikes in the historic temperature record.

    Steven Schneider first became famous predicting “nuclear winter”, which he later admitted turned out to be “nuclear autumn”. Similar alarmism about volcanos has led you to expect “volcanic winter” following a major eruption (“the year without a summer”), but calculations suggest “volcanic autumn”, a cooling difficult to detect against background noise.

    ∆T = lambda ∆F is sensible because the earth’s temperature is controlled by the amount of energy it receives from the sun and other factors. T is a function of F. If the function is well-behaved, a linear relationship between ∆T and ∆F can be used to approximate this function. The linear relationship will breakdown breakdown if you apply it to changes that are too big. All of the negative feedbacks that you sensibly argue could control climate can be represented by a smaller lambda, climate sensitivity – you don’t need to throw away the whole equation. And you don’t need to throw it away because ∆T can be hard to detect against a background of natural climate variability.

    If the function is not well-behaved, we may be in a chaotic region that can’t me approximated by a linear function. The physics of radiation doesn’t produce chaotic functions, but fluid dynamics does.

  254. Willis says :

    “One problem with doing this particular linear regression is that the volcanic forcing is approximately trendless, while the temperature has risen overall. We are interested in the short-term (within four years or so) changes in temperature due to the volcanoes. So what we can do to get rid of the long-term trend is to only consider the temperature variations around the average for that historical time. To do that, we subtract the Gaussian average from the actual data, leaving what are called the “residuals””

    You’re subtracting out part of the signal – potentially a large part firstly. Secondly your results are completely dependent on the type of smoothing you’re doing for the subtraction. Where are you sensitivity tests for various forms of smoothing? I see you smoothed out the long-term signal during the late 1800s and early 1900s where volcanic aerosols were prevalent… Your thesis is ultimately going to give you the answer you like just based on methodology.

  255. Willis Eschenbach says:
    September 22, 2013 at 1:01 pm
    “We are talking about the surface temperature … and as the poorly named “greenhouse effect” clearly shows, we can have identical conditions 1/4 mile above the TOA, and very very different conditions at the surface.
    That is why rain is a cooling mechanism … because we’re talking, not about the heat content of the planet, but the temperature of the surface, and rain definitely cools the surface.”

    Actually the conditions “above TOA” drive the conditions at the surface – if you have TSI rise it drives SST/OHC and land surface temperature/heat content directly, while at the same time changes in solar wind flux modulate GCM, you get less low-cloud seeding, while you get more water vapor in atmosphere (having considerably higher heat capacity than air) and you end up with higher both tropospheric and surface air temperatures. There’s no known “homeostasis” effect countering solar activity rising trends and its GCM/cloud cover positive feedback – if there would be something like that we wouldn’t observe the very significant rising of the SST anomaly in the 20th century – which is quite well in terms with the solar activity rise even in absolute surface heat content numbers.
    The water cycle cools the sea and land surfaces by taking considerable specific and latent heat (yes both – ~85% latent heat, ~15% specific heat) from it – more than third of the energy Earth surface absorbs from solar irradiation, converting it to heat (3.8×10^24 J yearly) leaves the surface via water evaporation, subsequently the specific heat contained heats the atmosphere all the way up (humid warm air has lower density than dry cold air) and the latent heat at the point where the water condenses, both changing ALR considerably. Eventually the water falls back as rain (or snow), but how it cools the surface is not so much by the temperature-of-the raining-water with temperature-of-the-surface difference (rather almost negligible effect), but again by its subsequent evaporation, which in average “costs” the surface at least 2.5 GJ per m^2 per year (or 2.6 MJ – 2.26 MJ latent heat+0.35 MJ specific heat per kilo of water evaporated – yes in average almost 1 ton of water is evaporated from 1 square meter of Earth surface, bulk of it of course from ocean). Which btw is result of water/air interface optical properties which considerably lower the real-world sea water emissivity and for example make the Kiehl-Trenberth energy budget with the “396” W/m^2 surface radiation an utter nonsense, because 2/3rds of the Earth surface is ocean which in average doesn’t radiate more than ~264 W/m^2.

  256. DocMartyn says:
    September 23, 2013 at 3:38 pm
    ====
    We’ve been doing this for decades….
    Look for “red tide” info, Miami has the longest running air quality for the Carib, etc

  257. wayne said:

    “I, as others, do greatly appreciate your work and insight into these areas though I hope you have the flexibility to possibly be incorrect in a few of your viewpoints where we do differ, one of those being that you do not feel the mass of atmospheres matter at all, I do. Keep an open mind, stay a real scientist. I’ll always give you that same courtesy. Even here on this post.”

    Seconded.

    Something has to set the energy content of the system around which any thermostatic mechanism must work.

    The variations in weather phenomena on Earth that Willis relies on emerge when the system diverges from that baseline energy content and operate as a negative system response rather than as the element that primes the system.

    Volcanic activity undoubtedly causes a divergence and the system response is circulation changes that affect the hydrological cycle as Willis correctly observes.

    If there were no water then the circulation system alone would have to achieve the thermostatic effect without the assistance of the highly efficient hydrological cycle and in that event the circulation changes would need to be far more violent to maintain system stability.

    Mars is a good example, water is absent and the atmospheric circulation periodically kicks up planet wide dust storms as the Martian equivalent of Earth’s hydrology based thermostatic process.

    On Mercury, the same effect is achieved by the phase changes of rock from solid to liquid and back again between the day and night sides.

    On Venus, the massive surface pressure causes the atmosphere to be so dense at the surface that it is halfway to being a solid and the consequent extreme power of the winds is the Venusian thermostatic process in action.

    Every planet has its own form of thermostatic process which is related to both the composition and quantity of atmospheric mass with the quantity of mass setting the baseline energy content and the composition determining the circulatory configuration required to achieve a successful thermostatic outcome.

    I look forward to the day when this ‘clicks’ in the minds of Willis and others who currently deny the role of atmospheric mass (and therefore surface pressure).

  258. Joe Born,

    Hi Joe,
    “As my excuse for choosing this time to pick that particular nit, all I can say is that you’ve given this formula more than once before, and each time I’ve failed to understand how it applies to the usual situation with which you’re dealing, i.e., to matching data representing averages over intervals that are potentially significant fractions of the system’s major time constant.”

    We are just making slightly different starting assumptions about the nature of the input and output data, I believe. The input stimulus (forcing) is normally given as an end-period cumulative, F0,F1, F2,F3 … for years 0,1,2,3…(say).
    The formula used by Willis uses the stimulus at the START of the timestep to yield the temperature value at the END of the timestep. So for year 1, your X(1) value should be set to (F1+F0)/2. For year 2, your X(2) value should be set to (F2+F1)/2 and so on. The difference term, X(2)-X(1), then approximates the estimated stimulus imposed at the start of the second timestep. The resulting temperature values on the other hand correspond to the time at the END of the timestep, not central year values.
    I have just tested this against analytic solutions for your choice of parameter values. There is a maximum error of 14% for the first timestep when tau is set to just twice the timestep. This rapidly diminishes to less than 2% error after 5 timesteps, and less than 1% error after 10 timesteps. (I used a linearly increasing forcing for the tests. For a step function input, the numerical error should be zero if the step is applied at the start of the first timestep.) The errors in the other cases with higher values of tau were negligible.
    Why don’t you modify your program to test this, and if you still have a problem, I will post a spreadsheet with results.

  259. Hi Willis,
    I have a couple of problems with the methodology you have applied here.
    First, your method of getting to the residual temps actually puts a systematic bias into the result because the volcanic effect in the temperature domain is not itself a zero-mean effect. It should pull down your Gaussian smooth and hence reduce the magnitude of the signal you are looking for.
    Secondly and more importantly, you can’t use the regression coefficient as a measure of sensitivity. The magnitude of response is frequency dependent. You can test this easily by setting up a sinusoidal forcing input and then looking at the results as you reduce the periodicity while maintaining the same amplitude.
    If I recall, Pinatubo had something like a 4 W/m2 forcing, but gave rise to (only) an estimated 0.6 degrees drop in temperature. If it had happened over a longer (shorter) period of time, then the temperature drop should have been bigger (smaller).

  260. “…Thanks, Steven. I’ve shown that I can duplicate the models’ global temperature output with a one-line equation which fits easily on the back of an envelope.
    So I fear that your continual claim that it’s oh-so-complex is completely contradicted by the ultimate simplicity of what the models are doing….I got an R^2 of about 0.98 between my simple emulation and the models.”

    Willis-san: I’ll respectfully remind you that it was stated (by others) at the time you posted the one-line equation that what you produced was not itself a model, but a model of a model. The 0.98 value for your r², instead of amusing you, should have raised a flag that you were not operating from first principles. It’s way too good a fit. When was climate science (or what passes for it in the world of Michael Mann) ever that precise? Never. Clearly your math, though perfectly valid, is operating at a different level than the models.

    Each fine-grain cellular detail of a model is ultimately summed, averaged, and replaced by a single global number, e.g., global albedo. Those finalized composite numbers have been incorporated in your model-of-a-model, but without the cellular detail that led to them. Thus there’s a level of detail in the climate models that your one-line method does not deal with. You really, really need to comprehend this to understand exactly what you’ve achieved and what you haven’t.

  261. RC Saumarez says:
    September 23, 2013 at 7:39 am
    @Tintoolman.

    Thankyou for clarifying your level of maturity.

  262. Willis writes “It simultaneously says that yes, volcanoes do affect the temperature … and yet, the effect is vanishingly small—only about a tenth of a degree per doubling of CO2.”

    IMO comparisons of volcanic aerosol forcings and CO2 forcings are not comparable. They’re most certainly not equivalent.

    A volcanic forcing means less energy arrives at the earth’s surface. Full stop. Its not there at the surface. On the other hand CO2 forcing is one factor in how the atmosphere moves energy from the surface out to space and the atmosphere can potentially do all sorts of things to change that rate.

  263. jorgekafkazar “Thus there’s a level of detail in the climate models that your one-line method does not deal with. You really, really need to comprehend this to understand exactly what you’ve achieved and what you haven’t.”

    What does it say to you then?

  264. Hi Stephen, agree with much you are saying that physics will always balance using proportionally the most effective methods available at the highest rate… but don’t want to stray too far off topic, a courtesy to Willis. We’ll discuss that later.

    But would you please take the time to read tumetuestumefaisdubien1’s comment for some further discussion at talkshop. This is bringing back memories of nearly a year ago, remember my figures showing the ocean’s emissivity much lower, believe near 0.65 IIRC, that everyone so objected to as unrealistic. Don’t know tumetuestumefaisdubien1’s qualifications but he is spitting right back the same figures I came up with, ~260 W/m², not 396, and a much lower emissivity. Maybe time to revisit that area. That was when I was so complaining that the Kiehl-Trenberth energy budget was completely incorrect, could not realistically be “balanced”, and then published that spread to show what I had come up with that did balance and here are the same values again out of the blue.

    tumetuestumefaisdubien1, if you happen to read this comment, thanks for your comment, I’ll take a look back on those points you are making, seems they so parallel what was found separately a year ago but I’ll have to dig back a bit.

  265. Using calculations from “Stratospheric loading of sulfur from explosive eruptions.” Bluth et al (1977), it seems that the sulfate peak after an eruption is between 2 to 3 months later. The sulfate then sediments out in about 30 to 40 months.

    In “Eruptions that shook the world”, Dr. Oppenheimer (session 8.3.2) suggests that “denser sulphur clouds grow larger particles” and, therefore, are “less effective at reflecting sunlight back to space”. Thus being, larger, explosive, (super)eruptions (like Toba’s) affect climate cooling in a proportionally lesser way than non-explosive ones, which release SO2 in finer particles which may float longer at the troposphere.

    My pet idea is that Carbonyl Sulfide (OCS) can contribute to the background levels of sulphate in the Junge layer, but it’s just an idea based on the long lifetime of OCS, and that it dissociates easily in 200 nm and 270 nm light. (stratosphere).

    Just some random thoughts should you wish to ruminate upon this further.

  266. Salvatore Del Prete says:
    September 23, 2013 at 3:36 pm

    Willis you are trying to show ever so hard that volcanic forcing is much less, when past evidence shows the contrary.

    For example look at the temperature data following the Mt. PINATUBO eruption from Dr. Spencer’s web-site where he shows month by mothn temp. data goiing back to 1979.
    Look at what happened to the temperatures following the Mt. Pinatubo eruption, they went down in the face of an El Nino.

    Provide a link, or I’m not interested. And I have shown that no, past evidence doesn’t show the contrary. If you think my evidence is wrong, QUOTE MY WORDS that you think are in error and tell us exactly what is wrong with them.

    Waving your hands at some study or other won’t do it. You can’t show that they’re right, that’s not possible in science, all you can do is falsify studies. So you need to show that I’m wrong.

    w.

  267. Salvatore Del Prete says:
    September 23, 2013 at 3:48 pm

    I would say the temperature data following the Mt.Pinatubo eruption showing a drop in world average temerature. despite an El Nino is very convincing counter evidence to what you conclude.

    Again there are many studies, and to try to get an accurate picture one must look at many studies and weigh the pros and cons.

    Meaningless. You haven’t said where and what I’ve done wrong. You haven’t provided a scrap of data, or a link to a single study. You’re just flapping your lips … sorry, not impressed.

    w.

  268. Frank says:
    September 23, 2013 at 4:34 pm

    WIllis wrote:

    “I’m using the canonical relationship of the current (and in my opinion incorrect) climate paradigm. This is that ∆T = lambda ∆F, where F is forcing, T is temperature, and lambda is climate sensitivity. As I tried to say up top, I don’t believe that’s how the world works … I’m just trying to show that by their terms the so-called “sensitivity” is very small.

    ∆T = lambda ∆F only applies to equilibrium situations where the temperature has had plenty of time to come into equilibrium with the change in radiation.

    Near as I can tell, it doesn’t apply at all to the real world. However, it applies 100% to the models, that’s how they work.

    w.

  269. jorgekafkazar says:
    September 23, 2013 at 7:35 pm

    “…Thanks, Steven. I’ve shown that I can duplicate the models’ global temperature output with a one-line equation which fits easily on the back of an envelope.
    So I fear that your continual claim that it’s oh-so-complex is completely contradicted by the ultimate simplicity of what the models are doing….I got an R^2 of about 0.98 between my simple emulation and the models.”

    Willis-san: I’ll respectfully remind you that it was stated (by others) at the time you posted the one-line equation that what you produced was not itself a model, but a model of a model.

    Since I stated from the start that it was a “black-box” analysis of a model, why on earth would you think I didn’t know that it is a model of a model?

    The 0.98 value for your r², instead of amusing you, should have raised a flag that you were not operating from first principles. It’s way too good a fit. When was climate science (or what passes for it in the world of Michael Mann) ever that precise? Never. Clearly your math, though perfectly valid, is operating at a different level than the models.

    “Operating from first principles”? I fear you truly don’t understand what a “black-box” analysis is. You might read my post “Life is like a black box of chocolates” which discusses that concept.

    Best regards,

    w.

  270. Willis, when I see a post by Salvatore Del Prete I usually read as far as “Del” then skip to the next post. I suggest you save your time. Paul_K has a very good level of understanding and has raised what look like important points.

  271. Paul_K : “If I recall, Pinatubo had something like a 4 W/m2 forcing, but gave rise to (only) an estimated 0.6 degrees drop in temperature. If it had happened over a longer (shorter) period of time, then the temperature drop should have been bigger (smaller).”

    If it had lasted longer the total forcing would have been greater. I would have thought this emphasises the need to convolve the input with the laplacian impulse response (decaying exponential) or equivalent like the recursive formula Willis mentions but does not use the main analysis. This also touches Frank’s comment about dimensions: integrating the forcing is the same as differentiating the response.

    Willis assures us that the results are about the same but I’d rather see a method that is dimensionally (ie physically) correct than one that is not.

    He says Fe is ECS equivalent of the instant forcing, but does not explain how it is derived or what assumptions are used to get there. (His spreadsheet nearly crashes Libre Office and his formulae don’t work, so I could not see what that included.) I suggested Willis’ Fe maybe the exponential integral but he has not commented on way or the other.

    Would you say that the regression coeff using the integral would give something that could be compared to sensitivity?

    re. Mean zero etc.:
    The gaussian high-pass filter Willis uses seems to be appropriate since several studies seem to show a time const of around 6 months. At five times that the residual will be <2% so the filter seems good. That will remove long term natural trends and steadily increasing CO2 (however big/small it may be) as well as any net downward trend induced by periodic volcanic forcing. I think that last point is your objection and makes sense. It would reduce the correlation.

    Would processing the volcanic forcing (in what ever form is used) to have a mean of zero , before doing the regression, sufficiently address that problem?

  272. Paul_K says:
    September 23, 2013 at 7:35 pm

    Hi Willis,
    I have a couple of problems with the methodology you have applied here.
    First, your method of getting to the residual temps actually puts a systematic bias into the result because the volcanic effect in the temperature domain is not itself a zero-mean effect. It should pull down your Gaussian smooth and hence reduce the magnitude of the signal you are looking for.

    First, Paul, it’s always good to hear from you.

    Next, I can see what you are getting at. It’s a valid issue. However, the effect on the results was not what I expected. The effect you’ve identified shifts the residuals vertically by the average of the volcanic effect in the temperature domain. To see how large this was, I shifted the residuals by the amount of the total effect (lambda = 0.05). All it did was slightly lengthen the time constant tau from 6.5 to 7 years, the value of lambda was almost unchanged.

    Now, the average of the fitted lagged forcing is 0.02 °C, and I’ve tested it with 2.5 times that amount. So I’ll stick with 0.05 as the value for lambda, and around 7 years for the time constant.

    Secondly and more importantly, you can’t use the regression coefficient as a measure of sensitivity. The magnitude of response is frequency dependent. You can test this easily by setting up a sinusoidal forcing input and then looking at the results as you reduce the periodicity while maintaining the same amplitude.

    I don’t understand this. I’m using the regression coefficient as the measure of the amount of variance in the temperature that is explained by the volcanic forcing, that is to say a measure of “goodness of fit”. I’m not sure what you mean by a “measure of sensitivity”. I’ve only used it to estimate the lag between the forcing and response. I haven’t used it at all in the lagged analysis.

    If I recall, Pinatubo had something like a 4 W/m2 forcing, but gave rise to (only) an estimated 0.6 degrees drop in temperature. If it had happened over a longer (shorter) period of time, then the temperature drop should have been bigger (smaller).

    According to the Sato dataset, the Pinatubo forcing (Figure 1) was 3.5 W/m2. You’ll have to point out the 0.6°C drop. Let’s play “Spot The Volcano” again. Here’s the data:

    There are two major volcanoes in that graph. Can you see your claimed 0.6°C drop in there? Can you identify the time of the volcanoes?

    In any case, this should make it clearer …

    I’m sure you can see the problem. Yes, there was a drop after Pinatubo … but it was only about 0.2°C, not the 0.6°C that is often claimed. That was the amount the models predicted for the drop.

    In addition, there are three or four similarly-sized drops in temperature which are NOT associated with an eruption.

    Finally, look at what happened after El Chichon, despite the forcing being nearly as large as that of Pinatubo … basically, nothing.

    Like I said … this myth of eruptions having some big cooling effect is hard to kill …

    Again, thanks for the ideas and questions,
    w.

  273. Greg Goodman says:
    September 24, 2013 at 12:39 am

    Willis assures us that the results are about the same but I’d rather see a method that is dimensionally (ie physically) correct than one that is not.

    None of these methods are dimensionally correct, because they all use a units fudge factor. Remember we are transforming forcing (W/m2) into temperature (°C), which are incommensurate.

    He says Fe is ECS equivalent of the instant forcing, but does not explain how it is derived or what assumptions are used to get there. (His spreadsheet nearly crashes Libre Office and his formulae don’t work, so I could not see what that included.) I suggested Willis’ Fe maybe the exponential integral but he has not commented on way or the other.

    Greg, as I mentioned above, questions such as that one are often answered in the underlying paper. In this case it’s the Sato paper, which provides a dataset of optical thicknesses “tau”, and says:

    The relation between the optical thickness and the forcings are roughly (See “Efficacy …” below):

    instantaneous forcing Fi (W/m2) = -27 τ
    adjusted forcing Fa (W/m2) = -25 τ
    SST-fixed forcing Fs (W/m2) = -26 τ
    effective forcing Fe (W/m2) = -23 τ

    And “Efficacy” refers to

    Hansen, J., M. Sato, R. Ruedy, L. Nazarenko, A. Lacis, G.A. Schmidt, G. Russell, et al. 2005. Efficacy of climate forcings. J. Geophys. Res., 110, D18104, doi:10.1029/2005/JD005776.

    (Again, remember I’m using their methods, but I’m not claiming that their methods are correct.)

    Since the effective forcing is smaller, that gives the higher value of the ECS. If you use the larger forcing Fi, it gives you the smaller value of the transient climate response (TCR).

    I should have clarified that in the head post, my bad. I’ll put it in as an update.

    w.

  274. @Willis Eschenbach

    You ask do I want an R^2 of 1.5? Is this supposed to be a serious question? If have pointed out that R^2 is not a proper metric of model fit. There are other more revealing methods, starting with the ACF. As regards ill-conditioning, fitting decay curves are notoriously ill conditioned and are subject to large errors in the face of variability of objective data. I would rather trust my experience and a wide range of scientific literature than your assertions.

    You have repeatedly asked me to tell you what is wrong with model, so I have told you,

    I have asked you to account for the fact that your model does not give results that are compatible with well known metrics of the temperature signal, namely the autocorrelation function. The implication is that your physical assumptions in your mpdel are wrong.

    You have completely failed to address this question, which is fundamental to modelling.. Instead, and forgive me if I have misunderstood you, you say that this is what other people do. I thought you claimed that this is “your” model.

    In this case you are simp;ly indulging in plagiarism. If you claim originality in the model, then you should try to defend it.

    On this, and several other posts, you have questioned my scientific ability as you have other people who clearly have a scientific background. I don’t give a fig what somebody like you thinks about my, or any other person’s scientific ability – you have zero authority with which to criticise any of us in the terms that you do.

    Since I have a PhD and have published a substantial number of papers using the techniques in question in high level journals and being a physician who applies these techniques in real time in invasive measurements in patients, I do this subject for real and you do not.

    Frankly, you come over as an opinionated, ill-educated, pretentious boor who simply rants at anybody who dares to question you..

  275. wayne says:
    September 23, 2013 at 9:41 pm

    “Don’t know tumetuestumefaisdubien1′s qualifications but he is spitting right back the same figures I came up with, ~260 W/m², not 396, and a much lower emissivity. Maybe time to revisit that area. That was when I was so complaining that the Kiehl-Trenberth energy budget was completely incorrect, could not realistically be “balanced”, and then published that spread to show what I had come up with that did balance and here are the same values again out of the blue.

    To explain my reasons for the claim: The Kiehl-Trenberth budget “396” (W/m2) surface radiation value is most probably based on blind theoretical calculation using Stefan-Boltzman law. But the result is physically impossible due to optical properties of water->air interface, which has average reflectivity 0.3-0.4 for the mid-IR waveleghts in question.The water simply doesn’t behave as a blackbody, to which the Stefan-Boltzman law applies – at least not without a considerable correction.
    In fact there’s estimated 505 km^2 of precipitation each year on Earth. The water must evaporate before. Bulk of it from ocean. The energy needed to evaporate such amount of water is more than third of the total solar shortwave energy the Earth surface absorbs.
    Given the absolute average sea surface temperature the Stefan-Boltzman law would determine the average ocean radiation to be about ~400 W/m2. But it can’t happen due to the reflectivity of the water/air interface which in average reflects 30-40% of the possible IR spectra (for liquid water at sea level pressure and possible temperatures) in question back – see here. But because the water is at the same time extremely opaque to the 290 K spectra (water transmittance for 10 micrometer spectra doesn’t allow the photons travel in the water more than ~0.1 mm) both the mid IR flux from sea and atmosphere basically keeps in the very surface skin of the water and contributes there chiefly to heating of the water to the point of evaporation – which substitutes there for direct IR radiation in direction to atmosphere. It can be calculated that ocean releases at least 34% of the heat resulting from solar shortwave radiation absorbtion this way instead of direct IR radiation into the atmosphere.
    (400 W/m2 – 34% = 264Wm^2. )
    Similarly it works for intermitent land water cover (well even for a water drop on a leaf), whenever it gets wet – because most of the possible water cover of land due to water surface tension is thicker than ~0.1 milimeter – a layer, which is able to absorb all over-10 micrometers mid-IR radiation encountering it. Basically all liquid water – fresh or salt, insolated or in dark evaporates due to concentration of the heat, causing IR emission in its thin skin – all caused by the optical properties of water. This properties aren’t so unique in nature, what is unique is the amount of liquid water on the surface of the Earth, which makes this effects being crucial for the surface/atmosphere complex behaviour interesting for climatology.

  276. Willis: “Thanks, Michael, but actually I’m not talking about feedback. Instead, I’m talking about a governed system, which is a very different critter. See my post called “It’s Not About Feedback” for a discussion of the difference.”

    “Regarding Thomas’s questions, the earth appears to be “bi-stable”, that is it has two different states, and it can shift from one to the other.”

    Willis, you need to make your peace with the word feedback. The bi-stable system you refer to implies a positive feedback bounded by -ve f/b. Your governor , like any governor, requires a negative feedback to work (as I pointed out in your original thread in relation to steam engine governors, which appear to be your conceptual base.).

    If you equate the word “feedback” with “linear negative feedback” only , of course it’s inadequate but if you refuse to recognise the whole concept of feedbacks, +/ve and -/ve, linear and non-linear, in describing the tropical storm “governor” and turn you back on the rest of science and engineering , I fear you will not get much traction for your hypothesis.

    As you know I am quite supportive of your basic idea.

    I again suggest you repeat this analysis on tropics and extra-fropical regions.

    If you extend the lag range I’m fairly sure that in the tropics you will find a similar negative correlation following the initial peak (which will itself be weaker).

    You will also find a stronger , more correlated signal in extra tropics (especially NH).

    That again will substantiate your hypothesis that it is the tropics that are stabilising global climate with respect to changes in radiative forcing.

    Best regards, Greg.

  277. Youkillmeandyoudomegood: says “The Kiehl-Trenberth budget “396″ (W/m2) surface radiation value ”

    What does that “value” represent? Surface upward IR would presumably include 0.7*S-B black body emission, 0.3*reflected solar, reflected atmospheric backscatter/re-emission.

    Not the same as “blind theoretical calculation using Stefan-Boltzman law.”

  278. Willis Eschenbach says: September 24, 2013 at 1:11 am
    I’m sure you can see the problem. Yes, there was a drop after Pinatubo … but it was only about 0.2°C, not the 0.6°C that is often claimed. That was the amount the models predicted for the drop.
    _______________________________________

    This is what I like about Willis – he goes back to the data, not the models. And even the layman can see what he is saying in that graphical data. I took an eyeball-squint of the lagged temperature-drop, and came out with 0.2 oc for Pinatubo and 0.15 oc for El Chichon, with a lag of one year. How does climate ‘science’ think it is anything different?

    And I like the way Willis will defend his ideas and assertions. There are so many so-called scientists (not just in climate ‘science’) who will make an assertion and then retire to a bunker – and actively prevent discussion by deleting adverse comments and banning dissenters. That, is not science. THIS, is science.

    .

  279. Willis, thanks for the extra details on Fe et al

    Here is an interesting quote from Douglass & Knox 2006, linked above :

    The sensitivity appears to be at variance with
    0.46 °C/(W/m2 ), corresponding to a T2x of 1.7 °C
    associated with a CO 2 doubling, expected by Wigley
    et al. [2005b]. There is no discrepancy if one acknowl-
    edges that the sensitivity for CO2 doubling is not
    necessarily the same as that for aerosol forcing. The
    reason for the difference in this case is that the models
    applied to CO2 doubling are designed for an approach
    to a long-term steady state (called ‘‘equilibrium’’) and
    that the assumed processes involved include positive
    delayed feedback. The CO2-doubling sensitivity is in-
    appropriate for the volcano forcing, which activates
    only rapidly occurring feedbacks.

  280. Willis: “None of these methods are dimensionally correct, because they all use a units fudge factor. Remember we are transforming forcing (W/m2) into temperature (°C), which are incommensurate.”

    Well they require the heat capacity of a mass to do the transformation which is why IRRC Frank was claiming the need to assess the interaction with the ocean mixed layer. The short cut you do here to go straight for the resultant temperature as seen in the surface data. I assume the ‘fudge factor” is sensitivity. It not not dimensional fudge but does assume a linear response, which is of questionable accuracy. The fudge is in choosing to do a linear regression not the units.

    That does not get around the need to integrate or differentiate as appropriate. With or without the fudge you can not regress power against energy. I think that point is unavoidable. This plays back into Paul_K’s point about the frequency dependency. This is saying the same thing.

  281. Hi again Willis,

    “I’m sure you can see the problem. Yes, there was a drop after Pinatubo … but it was only about 0.2°C, not the 0.6°C that is often claimed. That was the amount the models predicted for the drop.”
    I agree that it is remarkably difficult to pull out a volcano signal because the high frequency variance in the temperature datasets from ENSO, solar, cloud fluctuations and other effects (including measurement and methodological errors) leaves a low S/N ratio. That said, I think you are lowballing the estimate somewhat. The (MONTHLY) Hadcrut3 dataset shows a 0.38 deg C change in temperature from June 1991 to 3Q 1992 when the Pinatubo volcanic cooling was approaching its low point. This is before any correction for known ENSO and solar effects. The satellite data (UAH or RSS TLT data) shows a substantially larger temperature change – 0.5 to 0.6 deg C still without correction for ENSO and solar. See here for example:- http://wattsupwiththat.com/reference-pages/global-weather-climate/global-temperature/

    Corrections for ENSO and solar tend to increase the magnitude of the temperature change by order 0.1 deg C.
    Studies by sceptical scientists (Douglass and Knox 2005, Roy Spencer blog comments) and mainstream scientists (Soden et al 2002, Wigley 2005) alike have accepted a Pinatubo-induced TLT temperature change of around 0.6 deg C after correction for known ENSO and solar. This does not make it right, of course, but it should make you pause for thought given the very low temperature response that your work is implying here.

    In addition to the information that you use here, for Pinatubo, we also have satellite-derived measurements of changes in TOA radiation. The integral of the change in net flux gives us some idea of the total energy loss from the system after Pinatubo. This is substantially larger than can be explained using your fitted parameters for lambda and tau. Shameful personal plug – see here
    http://rankexploits.com/musings/2012/pinatubo-climate-sensitivity-and-two-dogs-that-didnt-bark-in-the-night/ for my calculations, with an update here:-

    http://rankexploits.com/musings/2012/pinatubo-climate-sensitivity-revisited/

  282. I always enjoy Willis Eschenbach’s posts, but I do feel wary about his ‘heretical’ belief in some kind of natural ‘thermostat’ in the Earth’s climate system. A thermostat requires negative feedback in both directions around some central level, and some kind of damping mechanism to ensure that the system does not oscillate wildly. It is obvious that the Earth’s climate system has both negative and positive feedback components, but there is no a priori reason to think that negative feedback will prevail, or that feedback will be ‘damped’ to give the required stability. (Lovelock’s Gaia hypothesis would give us such a reason, but I don’t recall that Willis has ever endorsed Gaia, and I really hope he is not resting on such a shaky foundation.) In the absence of any a priori reason (i.e. some well-established scientific principle, such as the laws of thermodynamics), it all comes down to evidence. And I really don’t see much evidence that global mean temperature is thermostatically controlled. If it is, how come we get the major long-term changes, such as the MWP and the LIA, which presumably most of us here believe in?

  283. David B said:

    “And I really don’t see much evidence that global mean temperature is thermostatically controlled. If it is, how come we get the major long-term changes, such as the MWP and the LIA, which presumably most of us here believe in?”

    Solar induced changes in cloudiness would produce the MWP, LIA and Modern Warm Period.

    If global albedo changes then the effect at the surface is the same as if ToA insolation had changed and so that does reset the thermostat.

    “A thermostat requires negative feedback in both directions around some central level, and some kind of damping mechanism to ensure that the system does not oscillate wildly.”

    The oceans do the damping because of their huge thermal capacity and long term internal variability.

    The central level has to be provided by something that is constant, otherwise no stability. That is why it must be atmospheric mass (leading to surface pressure when gravity is applied).

    Willis is contemptuous of that idea.

  284. “I hold that changes in forcing only marginally and briefly affect the temperature. Instead, I say that a host of emergent thermostatic phenomena act quickly to cool the planet when it is too warm, and to warm it when it is too cool.”

    Which begs the question, apart from cooling from volcanoes, what else at the scale of ESNO events is causing it to be warmer and cooler?

  285. Would still appreciate if somebody were willing to tell me why my earlier idea is wrong. I’m sure it’s got to be, since it’s so simple, but I’ve no clue why and am curious.

    Thanks in advance,

    Russ in Texas

  286. RC Saumarez says:
    September 24, 2013 at 1:35 am

    @Willis Eschenbach

    You ask do I want an R^2 of 1.5? Is this supposed to be a serious question?

    Naw, it was supposed to be this strange thing called “humor” …

    If have pointed out that R^2 is not a proper metric of model fit. There are other more revealing methods, starting with the ACF. As regards ill-conditioning, fitting decay curves are notoriously ill conditioned and are subject to large errors in the face of variability of objective data. I would rather trust my experience and a wide range of scientific literature than your assertions.

    RC, you’ve made that same claim above, based on your overall experience, that small changes in the data lead to large changes in the model parameter.

    Me, I’ve actually tried the experiment a number of times using real data, and found no such thing. Given the choice between my experiments and your theory, sorry, but I’m going with the experiments.

    I’ve been very clear about my methods and my data. How about you grab the data and actually try making some small changes, and see what happens.

    You have repeatedly asked me to tell you what is wrong with model, so I have told you,

    I have asked you to account for the fact that your model does not give results that are compatible with well known metrics of the temperature signal, namely the autocorrelation function. The implication is that your physical assumptions in your mpdel are wrong.

    “Account for the fact”? Your claim is that my results, when compared with the climate model results, have an R^2 of ~ 0.98 but the ACF is different? Could be … but how different, and why does that make a difference? And why would slight differences in the ACF, particularly at longer lengths, invalidate the model? I’m not trying to emulate the ACF of the results, I’m trying to emulate the results themselves, and doing quite well.

    You have completely failed to address this question, which is fundamental to modelling.. Instead, and forgive me if I have misunderstood you, you say that this is what other people do. I thought you claimed that this is “your” model.

    I have not “failed to address” anything. I’ve not ever seen you produce a scrap of evidence that my results have a different ACF than the model results that they emulate. If you can show such a thing, not just claim it but come up with data that shows that the ACF of my model results is significantly different than the ACF of the climate models, I’m happy to address it.

    In this case you are simp;ly indulging in plagiarism. If you claim originality in the model, then you should try to defend it.

    Whoa, there, cowboy. I didn’t plagiarize one damn thing. The quickest way to get your vote cancelled is to accuse me of lying or of theft. Perhaps you engage in those kinds of things. I don’t.

    On this, and several other posts, you have questioned my scientific ability as you have other people who clearly have a scientific background. I don’t give a fig what somebody like you thinks about my, or any other person’s scientific ability – you have zero authority with which to criticise any of us in the terms that you do.

    Authority? You think that scientific questions revolve around authority? I don’t give a fig for authority, I’m interested in data and observations and math.

    Since I have a PhD and have published a substantial number of papers using the techniques in question in high level journals and being a physician who applies these techniques in real time in invasive measurements in patients, I do this subject for real and you do not.

    RC, that’s wonderful. My congratulations. What does it have to do with my analysis?

    Frankly, you come over as an opinionated, ill-educated, pretentious boor who simply rants at anybody who dares to question you.

    And yet here you are … still reading my opinionated, ill-educated, pretentious words.

    RC, you have said any number of times that there is some giant error in my “black box” analysis of the climate models. Here’s one of your claims, for example:

    I have in the past been forced to write a post on WUWT to correct the complete rubbish you wrote about signal processing – a subject that you have not studied in any depth.

    But when I went to the post you claims “corrects my rubbish”, I and my work are not mentioned anywhere by you. Not one word, not one scrap of my data, no discussion of my model, nothing. Despite that, you clearly walked away thinking you had shown something …

    Now, perhaps there is some huge error in my “black box” model. But I fear that you declaiming that an error exists is far different from you demonstrating and explaining the error. I just did a search of your comments on my threads, and I don’t find such an exposition anywhere.

    As a result, to data you’ve given me nothing. Well, nothing but grief. You may be correct, but jumping up and down and saying that you have a PhD and you’re a published expert … well, I’m afraid that at WUWT that loses you points. We’ve had an unending string of PhDs who it turns out were clueless … you can start with Michael Mann and move on from there. So claiming expertise merely makes people wonder. As my great-grandfather used to say, If you have to hang your diploma on the wall, there’s something wrong with your education.

    Now, I’m perfectly happy to reboot this discussion. You can try starting over, and cutting way, way back on the arrogant, paternalistic tone. If you think I’m wrong, and I have been many times, then I invite you to QUOTE MY WORDS that you think are wrong, and then explain to us exactly where and why I’m wrong.

    Simply put, around here, we follow the motto of the Royal Society, even though they no longer do so all that much. Their motto is “Nullius in verba”, which means “Don’t take anyone’s word for anything” … and I’m afraid that means that your PhD-backed, “published a substantial number of papers” word is worth absolutely nothing. Not only that, but the more you hammer on your qualifications, the less you impress anyone.

    If you want to get traction around here, you need to demonstrate errors, not simply claim that they exist. And I’m happy for you to do that, it’s how I learn.

    Best regards,

    w.

  287. Willis wrote in his post: “For the value of the forcing, I have not used the instantaneous value of the volcanic forcing, which is called “Fi“. Instead, I’ve used the effective forcing “Fe“, which is the value of the forcing after the system has completely adjusted to the changes. As you might expect, Fi is larger than Fe.”

    According to the link he provided, the effective forcing appears to correct the aerosol forcing for latitude and may not reflect any adjustment to changes after the system. Using the effective forcing and a lagged analysis may not correct for the amount of time it takes the system to respond and therefore won’t produce an Equilibrium climate sensitivity.

  288. RC Saumaurez, here’s the ACF for the Forster climate model results, and for my emulation.

    Yes, the ACF of the emulation is not an exact match to the ACF of the climate model results … but I fail to see why that slight difference invalidates my emulation as you claim.

    w.

  289. Frank says:
    September 24, 2013 at 8:33 am

    Willis wrote in his post:

    “For the value of the forcing, I have not used the instantaneous value of the volcanic forcing, which is called “Fi“. Instead, I’ve used the effective forcing “Fe“, which is the value of the forcing after the system has completely adjusted to the changes. As you might expect, Fi is larger than Fe.”

    According to the link he provided, the effective forcing appears to correct the aerosol forcing for latitude and may not reflect any adjustment to changes after the system.

    Thanks, Frank. Actually, the difference between Fi and Fe has nothing to do with latitude. It’s all explained in the link to the “Efficacy” paper.

    w.

  290. tumetuestumefaisdubien1 says:
    September 24, 2013 at 2:04 am

    The Kiehl-Trenberth budget “396″ (W/m2) surface radiation value is most probably based on blind theoretical calculation using Stefan-Boltzman law. But the result is physically impossible due to optical properties of water->air interface, which has average reflectivity 0.3-0.4 for the mid-IR waveleghts in question.

    Say what? The emissivity of water in the IR spectrum is on the order of 0.96 for the spectral region from 9 to 12 microns … which makes the reflectivity about an order of magnitude smaller than you claim. See e.g. here.

    Unfortunately, as a result of that error, your entire argument collapses.

    w.

  291. Willis Eschenbach says:
    “Like I said … this myth of eruptions having some big cooling effect is hard to kill …”

    Speak to Toba as Pinatubo was a fly speck on what was Toba. The earth seems to have, indeed, always come back, even from the Milankovich cycles, but some volcanic eruptions have long lasting effects. The problem is that all of the truly major ones being so far in the past rely on paleo data of questionable worth for determining their effects. Plus all of your regressions necessarily and understandably ignore all of the unmeasured exogenous variables which are also having a contemporaneous effect. Chaotic is a word that comes to mind for the system. But we do indeed always seem to eventually readjust as you postulate. Eventually is a big word here.

  292. Greg Goodman says:
    September 24, 2013 at 2:24 am

    Willis:

    “Thanks, Michael, but actually I’m not talking about feedback. Instead, I’m talking about a governed system, which is a very different critter. See my post called “It’s Not About Feedback” for a discussion of the difference.”

    Willis, you need to make your peace with the word feedback.

    Thanks, Greg. You seem to think that the two terms are equivalent. They are not, which is why we say that an engine has a “governor”, and we don’t say it has a “feedback”. It’s also why we note that James Watt invented the “flyball governor”, he didn’t invent the “flyball feedback”

    The difference is simple. A simple feedback applies a force which is either positive or negative. Neither the amount nor the sign of the feedback are a function of the current output of the system. Instead, the amount and sign of the feedback are fixed in advance.

    A governor, on the other hand, applies a force which moves the output of the system towards some predetermined “set point”. You could think of it as a house with a furnace, an air conditioner, and a thermostat that turns one or the other on depending on the temperature. As a result, both the sign and the amount of the feedback applied by the governor are functions of the current output of the system.

    All the best,

    w.

  293. Willis Eschenbach says:
    “3) However, this is not the whole story. The reason that the temperature change after an eruption is so small is that the effect is quickly neutralized by the homeostatic nature of the climate.”

    Any ideas about the what and wherefore of your “governor” and the homeostatic nature of the climate other than it is “invisible”? Cite me a previous paper if you want. What about the repetitive glaciations the earth experiences? The fact that the earth was much warmer (supposedly) 500mm ybp? Again, not that I disagree with the models being wrong as that is indisputable based upon observations. Just an old engineer trying to understand a little better.

  294. Jim G says:
    September 24, 2013 at 9:34 am

    Willis Eschenbach says:

    3) However, this is not the whole story. The reason that the temperature change after an eruption is so small is that the effect is quickly neutralized by the homeostatic nature of the climate.”

    Any ideas about the what and wherefore of your “governor” and the homeostatic nature of the climate other than it is “invisible”? Cite me a previous paper if you want.

    A good question. My first paper on the subject was “The Thermostat Hypothesis“. You might also enjoy “It’s Not About Feedback“, “The Details Are In The Devil“, “The Tao of El Nino“, and “Emergent Climate Phenomena“.

    Come back if you have questions or ideas,

    w.

  295. Here’s my version of a “governor”(with some “overshoot”to make it more stable kinda like a suspension system on a …car).

    http://en.wikipedia.org/wiki/File:Phase_diagram_of_water.svg

    and also:

    http://chemwiki.ucdavis.edu/Physical_Chemistry/Physical_Properties_of_Matter/Atomic_and_Molecular_Properties/Intermolecular_Forces/Unusual_Properties_of_Water

    and also:

    http://www.engineeringtoolbox.com/air-psychrometrics-properties-t_8.htm

    …something about latent heat changing the moisture content with out changing the temperature.
    Maybe i’m “overshooting”…whatever,just something to think about.
    Thanks for all of your interesting articles and comments.

  296. “Thanks, Greg. You seem to think that the two terms are equivalent. ”

    No Willis, I do not think they are the _same_ but a governor will include a feedback. IIRC the Watts governor was centrifugal weights which actuate a vertical slider which controls the flow of steam to the piston. As the balls turn faster the pressure is reduced, this is a -/ve feedback. In fact it will be non-linear alteration of the slider with rpm because of the central pivot of the ball arms. This will be non-linear in the sense of less than linear. I would imagine that the valve action would be highly non linear with relaxation to the slider position but probably something like forth power with the turbulent flow of the stream through it.

    That’s all back of envelop description but it gives an overall strong, continuous, non-linear negative feedback. This will be sufficient to control the engine rpm within a very close tolerance over a wide range of load and boiler pressure.

    It needs some kind of negative feedback to perform the role of governor. Being non-linear will make it more responsive at the risk of some overshoot and oscillation. It would be incorrect to say a governor IS feedback but it requires a feedback to work.

    As far a I can trust my volcano stack graphs , it appears the tropics goes one better and controls the integral. This would be like the Watts governor also ensuring that if the engine was temporarily slowed by increased load, it not only restored the engine speed to the control value but it also overshot in rpm to ensure total counter per hour did not slip. This way you could run a clock at the same time as the plough without it losing time.

    This also is not the same as a feedback but relies on feedbacks. Is ‘governor’ define precisely enough to say this action is or is not a kind governor too, I don’t know but suspect it is not that precisely defined.

    The point you seem to rail about is whether such an action is a feedback on the climate system as a whole. Well since it seems to fairly effectively counteract changes in radiative forcing the answer has to be yes and it have to be negative in sign. Whether it is linear or non-linera requires further inspection.

    In terms of the tropics I would suggest it’s strongly non-linear (governor++) . Ex-tropics benefit from the stabilising influence and get basic governor regulation, I don’t see an obvious reason to say this more than linear, though it could be.

    I hope that helps to explain why saying tropics are “a feedback” on climate does not mean the same a them being a linear feedback and how it is not contrary to your suggestion of ‘governor’ type behaviour.

    My volcano graphs are not rigorous science but do suggest strong non-linear feedbacks controlling temperature in the tropics. This is why I thought it would be interesting to run your analysis here on tropics / ex-tropics to see whether is contradicted or corroborated what I found.

    best.

  297. Jim G says:
    September 24, 2013 at 9:03 am

    I would consider Tambora in 1815 to have been a major eruption. It’s the only unambiguous VEI-7 eruption (ejection of at least 100 cubic kilometers) to have been directly observed in recorded history. The effects of a few others were observed by ancient civilizations, & of course archaeology & geology record the effects of others during the Holocene & earlier epochs.

  298. Willis and Paul_K: There is a discrepancy between the surface and satellite temperature records after Pinatubo. The global Hadley anomaly record Willis is using peaked at +0.37 degC in January 1992 (7 months after the eruption), the sixth highest monthly reading to this time*. Looking at three-month averages, the maximum drop in the surface temperature record appears to be about 0.3 degC below the highs seen in 1990 and the first half of 1991, not the 0.6+ degC drop seen in the satellite record. June-Nov 1992 averaged about 0.0 degC at the surface. Unfortunately, annual averages for the second half of the 1980’s ranged between 0.0 and +0.3 degC, so the 0.30 “drop” associated with the maximum influence of Pinatubo doesn’t look like anything out of the ordinary. If Willis was feeling nasty, he could accuse those who compiled the data that Paul_K reanalyzed of cherry-picking the record that shows Pinatubo having the biggest effect on temperature. If Paul_K had analyzed the surface temperature record, he certainly would have obtained a different ECS than he reported at The Blackboard. The confidence interval for any result should reflect the uncertainty in the data used to generate that result.

    *Not surprisingly, all of the highest temperature readings occurred in Jan-Mar of El Nino years, (usually January) and were somewhat similar to the blip in Jan 1992. It sure looks like 1992 was an El Nino year. Perhaps it takes some time for the high values in SSTs to spread into the atmosphere, because the January 1992 blip is not apparent in the satellite record. The validity of any El Nino adjustment seems critical to any analysis of this event. Figure 1 of Wigley (2006) shows the 0.3 degC discrepancy between surface and satellite (TMT) records in January 1992. The adjusted TLT record used by Paul_K looks different from the TMT and surface records. )Perhaps the TLT record is the one that is least distorted by El Nino events and therefore easiest to analyze.) http://journals.ametsoc.org/doi/pdf/10.1175/JCLI3611.1

  299. Paul_K: “The input stimulus (forcing) is normally given as an end-period cumulative . . . The formula used by Willis uses the stimulus at the START of the timestep to yield the temperature value at the END of the timestep.”

    Thank your for the clear explanation, which makes perfect sense if one assumes (appropriately, in my view) the fiction that the forcing is a superposition of step functions that commence at the beginnings of respective time intervals n, continue forever, and have respective magnitudes F_n – F_{n-1}. That is, the forcing is piecewise constant, at value F_n in each interval n.

    In Mr. Eschenbach’s formula T2 = T1 + lambda (F2 – F1) (1 – exp(-1/tau)) + exp(-1/tau) (T1 – T0), that is, the temperature T2 that prevails at the end of Time Interval 2 is the sum of the following right-hand-side terms:
    (1) the temperature T1 at the start of time interval 2,
    (2) the response lambda (F2 – F1) (1 – exp(-1/tau) to the step function (of magnitude F2 – F1) that started with that interval, and
    (3) the increment exp(-1/tau) (T1 – T0) in the response to the step functions that commenced in previous intervals and whose sum was F1 in the previous interval.

    Actually, I had already understood that the forcing data used (such as those in Column G in Mr. Eschenbach’s “Forcing” table) would be “end-period cumulative” if what you mean by that is consistent with the exegesis above.

    The source of my confusion is that I would never have thought that the temperature data the models are to match take the form you describe. I would have thought, for instance, that each monthly temperature datum (Column J) would represent an average over the respective month rather than, as I now see that Mr. Eschenbach’s formula implies, the value the temperature had reached at month’s end.

    And I confess that I’m reluctant to abandon that interpretation.

  300. Greg Goodman says:
    September 24, 2013 at 2:45 am
    Youkillmeandyoudomegood: says “The Kiehl-Trenberth budget “396″ (W/m2) surface radiation value ”
    What does that “value” represent? Surface upward IR would presumably include 0.7*S-B black body emission, 0.3*reflected solar, reflected atmospheric backscatter/re-emission.
    Not the same as “blind theoretical calculation using Stefan-Boltzman law.”

    Let’s stop presuming and see what K. Trenberth really says: “…we have taken the surface skin temperature from the NRA at T62 resolution and 6-h sampling and computed the correct global mean surface radiation from (I) [R = εσT^4] as 396.4 W m−2. (a href=”http://www.cgd.ucar.edu/staff/trenbert/trenberth.papers/TFK_bams09.pdf”>here, page 315)

    So Trenberth himself says he applied the Stefan-Boltzman law on measured surface temperature and that it is the way how he obtained the 396 W figure.

    And as I think about, it must not necessarily be a wrong approach if we would assume the energy for the surface evaporation is chiefly provided by the solar irradiation and atmospheric mid-IR backradiation – which seems to me even quite likely after all. In such case we in fact at all don´t need to postulate much lower water emissivity than 1, the Trenberth surface radiation value would be more or less correct and only remaining major problem with the budget would be that it overestimates the TOA insolation some 4 W/m^2 if compared to TSI average measured by SORCE-TIM instrument last decade, which doesn’t seem to be a result of a systematic error.

  301. Jim S, I already commented on why glaciation does not disprove Willis ‘governor’.
    September 22, 2013 at 10:43 am
    No one is claiming climate had been perfectly stable for 4.3 billion years and that this is thanks to tropical storms.

  302. [R = εσT^4] , what’s ε ?

    [Emissivity, the amount of radiation emitted/absorbed by a substance. For first-order calculations it can be taken as 1 for the surface of the planet. Use 0.9 or so if you want more accuracy. It doesn’t affect the energy based calculations for balances and the like, because it scales both incoming and outgoing the same. -w.]

  303. “And I confess that I’m reluctant to abandon that interpretation.”
    Joe, the point is that the start of month, end of month thing only matters if you have tau=2*step as in your exaggerated example. Paul_K reported specific figures and the error is minimal in the context of this data and tau of about 6 or 7 steps.

  304. There was another thread awhile back that discussed in part the effects of volcanoes. I see it written by you. One thought that I had at that time was perhaps the severity or lack of it depends on where the solar cycle is when the volcano erupts. Could it be that the major solar and atmospheric mechanisms that are at work at the time of the eruption, end up modifying the effects of the volcano by either enhancing or diminishing the effect?

    There is another thought that came to me several weeks. I mentioned it once before and I still am wondering about the ‘why’ of what it appears to show. The reason that it might fit into some of the conversation here has to do with comments that mention governors, or modulators. This chart from SIDC shows the excess ssn count between the hemispheres…http://sidc.oma.be/sunspot-index-graphics/wnosuf.php
    When the North is dominant it looks like it corresponds closely with the cooling trend of our planet, and when the South is dominant it corresponds closely with the warming trends. How could this be? Is it possible that there is a slight directional change in the output or focus of the Sun from this? Is this a governor/modulator mechanism?

  305. Greg Goodman: “You may have an interesting point. But how about you post a graph produced by you code so that we can see what the effect is without every man jack of us having to do it just to see your point?”

    I apologize for failing to respond; I somehow missed your comment before. Yes, I probably should have gone to the trouble of figuring out to post a drawing. But with so many folks here saying so many abstruse things, I guess my mental picture of the readership is of a bunch guys who have R consoles already open and waiting for code to be slapped into them.

    In any event, I believe we’ve been overtaken by events; Paul_K graciously gave an explanation above (and I followed up).

    As to the “spin up” issue, the approach I’d take would be to use the above routine’s y0 as one of the parameters the modeling routine tweaks. Although I assume that such an approach could be costly in some situations, I don’t think it would ordinarily present a problem in the case of a simple single-order, scalar system.

    Again, my apologies for ignoring your comment.

  306. There is the link Willis, which shows the Mt. Pinatubo eruption had a much greater influence on the climate then you try to convey. Note this was in the face of an El Nino at that time.

    Again Willis it is impossible to come up with what you came up with due to all the variables of the climate system, let alone the variability of each individual volcanic eruption.

  307. @Willis Eschenbach.

    I can’t comment on your ACFs becauase the time axes has no units.

    You don’t understand why ACFs are important? Let me explain.

    The ACF is a measure of lag in a system as measured in this contect. Climaye temperature data shows persistance. This means that there is correlation between samples at long time intervals. It is well known, eg Luck & Lederer and others, McIntyre, Tol for example.

    It is well known that the ACF of temperature does not conform to an 1st Order ARMA process. The fall off in the early lags is too great and the persistence in long lags is too great to be accounted for by the behaviour of a single capacitance. There has been musch speculation about representing the temperature series as a Hurst process, which is of course non-linear. Alternatively, the ACF can be approximated by a multi-compartment model and given the quality of the data it is difficult to distinguish between the two.

    Persistance has a very important role in system dynamics, especially in non-linear systems and cannot be ignored. It is a fundamental property of a system and is often only minimally represented, if at all, in the use of standard entropic measures such as R^2. This is not an appropriate measure to characterise a system and unless one is very precise in one’s definitions..

    You told me to grab some data and do an analysis. Thank you, I already have. See:

    http://judithcurry.com/2012/02/19/autocorrelation-and-trends/

    I wrote this after Luck et al presented their analysis. I was somewhat unconvinced and did the calculations as a reality check. Since this was for a blog, not a journal, I have gone to some length to explain the principles of the analysis in a simple way as I am aware that signal processing is a black art to many people.. I am perfectly happy to discuss the underlying mathematics with anyone and I am well aware that there are several approximations in the analysis, as there are of course in Luck et al’s
    .I would add that estimation of persistance from ACFs with a strong seasonal signal is not straightfoward. Low pass filtering the signal will give wildly incorrect results. Highly precise multi-notch filtering is highl problematic. The method that I used is one that is used in some biomedical signals where one is looking for subtle persistances in the presence of wide variability. It also has its limitations, bit I go to some length to establish the effect of the method on the ACF.

    Since you told me to put up or shut up, I wrote an essay on the difference between statistical and physical modelling and emphasised how one could be mislead into thinking that a non-linear system is linear, again, in a very simplistic way. It didn’t see the light of day on WUWT, but I am sure that you can recover the document from Anthony Watts.

    .
    I

  308. Willis – If you’re still monitoring this thread: It could be interesting to see if there’s a volcano effect on diurnal temperature range (DTR).

    http://australianclimatemadness.com/2011/09/12/another-paper-suggests-cosmic-ray-influence-on-clouds/

    Don’t know quite what it would mean if there was or wasn’t one, but it could be interesting…..

    —–

    Yhanks for replying to my previous query. I want to do a bit of checking on what you said about RCS but haven’t been able to put the time into it yet (I too have been and am on holiday in the UK).

  309. Willis Eschenbach says:
    September 24, 2013 at 8:58 am
    “Unfortunately, as a result of that error, your entire argument collapses.”

    You’re right. As I thought about it again there is in fact no need to postulate much lower water emissivity than 1 (Trenberth even uses emissivity value 0.9907) – the energy for the surface evaporation can easily be provided both by the solar irradiation and atmospheric mid-IR radiation. So the Trenberth surface radiation value can be more or less correct after all.
    I appologize for the confusion, some wrong ideas fortunately don’t live long.
    But this anyway somehow doesn’t seem to collapse the argument that the water cycle cools Earth surface by taking the latent and specific heat from it and transporting it up to the atmosphere and not much by cooling the surface by subsequent rain, nor it seems to collapse the argument that Earth climate system doesn’t seem to have effective surface temperature “homeostasis” or “thermostat” mechanisms able to counter the effects of solar and GCM activity changes – which seem being partially dependent on the solar activity. It even very much looks like the surface temperature anomalies are driven by this changes both on short and longtime scales, because the amount of absorbable incoming solar energy reaching the surface changes with changes of this factors both directly and indirectly – affecting the low-cloudiness, changing albedo, factor, which can have even considerably higher magnitude than just changes of the TSI.
    But I’m really not sure what is exactly your “heretic” idea, maybe you mean only purely terrestrial forcing factors? So I appologize if it is a misunderstanding.
    In any case I fully agree that the effects of volcanic erruptions are likely overestimated and in my opinion they can even have a phase where the ejected particles cause dimming when still in the air, but after eventually settling down having opposite forcing effect by at least intermittently lowering the albedo of the surfaces where it settled down (especially at high albedo surfaces as snow or ice) – which can at least partially offset the previous dimming effect. (Just an instant idea)

  310. happycrow says:
    September 24, 2013 at 8:20 am

    Would still appreciate if somebody were willing to tell me why my earlier idea is wrong. I’m sure it’s got to be, since it’s so simple, but I’ve no clue why and am curious.

    Thanks in advance,

    Russ in Texas

    happycrow says:
    September 24, 2013 at 10:54 am

    Damn, you guys are cold. There a blog around here where they’ll actually talk to laymen?

    Russ, I did a search for your “earlier idea”. I find only two comments from “happycrow”, the two I show above …

    And far from being cold, new ideas get discussed here all the time.

    You got to post them first, though …

    Give it another shot.

    w.

  311. Willis some day give us your explanation or explanations for why the climate changes abruptly at times ,while at other times it is more stable and yet at other times it fips from a glacial regime to an inter glacial regime.

    You say it is not the sun, not co2, not volcanic activity, etc.etc. so what is it?

  312. Mark Wright: Which begs the question, apart from cooling from volcanoes, what else at the scale of ESNO events is causing it to be warmer and cooler?

    If La Nina causes an upwelling of colder water, a tropical governor will cause it to be warmed by additional solar. When El Nino hits this energy will be dumped to the atmosphere. Hence frequency and amplitude of El Nino/Nina cycle has the means to cause heat gain or loss to the climate system. Notwithstanding the tropical a governor action. In fact because of the latter.

  313. tumetuestumefaisdubien1 says:
    September 24, 2013 at 11:43 am

    Willis Eschenbach says:
    September 24, 2013 at 8:58 am

    “Unfortunately, as a result of that error, your entire argument collapses.”

    You’re right.

    Ladies and gentlemen, I give you the action of a true seeker and scientist. When you are wrong, say so, digest the implications and move on. Tume, if I might call you that, well done.

    As I thought about it again there is in fact no need to postulate much lower water emissivity than 1 (Trenberth even uses emissivity value 0.9907) – the energy for the surface evaporation can easily be provided both by the solar irradiation and atmospheric mid-IR radiation. So the Trenberth surface radiation value can be more or less correct after all.
    I appologize for the confusion, some wrong ideas fortunately don’t live long.

    That’s one reason that I enjoy writing for the web, any illusions I entertain have a short half-life in the hot glare of public discussion.

    But this anyway somehow doesn’t seem to collapse the argument that the water cycle cools Earth surface by taking the latent and specific heat from it and transporting it up to the atmosphere and not much by cooling the surface by subsequent rain, …

    Curiously, the water cycle of “evaporation, condensation, rain, and repeat” cools the earth in lots of ways. In addition to reflecting sunlight from their top surface as cumulus clouds do, and transporting heat from the surface directly to the upper troposphere where it radiates easily to space, thunderstorms cool the surface in a variety of other ways, particularly (but not exclusively) over the ocean.

    1. Wind driven evaporative cooling. Once the thunderstorm starts, it creates its own wind around the base. This self-generated wind increases evaporation in several ways, particularly over the ocean.
    . a) Evaporation rises linearly with wind speed. At a typical squall wind speed of 10 mps (20 knots), evaporation is about ten times higher than at “calm” conditions (conventionally taken as 1 mps).
    . b) The wind increases evaporation by creating spray and foam, and by blowing water off of trees and leaves. These greatly increase the evaporative surface area, because the total surface area of the millions of droplets is evaporating as well as the actual surface itself.
    . c) To a lesser extent, surface area is also increased by wind-created waves (a wavy surface has larger evaporative area than a flat surface).
    . d) Wind created waves in turn greatly increase turbulence in the boundary layer. This increases evaporation by mixing dry air down to the surface and moist air upwards.
    . e) As the spray rapidly warms to air temperature, which in the tropics is often warmer than ocean temperature, evaporation also rises above the sea surface evaporation rate.

    2. Wind driven albedo increase. The white spray, foam, spindrift, changing angles of incidence, and white breaking wave tops greatly increase the albedo of the sea surface. This reduces the energy absorbed by the ocean.

    3. Cold rain and cold wind. As the moist air rises inside the thunderstorm’s heat pipe, water condenses and falls. Since the water is originating from condensing or freezing temperatures aloft, it cools the lower atmosphere it falls through. It also cools the surface when it hits. In addition, the falling rain entrains a cold wind. It is cooled by the evaporation of the falling drops This cold wind blows radially outwards from the center of the falling rain, cooling the surrounding area.

    4. Increased reflective area. White fluffy cumulus clouds are not tall, so basically they only reflect from the tops. On the other hand, the vertical pipe of the thunderstorm reflects sunlight along its entire length. This means that thunderstorms shade an area of the ocean out of proportion to their footprint, particularly in the late afternoon.

    5. Modification of upper tropospheric ice crystal cloud amounts (Lindzen 2001, Spencer 2007). These clouds form from the tiny ice particles that come out of the smokestack of the thunderstorm heat engines. It appears that the regulation of these clouds has a large effect, as they are thought to warm (through IR absorption) more than they cool (through reflection).

    6. Enhanced nighttime radiation. Unlike long-lived stratus clouds, cumulus and cumulonimbus generally die out and vanish as the night cools, leading to the typically clear skies at dawn. This allows greatly increased nighttime surface radiative cooling to space.

    7. Delivery of dry air to the surface. The air being sucked from the surface and lifted to altitude is counterbalanced by a descending flow of replacement air emitted from the top of the thunderstorm. This descending air has had the majority of the water vapor stripped out of it inside the thunderstorm, so it is relatively dry. The dryer the air, the more moisture it can pick up for the next trip to the sky. This increases the evaporative cooling of the surface.

    In part because they utilize such a wide range of cooling methods, cumulus clouds and thunderstorms are extremely good at reducing the surface temperature of the earth. Together, they form the governing mechanism for the tropical temperature.

    nor it seems to collapse the argument that Earth climate system doesn’t seem to have effective surface temperature “homeostasis” or “thermostat” mechanisms able to counter the effects of solar and GCM activity changes – which seem being partially dependent on the solar activity. It even very much looks like the surface temperature anomalies are driven by this changes both on short and longtime scales, because the amount of absorbable incoming solar energy reaching the surface changes with changes of this factors both directly and indirectly – affecting the low-cloudiness, changing albedo, factor, which can have even considerably higher magnitude than just changes of the TSI.

    Pass. Not sure I understand that.

    But I’m really not sure what is exactly your “heretic” idea, maybe you mean only purely terrestrial forcing factors? So I appologize if it is a misunderstanding.

    The currently accepted paradigm for the evolution of the climate is that changes in temperature are a linear function of the changes in forcing. I say the earth’s temperature is not set by the forcing, but by interlocking thermostatic mechanisms. In other words, I say the earth has a thermostat.

    In any case I fully agree that the effects of volcanic erruptions are likely overestimated and in my opinion they can even have a phase where the ejected particles cause dimming when still in the air, but after eventually settling down having opposite forcing effect by at least intermittently lowering the albedo of the surfaces where it settled down (especially at high albedo surfaces as snow or ice) – which can at least partially offset the previous dimming effect. (Just an instant idea)

    Mmmm … interesting question about fallen particles. Most of the stuff that is injected into the stratosphere is not ash, but chemical compounds, especially the reflective sulfates. It is these which circle the globe fairly rapidly, stay airborne, and lower the incoming sunlight.

    Not sure what effect sulfates would have on the ice, however. And while the ash is darker than snow, it’s far from black like soot, take a look at the pictures of the ash-falls after big explosions.

    All the best,

    w.

  314. Willis: I repeated my calculations to correct for the fact that the oceans only cover 70% of the surface and added in the heat capacity of the air. WIth a 50 m mixed layer ocean and atmosphere, a -1 W/m2 forcing causes temperature to drop at an initial rate of 0.2 degC/yr, or about 0.01 degC per month.

    A no-feedbacks climate sensitive of 1.2 degC for 2XCO2 is sometimes described as a Planck feedback of -3.2 W/m2/degC*. Once the temperature has dropped 0.1 degC, the reduction in outgoing blackbody radiation will cancel 0.32 W/m2 of current (negative) volcanic forcing. Therefore any temperature drop quickly begins to diminish the net radiative forcing associated with volcanic aerosols (which also dissipate). However, the system is out of equilibrium for several years after an eruption. No lagged relationship between forcing and temperature is an equilibrium relationship.

    * If you believe in positive feedback and a climate sensitivity of 2.4 degC for 2XCO2, the “effectiveness of the Planck feedback” is reduced to -1.6 W/m2/degC, enough to still begin to counter the volcanic forcing, but half as quickly. The higher climate sensitivity allows larger temperature swings.

    Since you believe overall feedback is negative, a climate sensitivity of 0.6 degC for 2XCO2 would double the “effectiveness of the Planck feedback” to -6.4 W/m2/degC. An 0.1 degC temperature drop will negate 0.64 W/m2 of volcanic forcing and severely limit the amount of cooling.

    You don’t have to reject the conventional linear relationship between forcing and temperature change simply because you believe volcanos have less impact than natural variation on temperature or because you believe feedback is negative. The Planck feedback IS negative (-3.2 W/m2/degC) and it could be made more negative by clouds. The conventional explanation is that water vapor feedback cancels about half of the Planck feedback. If all of the Planck feedback were negated by positive feedbacks, one has a run-away greenhouse effect.

  315. Salvatore Del Prete says:
    September 24, 2013 at 11:53 am

    Willis you have the data(I presented) it is in black and white ,unless you don’t believe satellite temperature data..

    I don’t recall saying anything about the temperatures of the stratosphere or the troposphere. The discussion has all been about surface air temperature. Yes, injecting tonnes and tonnes of reflective compounds into the stratosphere changes the temperature of the stratosphere.

    But the odd part is, it doesn’t change the surface temperature of the earth very much at all. That’s the curiosity.

    I’ve shown elsewhere that a ten-minute shift in the average time of tropical cumulus onset is enough to totally counteract a doubling of CO2 (annual average 3.7 W/m2), and the volcanic forcing is only a tenth of that (annual average 0.3 W/m2).

    Volcanoes are important because we can see the effect of a short sharp pulse. It’s an ideal way to measure the effect of a known change in forcing … which for the surface turns out to be very little, and which integrates out to zero. I say it’s because of emergent phenomena which govern the temperature. When the tropical ocean is a bit cool, the clouds don’t form until a bit later … and the balance is maintained.

    In the case of volcanoes, when the surface is cold because the sun is a bit weak, the system just allows more sunshine in by delaying cloud formation. It happens automatically, since when it’s cold the clouds don’t form.

    Nifty system, huh? A variable sun-shade that moves earlier or later to keep the temperature constant.

    w.

  316. RC Saumarez, I wanted to extend an invitation to you. Here’s the puzzle. We have the HadCRUT4 temperature dataset, and the volcanic forcings dataset. For convenience, I’ve put a simple Excel worksheet containing the Sato data (global, NH, and SH) and the HadCRUT4 data (global, NH, and SH) here. IF we make the mainstream assumption that the change in temperature is a suitably lagged linear function of the change in forcing … then what numbers do you come up with for the “climate sensitivity” and the time constant “tau” of the lag?

    I guess what I’m asking is, you seem very dismissive of my method of attack on the puzzle … so here you are with a free hand. How would you numerically analyze the volcanic effect on the temperature? What methods would you use? How would you judge your success?

    All the best,

    w.

  317. Repost:

    Russ in Texas says:
    September 23, 2013 at 8:36 am
    Willis or somebody else with good science-fu… can somebody help the liberal arts major here?

    It seems to me (but sometimes my ideas accidentally violate the laws of thermodynamics), that part of the equilibrium might have something to do with axial tilt? No, I’m not trolling, hear me out for a second, and then illuminate/slap-me-around as needed:

    You get something like a volcano: it’s going to produce a lot of localized cooling. But then it becomes winter, and the local area gets dropped even further below what the volcano would be doing…end result being more heat circulation from the rest of the globe, bang, it’s evened out, resulting in, as measured globally the forcing mechanism is destroyed at a global energy cost that is distinct, but miniscule.

    I’m assuming the 305k ocean-temp limit combined with lots and lots of summer sunlight handles the other side of said equation, but… I was a history major. You wanna know why Hugh de Beaux kidnapped Mary of Siciliy in 1350, I’m your dude…. this stuff, not so much (but interested).

    What am I not getting here? Thanks in advance.

  318. Frank says:
    September 24, 2013 at 12:39 pm

    Willis wrote:

    Thanks, Frank. Actually, the difference between Fi and Fe has nothing to do with latitude. It’s all explained in the link to the “Efficacy” paper.

    The explanation I suggested came from the link under “Sato paper” in your post. That link points to here: http://data.giss.nasa.gov/modelforce/strataer/

    Please push the “Efficacy” button as requested, Frank. All questions are answered there.

    w.

  319. Willis,
    Thank you for your kind response. Have read the first installment. Very interesting. You indicate that the earth’s temp has only varied approximately +/- 3% over the last half billion years. What would your source be for this info? I found this:

    http://www.physics.rutgers.edu/ugrad/140/lectures/lecture_4.pdf

    Also, a thought , what is the difference between a self regulating system as you describe and simply a system in which all the feedbacks sum to negative (or zero)? And, yes I read your description of the “governor” in a previous response to another poster. The effect would be the same though would it not?

  320. happycrow:

    At September 24, 2013 at 10:54 am you say in total

    Damn, you guys are cold. There a blog around here where they’ll actually talk to laymen?

    OK. I will bite.
    If you state what you want explained then I will try to ‘translate’ (as e.g. I did above concerning the R&C Effect).

    That way you may get what you want and others can concentrate on their debate without need to to talk to anyone other than each other. Your desire and their debate are both important.

    Richard

  321. Greg Goodman: “Joe, the point is that the start of month, end of month thing only matters if you have tau=2*step as in your exaggerated example. Paul_K reported specific figures and the error is minimal in the context of this data and tau of about 6 or 7 steps.”

    Well, I guess it depends on what “matters” means. It’s true that even under my assumptions as to what the data mean (i.e., the F and T values are both averages over the interval) Mr. Eschenbach’s formula will produce a curve that fits the data well; for a five-time-step time constant it produced less than 6% error in the fit when I applied fifty steps’ worth of red noise, and it was only 8% off in its equilibrium-gain (“lambda”) estimate. So in the context of Mr. Eschenbach’s post it’s certainly close enough.

    Still, the time-constant estimate was off by 25%, giving an estimate of six and a quarter time steps, i.e., in the range Mr. Eschenbach saw, rather than the five it should have exhibited. To me that could justify perhaps adopting a different formula in the future, which is why I brought it to Mr. Eschenbach’s attention.

    I hasten to add that the approach I generally use also seems to be something wrong; although it gets closer (fit error 2%, lambda error 0.2%, time-constant error 5%), the time-constant estimate is consistently low. So I will welcome any suggestions for improvement.

  322. Willis,
    One more thought, even if the change in temp over the past 500mm years has only been +/- 3% (the source I quoted would indicate more) we are talking about a difference in reality of five to ten thousand feet of ice right here where I live compared to cattle grazing on the slopes. Irrespective of the mechanism the governor would have little practicle use in the real world of climate as it ain’t governing enough over the “short term” of 100,000 years or so to make much difference to the viability of the environment. I will invent a term, cyclically chaotic.

    Thanks again

  323. Willis said:

    “Yes, injecting tonnes and tonnes of reflective compounds into the stratosphere changes the temperature of the stratosphere.
    But the odd part is, it doesn’t change the surface temperature of the earth very much at all. That’s the curiosity. ”

    That is because changing the temperature of the stratosphere also changes the height of the tropopause beneath those stratospheric temperature changes. A warmer stratosphere pushes the tropopause down and a cooler stratosphere pulls it up. The products of volcanoes injected into the stratosphere absorb solar shortwave directly, warm the stratosphere temporarily and push the tropopause down regionally.

    Consequently the net gradient of tropopause height changes between equator and poles, the global air circulation then changes with net latitudinal shifting of jets and climate zones which then change global albedo and the amount of ToA insolation able to enter the oceans to fuel the system.

    Volcanoes have a local or regional effect on the stratosphere depending on where they erupt.
    The sun has an effect on the stratosphere preferentially towards the poles when solar activity changes.

    In both cases it is the alterations to the global air circulation that neutralise the thermal effects for the system as a whole.

    The same for changes in GHGs. They change the circulation but not the total system energy content (after a suitable adjustment period) and the effect on the circulation of our emissions is miniscule compared to the effects of solar and oceanic influences.

    One still need the baseline system energy content to be set by something that is constant over geological periods of time, Hence the significance of gravity and atmospheric mass. Anything other than those two elements is capable of varying too much for the long term stability that we observe.

  324. The thunderstorm based hypothesis of Willis is but a small fraction of the global thermostatic process. One needs to involve the entire global circulatory system from ToA to the bottom of the oceans.

  325. Russ in Texas:

    This is my promised attempt to answer your post at September 23, 2013 at 8:36 am

    http://wattsupwiththat.com/2013/09/22/the-eruption-over-the-ipcc-ar5/#comment-1425599

    You say:

    It seems to me (but sometimes my ideas accidentally violate the laws of thermodynamics), that part of the equilibrium might have something to do with axial tilt? No, I’m not trolling, hear me out for a second, and then illuminate/slap-me-around as needed:

    Axial tilt does have both hemispheric and a global temperature effects

    The seasons of each hemisphere are provided by the axial tilt.

    The Northern Hemisphere (NH) is covered by much more land than the Southern Hemisphere (SH). And land varies its temperature much more than ocean for a similar change to solar energy. Therefore, the NH has larger seasonal temperature variation than the SH. Summer in the NH coincides with winter in the SH, and the global temperature at any moment is the sum of NH and SH temperatures. Thus, axial tilt causes global temperature to vary. The global temperature rises by 3.8°C from January to June each year and falls by 3.8°C from June to January each year.

    Clearly, any mechanism which stabilises global temperature must act to adjust global temperature towards an equilibrium temperature. But axial tilt varies that equilibrium by 3.8°C during each year.

    You then say

    You get something like a volcano: it’s going to produce a lot of localized cooling. But then it becomes winter, and the local area gets dropped even further below what the volcano would be doing…end result being more heat circulation from the rest of the globe, bang, it’s evened out, resulting in, as measured globally the forcing mechanism is destroyed at a global energy cost that is distinct, but miniscule.

    That concurs with what Willis is arguing in his above article. He says (in my opinion, he cogently says) that the cooling effect of volcanos is much less than is often claimed. Your assumption that the volcanic cooling is localised is a possible explanation for why the global cooling from volcanism is observed to be small.

    You continue saying

    I’m assuming the 305k ocean-temp limit combined with lots and lots of summer sunlight handles the other side of said equation, but… I was a history major. You wanna know why Hugh de Beaux kidnapped Mary of Siciliy in 1350, I’m your dude…. this stuff, not so much (but interested).

    Firstly, the 305K limit exists to sea surface temperature in the tropics. It does not apply to all oceans and it does not apply to land.

    Secondly, I don’t understand what you mean by “the other side of said equation”. I am assuming you are suggesting that the 305K limit fixes maximum temperature and “lots of summer sunlight” forces global temperature up towards that limit. Now, I may have misunderstood you and if so then I apologise, but if you did mean what I have interpreted then that is very improbable. What would be the ‘stop’ which prevents global temperature rising above a maximum limit? It cannot be the R&C Effect because that only operates for sea surface temperature in the tropics. And the ‘stop’ varies the maximum temperature it imposes by 3.8K during each year. . It is much more likely that there is a constantly changing equilibrium, and effects such as the R&C Effect and the Eschenbach Effect would then be mechanisms forcing the system towards the equilibrium it has at any moment.

    Thirdly, please do NOT apologise for not knowing every subject. All subjects have their importance and your expertise in the kidnaping of Mary of Siciliy can be of interest to us as our attempts to understand climate can be of interest to you.

    I hope this answer is clear, adequate and what you wanted.

    Richard

  326. Hansen, J., Mki. Sato, R. Ruedy, “Fe, which is the product of the forcing and its efficacy. Fe requires calculation of the climate response and introduces greater model dependence”

    I don’t see any mention that this has anything to do with TCS/ECS.

    Just more fudge factors. Take the observed data then ‘correct’ it fit the model.

  327. Thank you, Richard, that was extensive and helps me quite a bit… including that a volcano’s effects being local is indeed a dangerous assumption. Does the fact that the northern hemisphere is mostly land have any notable effect on the heat-transfer in the atmosphere from region-to-region and day-side/night-side and winter/summer once we’re past surface temps? I can see fronts move around in my weather report, of course, but I’m not well versed on how air moves *after* it’s displaced vertically.

    Thanks again,

    Russ in Texas

  328. re Paul_K’s two points and Frank’s dimensional complaint.

    T2 = T1 + lambda (F2 – F1) (1 – exp(-1/tau)) + exp(-1/tau) (T1 – T0)
    it seems your regression fit is:
    T2 = T1 + lambda (F2 – F1) ; not the same thing.

    You pointed out in an earlier thread that for a constantly rising forcing, the response will settle to a delayed image of the input where delay is equal to tau, once the transient has settled. But here we are specifically choosing to study when a transient hits. The biggest ones we can find.

    The dimensions of lambda are the same as sensitivity since each term in the expression must have the same dimensions. That is not sufficient to grab any term with those dimensions and equate it with climate sensitivity. You also lose a factor of (1 – exp(-1/tau)) = 0.14

    I think trying to correlate all volcanoes is a good way to reduce spurious coincidental changes that lead to false attribution. However, that are some notable problems with the implementation still.

  329. The base relationship of all this is ;
    C dT/dt = F(t) – λ *T

    a multivariate regression may be one approach.

  330. tumetuestumefaisdubien1,

    Yes, I believe you and I are saying the same thing but using different words to describe what is happening at the air-water interface in an attempt to get others to see why this is so. And it very much has to do with the reflectivity and angles. Sorry for the pause but I was out of place for a while.

    I have in the past used the weighted cosine integral of the angles of infrared leaving the oceans surface, it integrates to be equal to 1/pi or 31.8% that is self-absorption due to the angles, weighted that is. That take 400 * (1-0.318) to be 272.8 W/m². But you also have to allow of waters actual emissivity of about 0.96. So 272.8 * 0.96 is 262 W/m² actually leaving the air-water interface and you and I end up at the same point. I tend to call it the effective emissivity though it is really the mixture of two separate effects wrapped into one.

    There are some here that just cannot see that e/m radiation is composed of waves, following the duality, and these waves can and do cancel if there is a factor of angles involved. If A sends radiation to B but B is sending the exact same amount of radiation to A and at the same frequency, it is a null operation due to the angles making them exactly oppose. The angle there is 90° and the cosine is zero. At different angles you have to weight the cosines and integrate to get the overall upward sum total passage of the infrared radiation. You can do the same inverting the angles and use sines.

    Think we are saying the same thing. Think this is the basis for what you are using as the reflectivity and the optical properties.

    What is important is that by using the vaue that both of us are coming up with you can actually get the T-K buget to balance with the proper window frequencies radiation taken into account.

  331. The base relationship of all this is ;
    C dT/dt = F(t) – λ *T

    the solution to that ODE is (as you correctly state) :
    T2 = T1 + lambda (F2 – F1) (1 – exp(-1/tau)) + exp(-dt/tau) (T1 – T0)
    eg. http://rankexploits.com/musings/2011/noisy-blue-ocean-blue-suede-shoes-and-agw-attribution/
    (for the data here dt=1 if tau is also in months)

    for brevity let’s say
    a=1-exp(-dt/tau)

    then
    (T1-T2)-(1-a)(T1 – T0)=a*lambda*(F2 – F1)
    that can be simply regressed for a number of values of a (derived from tau) and the lag regression plot produced.

    It can be seen that LHS is close to being the temperature acceleration RHS is rate of change of forcing. This should satisfy Frank and Paul_K’s criticisms and is a full solution under the assumption this this is all linear to start with ;)

    I think this also removed the Joe Born “start-end” thing too.

  332. Re: Greg Goodman
    September 24, 2013 at 5:34 pm
    “The base relationship of all this is ;
    C dT/dt = F(t) – λ *T

    a multivariate regression may be one approach.”

    Hi Greg,
    Well that is ONE CHOICE of base relationship and it is the one that Willis is using in his forward model of temperature. The more general base relationship is:-
    Change in TOA Net flux = F(t) – λ *T

    Forster and Gregory 2006 did estimate λ from a regression approach over various time periods, including a go at Pinatubo.

    http://journals.ametsoc.org/doi/pdf/10.1175/JCLI3611.1

  333. Willis: Thank you for your reply, which corrects my misunderstanding. Hansen et al don’t believe that the radiative forcing for all agents will produce the same temperature change. They have put a fixed amount of Pinatubo aerosols into their MODEL for a century and then calculated the equilibrium temperature change and the radiative forcing. The MODEL has a high climate sensitivity (that you believe is wrong) arises from feedbacks that are produced by the model. The correction factor is a fairly negligible 0.91 and you applied it to the observed forcing. So what?

    An analogy: If you look for a correlation between the force applied to an object and its velocity, you may find some correlation. It may improve by including a lag. However, some of the time the force may be negative while the velocity is still positive. At a moment when zero force is being applied, the velocity can have many different values. Consider a ball thrown up into the air. The force is constant, but the velocity changes with time. Does that mean there is no relationship between force and velocity? Of course not. You need to integrate the force or take the derivative of the velocity before looking for a linear relationship. That linear relationship can be used to calculate the mass of the object.

    Another analogy: You might look for a correlation between the power being applied to an object and its kinetic energy, but you won’t discover the correct relationship without integrating the applied power and taking the square root of the energy. Your analysis will work better if you include friction.

    Radiative forcing “is” energy per unit time while temperature “is” energy. By analogy, you need to integrate the radiative forcing or take the derivative of the temperature before looking for a linear relationship and using it to determine lamba. However, the response to a volcano is more complicated, because a change in temperature also produces a change in outgoing radiation, which modifies the forcing. And the relationship between temperature and energy depends on the heat capacity of the mixed layer.

    Hope this helps. Thanks again for your replies.

  334. @Willis Eschenbach.
    I would not attempt what you suggest through linear modelling, because all the data involved suggest that these are non-linear processes, I would think it would take several months to get my teeth into the problem properly. To do so would involve a considerably better model than you propose..

    Frankly I think you are behaving very badly.

    I have made substantive criticisms of your model that is based on values in the literature and on proper analysis. You tend to reply with abuse. When finally challenged you come up with some autocorrelation functions that are incompetently plotted and you say that you can’t see the significance of them.

    Let me remind you that you are setting yourself up as the expert on signal processing and numerics, not I. Assuming that the ACFs are calibrated in years, then I have two comments:
    a) I am suspicious about the bandwidth and I think some of the shape may be artefactual..
    b) It is immediately obvious that they cannot possibly be generated by the model you propose and if you had the expertise you claim, it should have been immediately obvious to you.

    You challenged me to grab some data and do some sums. As it happens, I have and this is publically available. Whether you agree with my mathematical approach or not or whether you agree with Luck et al, McIntyre, Tol or not it is quite clear from the data that your model do not conform to data.

    In short I have made the argument, refuted your model and you are incapabable of making any sort of response that one might expect from a genuine “scientist”..

    I have read your various posts, you come over as completely imcompetetent. You post on the periodic transform is a particularly delicious example and you will note that everyone on the thread who had a knowledge of the subject concluded that you were talking complete bollocks.

    I did not comment on your “science” of treadmill exercise. Let me remind you that I am an investigative physician with 30 years experience in cardiopulmonary field. Your explanation was lamentable – do you understand the highly non-linear pressure flow relations of the lung that is dependent on lund volume? Do you understand the highly non-linear pressure flow relationships in the pulmonary circulation that is highly cardiac output dependent? Do you understand how ventilation and cardiac output is matched? I could go on for hours but the fact that you describe your presentation as science is unimpressive.

    Frankly I think you are self agrandising and lack insight into your scientific limitations. While you may enjoy the admiration of the credulous you hardly enjoy the respect of the educated.. The intemperance of your responses and your failure to respond to valid criticism in an objective manner reinforces this view. I would suggest that you really learn some maths and do the work in learning about signal processing rather than producing naive calculations whose basis you do not understand.

  335. Frank,
    “If Willis was feeling nasty, he could accuse those who compiled the data that Paul_K reanalyzed of cherry-picking the record that shows Pinatubo having the biggest effect on temperature. If Paul_K had analyzed the surface temperature record, he certainly would have obtained a different ECS than he reported at The Blackboard. The confidence interval for any result should reflect the uncertainty in the data used to generate that result.”

    Frank,
    There are certainly some differences between the surface-measured and satellite temp datasets but the choice of dataset is less important than you might think. This is because the abstraction of parameter values is not typically dependent JUST on the peak change in temperature. If you track back through my original article (which you graciously referenced), you will see that it includes a little story about Roy Spencer using TLT data, and Troy Masters testing the result on surface-measured data. The Forster and Gregory 2006 approach was also based on surface data from Hadcrut and GISS. The results all seem to yield values in the same broad range, which includes the no-feedback Planck response. However small positive or small negative feedbacks cannot be excluded by the data.

  336. Paul_K
    Hi Greg,
    Well that is ONE CHOICE of base relationship and it is the one that Willis is using in his forward model of temperature. The more general base relationship is:-
    Change in TOA Net flux = F(t) – λ *T

    Forster and Gregory 2006 did estimate λ from a regression approach over various time periods, including a go at Pinatubo.

    http://journals.ametsoc.org/doi/pdf/10.1175/JCLI3611.1

    ===
    re “The base relationship of all this is ;”
    Thanks, I meant the base relationship of the linear model is… I was not meaning it was the only possible model.

    It does appear that this has been done many times and more rigorously in the literature, so this whole thread is probably a bit of a red-herring.

    Lindzen and Choi 2011 , Dessler 2010 etc. etc. It seems we can produce pretty much whatever result we would like to see !

    L&C studies 20S-20N , which is what I suggested above, and finds much lower sensitivity than other studies.

    One mistake they all seem to have in common is the applying linear regression to two variables with significant experimental error, which is invalid and systematically produces slope below the true value. Lower slope = higher CS.

    L&C2011:
    As there are many intervals, the final ∆Flux/∆SST is a regression slope for the plots (∆Flux, ∆SST) for a linear regression model.

    Dessler 2010:
    Fig. 2. (A) Scatter plot
    of monthly average val-
    ues of DRcloud versus DTs
    using CERES and ECMWF
    interim data. (B) Scatter
    plot of monthly averages
    of the same quantities
    from 100 years of a control
    run of the ECHAM/MPI-
    OM model. In all plots,
    the solid line is a linear
    least-squares fit and the
    dotted lines are the 2s
    confidence interval of
    the fit.

    This maybe at least one reason why many studies are producing higher CS, simple basic processing errors. Least squares only gives the ‘best estimate’ when there is a “controlled variable” whose error/uncertainty is negligible compared to that of the dependant variable.

    The number of PhDs that are apparently totally oblivious to the assumptions and rules that need to be examined before clicking the “fit trend” button is astounding.

  337. Russ in Texas:

    In your post at September 24, 2013 at 4:38 pm you ask me

    Does the fact that the northern hemisphere is mostly land have any notable effect on the heat-transfer in the atmosphere from region-to-region and day-side/night-side and winter/summer once we’re past surface temps?

    Again, I am not clear what you are really asking, so I will answer what I think you mean. Please get back to me if I have misunderstood.

    The Earth receives energy from the Sun and – averaged over time – must radiate the same amount of energy back to space (if the energy from space did not equal the energy to space then the Earth’s global temperature would change). ‘Hot’ air rises and contains GHGs which radiate heat. At high altitudes most of that heat will go to space. So, the greater heating of air by land than from the sea in Summer means the summer heat loss to space is greater for large land areas than for large ocean areas.

    ‘Hot’ air has lower pressure than ‘cold’ air. This difference induces winds and other weather. It causes heat to be moved around by winds. I think you may find a primer book on meteorology would be helpful to you. From your questions, I suspect it may interest you.

    I hope this answer is some help.

    Richard

  338. RC Saumarez:

    Your post at September 25, 2013 at 1:40 am is a waste of space in the thread.

    Your post is abusive, offensive and arrogant but it is devoid of any useful information.

    Please say if you have genuine criticisms of Willis’ work because many – including me – would like to know of them. But assertions that you would have done something else, your “suspicions”, and your unfounded insults of Willis only provide information about you.

    This may surprise you, but nobody is interested in you. Many are interested in Willis’ analysis.

    Richard

  339. “This may surprise you, but nobody is interested in you. Many are interested in Willis’ analysis.”
    Richard, this may surprise you but you are not authorised to speak for everyone reading and commenting here.

    I am very interested in RC Saumarez’s comments which I generally find technically interesting. As with all I read, I check whether is makes sense and ties in with my understanding.

    I find Willis has been unnecessarily off hand in a lot of his comments to RC, even though he may be correct on some points. A bit of cool and civility on all sides would be appreciated.

    Hopefully Willis will have time to ignore posters like Salvatore and time to address technical criticism raised by Paul_K, Frank and myself.

  340. Greg Goodman:

    At September 25, 2013 at 4:58 am you say to me

    I am very interested in RC Saumarez’s comments which I generally find technically interesting. As with all I read, I check whether is makes sense and ties in with my understanding.

    Thankyou for that. Clearly, in reading the post from RC Saumarez which I was replying (it is at September 25, 2013 at 1:40 am) I must have missed something. I have read it again, and I still fail to discern anything in that post which is “technically interesting”. I absolutely fail to see any specific criticism of the analysis by Willis.

    Perhaps you could explain what RC Saumarez said in that post among his abusive, offensive and arrogant remarks which you found to be “technically interesting” because that would be helpful?

    Thanking you in advance.

    Richard

  341. Richard,

    Thank you for the replies — if most of the heat over top of a land surface radiates into space prior to diffusing towards colder sections of the atmosphere, that would indeed invalidate my hypothesis completely. I’ll hunt down a primer and start cracking so I can follow this more competently.

    Thanks again for taking the time and effort, I appreciate it.

  342. This is a demonstation of overshoot involving two coupled boxes of infinite volume. Box 1 is filled at a constant rate of an arbitrary unit. Box 2 is filled from box 1 at an accelerating rate that is calculated as the previous rate plus a fraction of the difference between the volumes in boxes 1 and 2. The chart shows boxes 1 and 2 swapping overshoot/undershoot positions.

    https://docs.google.com/spreadsheet/ccc?key=0AiP3g3LokjjZdFN0d1pfWUNTMk5DVVUwcTNnTmEwdmc&usp=sharing

    I see this as a sort of “demonstration of concept” which could help explain “the pause”. Box 1 being the well mixed layer and box two the ocean depths. Due to inertia convection is at first slow but then accelerates caring the surface heat to the depths.

  343. Paul_K: Thanks for the reply. If one is playing Willis’s game of “spot the volcano”, it appears to be much easier to do so using the satellite records (some of which have been adjusted for El Nino) than the surface temperature record Willis is showing. Since you analyze several years worth of monthly data, it is possible that you would find that your parameters provide a reasonable good fit for all of the temperature records. However, a quick glance at the surface temperature record in Troy’s post (which appears to have been amplified two-fold “2*GISS) might create some doubt.

    Troy, Wigley and probably other have analyzed the surface temperature record, but you may have been the only one to apply a model that fits a model with all of the relevant heat energy fluxes to all of the observations. I hope you have tried to publish: your work appears to be unique in this regard. And Pinatubo is the largest and best observed perturbation of climate in history. If you plot your calculated temperature data vs the observed temperatures from several different sources, you might be able to narrow the range of ECS. If you plot your calculated temperature data vs the observed data from several different sources, you might be able to narrow the range of ECS

  344. Grrr. I just spent three hours writing an extended response on this thread only to get the dread “failed to upload” error from WordPress. Double Grrr.

    I will summarize. RCS’s blog post was excellent, but fails to properly recognize that e.g. the Laplace transform of the ACF is very likely “rich” and that Laplace transforms connect power laws to multiple exponential laws to arbitrary decay function laws that aren’t really either one. RCS also fails to note that Willis’ result of “much less effect than existing GCMs incorporate” for volcanic forcing is very likely to survive tweaking the specific assumptions made concerning autocorrelation, even if some part of it is self-fulfilling prophecy caused by the selected filtering. That much is already apparent from playing the “spot the volcano” game. I think he has successfully shown that there is a quantitative basis for rejecting the assertion that volcanoes have a large or persistent impact on the climate, and I personally heartily applaud the application of the fluctuation-dissipation theorem to climate science, which I think RCS would agree we need more of, not less.

    Willis does not, I think, claim that his results are unique and correct (which would be a ridiculous claim given that he’s looking at only one dimension of a highly multivariate system with lots of confounding climate “impulses” e.g. ENSO events that overlap with the volcanic events), only that they are reasonable and defensible and likely to NOT be particularly sensitive to his specific assumptions concerning the climate autocorrelation, although given the paucity of the data one cannot rule out the possibility that confounding impulse-driven variation may have lowered his estimates of the system response by adding similar-scale noise. It would be very interesting to do the same analysis on volcanoes AND ENSO at the same time (for example) although there may not be enough data on ENSO to support this over a long enough time frame, and still more interesting to include still more “impulse” forcing changes if they can be found in the data. Examination of autocorrelation of these impulses is surely a common goal of RCS and Willis, as that is HOW once can objectively determine what the structure of the ACF, even though the analysis is doubtless significantly hampered by the (lack of) data — at most a short time series of higher quality data and an intermediate time series of much lower quality data.

    Now, if we could all stop being pointlessly rude to one another, that would be lovely. Yes, RCS, you have some mad skills in statistics. I’m perfectly happy to acknowledge them, after reading your paper. Indeed, I share your interest in Hurst-Kolmogorov processes and think that in the end, it may be Koutsoyiannis who leads us out of the desert of ill-founded climate models and into the green oases of models that at least are capable of getting the various correlation times right, or as right as we can so far determine given the inadequate, sucky, indifferently manipulated data.

    In a sane world, one would have done the work you and Koutsoyiannis appear to be doing long BEFORE building the FIRST climate model, as without it one doesn’t even have an empirical handle on what reservoirs are actually important, what timescales matter. The GCM community is paying the price by having to postulate a huge reservoir for “missing heat” with very long time constants and so far moderately unbelievable mixing dynamics: “The Ocean”. This, in turn, will make ALL of their past computations wrong as well, as the Ocean wasn’t born yesterday and it didn’t “suddenly” decide to confound GCMs, they were wrong even before they were wrong if indeed it is an important and formerly neglected factor.

    I especially like your “reservoir” model. It will join the “coupled capacitor” model that might be used to model the multiple exponential alternative to the power law. Indeed, it would be fun to build a reservoir model not with smooth sloping sides but with smooth sloping sides with e.g. holes drilled through them, with multiple reservoirs with seepage from one into another (and down to “ground”), with a stream that fills them through different pipes that themselves work better or worse as the water level(s) in the various reservoir(s) change(s). Then one might get SOME idea of just how complex things like the KT diagram really are, and how egregious the splitting up of insolation into the various channels under the assumption that the channel couplings don’t vary with the state of the system really is.

    Now, if we could all try very hard to be excellent to one another instead of rude and reactive, perhaps we could make better progress. I’m just saying…

    rgb

  345. RC Saumarez says:
    September 25, 2013 at 1:40 am

    @Willis Eschenbach.
    I would not attempt what you suggest through linear modelling, because all the data involved suggest that these are non-linear processes, I would think it would take several months to get my teeth into the problem properly. To do so would involve a considerably better model than you propose.

    Frankly I think you are behaving very badly.

    [followed by much abuse]

    RC, I’ve very politely invited you to do the analysis your way. You’ve declined … no surprise there. Instead, you say that given two simple datasets and being asked to determine how much one of them affects the other “it would take several months” just to get your teeth into the problem … OK, in that case I’m glad I didn’t hire you to do the job.

    But for a man who appears to have no idea how to actually solve the problem, you are surprisingly willing to take random meaningless potshots at the two different ways that I’ve gone about it. I used standard linear regression, and I’ve used a lagged exponential model.

    You responded, for example, that my model did not properly account for autocorrelation. I went to the trouble of posting the ACF of the results and my emulation … they are only slightly different. In response you’ve said:

    I can’t comment on your ACFs becauase the time axes has no units.

    Right … we’re looking at monthly data, we have been from the start, and now you don’t answer questions because I didn’t say the units are months. Do I have to say “Mother may I” as in the kid’s game as well? RC, I’m not going to hold your hand and blow in your ear. If you want to pass on answering and that’s your ludicrous excuse, I’ll take it that you are unwilling to admit I’m right, that there is very little difference between the ACFs of the model results and my emulation of those results.

    In any case, you’re positive both of those methods are the wrong way to go about it (despite the fact that they give very, very similar answers), but you don’t know the right way, and it would take you a couple of months to get your teeth into the problem … got it.

    Please come back when you’ve found your teeth, put them in, and gotten those choppers sunk into the problem. At that point I’m sure your comments will make more sense.

    Until then? Well, I suppose you could continue to spew abuse at me if that’s what floats your boat. I have to warn you that hurling vitriol doesn’t do much good for your reputation as a brilliant PhD signal analyst … but I suppose if it takes you two months to compare two simple signals, your reputation isn’t doing all that well anyhow.

    RC, I asked you how you would do it because from the way you’ve been going on here, I actually thought you knew a better way, and I’m always willing to learn.

    However, it seems you don’t know a better way.

    When you find one, I’m still more than happy to hear it. My suggestion would be to concentrate on the puzzle, and let the ad hominems slide. Yeah, I’m a jerk, and I don’t take insults well, and I don’t like being patronized, and I don’t suffer fools gladly … so what?

    Me being a jerk doesn’t help your position at all. You’ve made lots of claims that you knew much, much better than me how to go about solving the puzzle … and despite all your claims, when I say “OK, show us your better way”, you say it will take you two months to just start chewing on the problem.

    And fair enough, I may not see the true complexity, it may take that long, I’m not a PhD signal analyst … so when you’ve not only gotten your teeth into the problem, but have chewed it up and digested it, come back and reveal all. That should at least give us some peace until the end of November, at which point you can truly put me in my place with the elegance of your solution.

    w.

  346. One more comment: Indeed, the interesting thing about the “sloping side” reservoir model is that one can actually map it, I think, into things like the SBE by selecting the right shape for the reservoir sides. This still won’t give you nonlinear coupling and will only crudely model feedback — where increasing CO_2 might be equivalent to increasing the thickness of one PART of the side(s) of the reservoir, which might actually effectively decrease the thickness of other sides as increased flow metaphorically forces resistive particles apart or it might equally metaphorically increase drag — positive or negative feedback ASIDE from the variation in the height of water in the reservoir.

    I’ll have to think about this some more. In the end, the actual model is oversimplified and wrong no matter what, as the actual dynamics are going to be closer to the transport model with multiple local “reservoirs” all with their own distinct time constant(s) you allude to in your blog post and that correspond crudely to GCMs, but to the extent that one wishes to describe some sort of “average” flow through an open system with crude linearizations as a toy model or metaphor it is still useful.

    rgb

  347. AJ: “Due to inertia convection is at first slow but then accelerates caring the surface heat to the depths.”

    I have no basis for commenting on how plausible a significant inertial component is, but in the absence of comments from any of this site’s heavy hitters I’ll put my two cents’ worth in: to my untutored eye the fact that your (third-order) system includes no dissipative elements somewhat detracts from its persuasiveness as a proof of concept. What would help is some basis for believing that the inertia involved is large in comparison with dissipation. (I’m perfectly willing to be convinced that it is, but it doesn’t as an initial proposition commend itself to my intuition.)

    Just in case you find it relevant, it seems to me that a somewhat separate question is whether lower depths can be increasing in temperature while the surface’s temperature drops. I would say such a proposition ought not to be controversial, even if you don’t assume much inertia. If you reproduce the “tautochrone” shown here: http://wattsupwiththat.com/2012/06/18/time-lags-in-the-climate-system/ for different times of the day, you’ll be able to infer from the different depths’ phase differences that such a result can follow from diffusion alone.

  348. Robert Brown says:
    September 25, 2013 at 9:45 am

    Grrr. I just spent three hours writing an extended response on this thread only to get the dread “failed to upload” error from WordPress. Double Grrr.

    It seems to be connected to the “Copy” key, that if I don’t copy it before hitting the “POST COMMENT” button the odds of it failing to upload increase dramatically … need a statistical analysis of that to be sure.

    I will summarize. RCS’s blog post was excellent, but fails to properly recognize that e.g. the Laplace transform of the ACF is very likely “rich” and that Laplace transforms connect power laws to multiple exponential laws to arbitrary decay function laws that aren’t really either one. RCS also fails to note that Willis’ result of “much less effect than existing GCMs incorporate” for volcanic forcing is very likely to survive tweaking the specific assumptions made concerning autocorrelation, even if some part of it is self-fulfilling prophecy caused by the selected filtering. That much is already apparent from playing the “spot the volcano” game. I think he has successfully shown that there is a quantitative basis for rejecting the assertion that volcanoes have a large or persistent impact on the climate, and I personally heartily applaud the application of the fluctuation-dissipation theorem to climate science, which I think RCS would agree we need more of, not less.

    The response of the climate models to volcanoes has long been held up as evidence that they do well. Unfortunately, their response matches the volcanoes quite poorly. In addition, some models appear to treat volcanic forcing differently (different sensitivity) than other forcings. I believe that the technical term for that is “cheating”.

    Willis does not, I think, claim that his results are unique and correct (which would be a ridiculous claim given that he’s looking at only one dimension of a highly multivariate system with lots of confounding climate “impulses” e.g. ENSO events that overlap with the volcanic events), only that they are reasonable and defensible and likely to NOT be particularly sensitive to his specific assumptions concerning the climate autocorrelation, although given the paucity of the data one cannot rule out the possibility that confounding impulse-driven variation may have lowered his estimates of the system response by adding similar-scale noise. It would be very interesting to do the same analysis on volcanoes AND ENSO at the same time (for example) although there may not be enough data on ENSO to support this over a long enough time frame, and still more interesting to include still more “impulse” forcing changes if they can be found in the data. Examination of autocorrelation of these impulses is surely a common goal of RCS and Willis, as that is HOW once can objectively determine what the structure of the ACF, even though the analysis is doubtless significantly hampered by the (lack of) data — at most a short time series of higher quality data and an intermediate time series of much lower quality data.

    I would disagree about the volcano / ENSO joint analysis, unless the analysis recognizes that one is a forcing (volcanoes) and the other is a response. The El Nino/La Nina alteration are the instroke and outstroke of a pump which pumps excess heat to the poles. So when a volcano (or anything else) cuts down the incoming heat, the pump should operate less frequently.

    Now, if we could all stop being pointlessly rude to one another, that would be lovely. Yes, RCS, you have some mad skills in statistics. I’m perfectly happy to acknowledge them, after reading your paper. Indeed, I share your interest in Hurst-Kolmogorov processes and think that in the end, it may be Koutsoyiannis who leads us out of the desert of ill-founded climate models and into the green oases of models that at least are capable of getting the various correlation times right, or as right as we can so far determine given the inadequate, sucky, indifferently manipulated data.

    In a sane world, one would have done the work you and Koutsoyiannis appear to be doing long BEFORE building the FIRST climate model, as without it one doesn’t even have an empirical handle on what reservoirs are actually important, what timescales matter. The GCM community is paying the price by having to postulate a huge reservoir for “missing heat” with very long time constants and so far moderately unbelievable mixing dynamics: “The Ocean”. This, in turn, will make ALL of their past computations wrong as well, as the Ocean wasn’t born yesterday and it didn’t “suddenly” decide to confound GCMs, they were wrong even before they were wrong if indeed it is an important and formerly neglected factor.

    Agreed on all points.

    I especially like your “reservoir” model. It will join the “coupled capacitor” model that might be used to model the multiple exponential alternative to the power law. Indeed, it would be fun to build a reservoir model not with smooth sloping sides but with smooth sloping sides with e.g. holes drilled through them, with multiple reservoirs with seepage from one into another (and down to “ground”), with a stream that fills them through different pipes that themselves work better or worse as the water level(s) in the various reservoir(s) change(s). Then one might get SOME idea of just how complex things like the KT diagram really are, and how egregious the splitting up of insolation into the various channels under the assumption that the channel couplings don’t vary with the state of the system really is.

    While the “reservoir” model is indeed interesting, as you point out it does not address the fact that flow systems like the climate follow the Constructal Law, which says that the flow system will constantly evolve to increase the interaction between the various flows.

    As a result, your example with multiple interactions between the reservoirs that are constantly changing is much closer to the actual reality. See Bejan’s papers here, here, and here on the subject.

    And yes, I know I shouldn’t be rude to RC, he just rubs me the wrong way, coming in to tell us very patronizingly that the master has arrived and we should all listen to him and take his words as gospel. Nullius in verba is my motto. But you’re right, I’ll cut him some slack when he comes back in November to tell us how he’d solve the puzzle.

    w.

  349. Willis you are correct to point out that the contribution from volcanic activity from the period 2008-2010 or so had no impact on the temperature trend as the IPCC so wrongly points out. I am with you on that.

  350. Joe Born says:
    September 25, 2013 at 10:12 am
    “… your (third-order) system includes no dissipative elements …”
    #########
    Yep… it’s an undamped model. Hit it with a pulse and it will ring forever. I find this distracting as well. Any help would be appreciated. Thanks for the comment.

  351. Joe Born says:
    September 25, 2013 at 10:12 am
    “What would help is some basis for believing that the inertia involved is large in comparison with dissipation.”
    ##############
    I don’t have much to support the proposition that inertia is large in comparison to dissipation, but maybe a couple of plots in the link below might support this. They show temperatures and sea-levels both with a ~60yr cycle, with sea-level trends lagging temperature trends by ~20yrs.

    https://sites.google.com/site/climateadj/multiscale-trend-analysis—hadcrut4

    In a traditional one-box model, the response cycle can’t lag the forcing cycle by 1/4 cycle and retain an amplitude. If I modify the exponential decay model to introduce a relative acceleration, say exp(-(t^1.5))/tau), then I can push the lag out past 1/4 cycle and still retain an amplitude. Presumably this relative acceleration indicates inertia. Of course I’m assuming that sea-level is a proxy for ocean heat content as well.

    So I have two models, one that overshoots but doesn’t dissipate and one that indicates inertia but doesn’t overshoot. Now if only someone could come up with a model that overshoots, dissipates, and creates a response lag of 1/3 cycle. There would still be the matter of only be ~2 cycles of observational data though and the first one is sorta iffy.

  352. AJ: “Any help would be appreciated.”

    I can’t think of what I could contribute; as I said, this does not seem to be a very promising avenue (and I didn’t really follow all your comments). But, as I also said, I have no expertise in this area, so it may turn out to be more fruitful than I think.

  353. Joe Born says:
    September 25, 2013 at 10:12 am
    “If you reproduce the “tautochrone” shown here…”
    ##############
    Well, while I’m playing show and tell, here’s my first attempt using R for an “analysis”. It was simply doing yearly sine fitting on the ARGO data.

    https://sites.google.com/site/climateadj/argo-sine-fitting

    I started off by looking at 45S, determining the amplitude, phase shift (relative to relative equinox), and mean. Later I expanded it to produce image plots for 55S to 55N. At first I thought the phase might tell me something about uptake, but surmised that it probably wouldn’t (note that my phase plot is mislabeled “ARGO R2 Phase” ugghh..).

    One thing to note in this collection is the image plot of mean temperature. You’ll notice a relatively warm dome that goes from the tropical surface to the mid-latitudes at depth. I don’t know if this feature has a name, so’ll I’ll just call it the “warm front” (Ekman Pumping Zone, maybe?). When I plotted some temperature trends, the most positive trends were at the poleward edge of the warm front and the most negative trends were at the equatorial edge of this warm front. This indicates that the warm front in moving poleward (note only 6 years of data at the time). In an “overshoot” context, this could imply that the convection profile is changing, possibly accelerating the heat transport to the depths. If this is correct, then sometime in the future this warm front movement could either decelerate or possibly start moving equatorward. Here’s the trend plots I speak of:

    https://sites.google.com/site/climateadj/argo-analysis

    Anyway, that’s all I have. My guess is that there is a true overshoot/undershoot phenomenon that the model don’t capture.

  354. Joe Born says:
    September 25, 2013 at 3:12 pm
    “I have no expertise in this area…”
    #####
    Neither do I, so who knows? Once again, thanks for the comment.

  355. Willis Eschenbach says:
    “September 22, 2013 at 8:17 pm
    Chuck Nolan says:
    September 22, 2013 at 6:19 pm………………….This is why I don’t like what I call “pressureheads” on my threads. I’ve shown you can’t do it, but you keep showing up.”
    —————————————————————
    Sounds like Willis is channeling Mikey Mann.
    I thought this was about learning the answers I didn’t know Willis already had them all.
    I guess I’ll get off of “His” thread.
    Thanks for all the info Anthony.
    Willis can shout at someone else and I’ll “TAKE IT ELSEWHERE!”
    Save the name calling for your enemies.
    cn

  356. AJ says:
    September 25, 2013 at 4:13 pm

    Joe Born says:
    September 25, 2013 at 10:12 am

    “If you reproduce the “tautochrone” shown here…”

    ##############
    Well, while I’m playing show and tell, here’s my first attempt using R for an “analysis”. It was simply doing yearly sine fitting on the ARGO data.

    https://sites.google.com/site/climateadj/argo-sine-fitting

    I started off by looking at 45S, determining the amplitude, phase shift (relative to relative equinox), and mean.

    Man, if that’s your first attempt in R, I’m very, very impressed. Nice graphics, clean code. I am so stealing great chunks of that.

    I’ll have to give that greater study, as I’m not sure what all of that implies. But what a piece of work. I encourage you to write it up so everyone can understand it and submit it as a post to WUWT.

    Well done, that man!

    w.

  357. Willis,
    While I’m being encouraged, here’s a follow on. Due to noise, I noticed that sine fitting wasn’t very effective at picking up the power of the annual signal. So I switched to a variance method instead and compared my plots against the AR4 models:

    https://sites.google.com/site/climateadj/ocean_variance

    Most models got the top 100-200m right. Below the thermocline… not so good.

    Thanks for the encouragement, but I don’t really do blog authoring. I’d rather write up my own notes and post links in comments where appropriate. If you feel something is blog worthy, feel free to steal away. If you need clarification on something… you got my email address.

  358. Chuck Nolan says:
    September 25, 2013 at 4:46 pm (Edit)

    Willis Eschenbach says:
    “September 22, 2013 at 8:17 pm
    Chuck Nolan says:
    September 22, 2013 at 6:19 pm………………….

    This is why I don’t like what I call “pressureheads” on my threads. I’ve shown you can’t do it, but you keep showing up.”

    —————————————————————
    Sounds like Willis is channeling Mikey Mann.
    I thought this was about learning the answers I didn’t know Willis already had them all.
    I guess I’ll get off of “His” thread.
    Thanks for all the info Anthony.
    Willis can shout at someone else and I’ll “TAKE IT ELSEWHERE!”
    Save the name calling for your enemies.
    cn

    Chuck, I’ve given a very clear scientific proof why pressure can’t warm the surface. Here’s the elevator speech. Imagine a blackbody planet evenly heated by a thousand suns spaced equally around the sky. The amount it radiates is equal to the amount it absorbs from the suns.

    Now add an argon atmosphere, which is transparent to infrared. The pressureheads claim that the pressure of that atmosphere will warm the surface.

    But if it does so, the surface will radiate more … and at that point it’s radiating more than it is absorbing, and that’s not possible on a continual basis.

    That’s the proof that no such mechanism based on pressure can heat the surface. It would violate the law of conservation of energy, constantly emitting more than it is receiving.

    Now, if you want to dispute that proof, fine. No one has done so successfully to date. But I’m not interested in claims that pressure warms the surface, I’ve proven, not alleged but proven, that it’s not possible. It’s actually one of my cleverest scientific proofs.

    Do I have all the answers? By no means, I have more questions than answers. But one of the things I do have the answer to are claims that pressure can warm the surface … it can’t.

    Finally, is this “my” thread? Well, I wrote it … and I won’t sit by and see it hijacked by pressureheads. Got no time for that …

    So you’re welcome to hang around and join the discussion … but not to make impossible claims. This site is dedicated to science.

    All the best,

    w.

  359. AJ, thanks for linking the graphs and code, very interesting.

    I did not understand what graph you were referring to with the warm bump/ Ekman comment though.

    One thing that seems notable in the temp/depth plots that you don’t comment on is that below 200m the cycle appears to 6m not 12m , or at least has a strong 6m component. Now the 6m variation is equatorial/tropical insolation so it seems that mixed layer at 45S had the annual cycle we would expect but deeper is more influenced by tropical energy input. That is a valuable result and surely informs us of something about the system.

  360. AK ” There is a lot of variation in the individual years and an interesting oscillation between a short phase one year and a long phase the next.”

    A one year anti-correlation has been noted in arctic ice, though I don’t have a reference to hand.

    There is a triplet of frequencies centred on 2y in the power spectrum of ice area:

    http://climategrog.wordpress.com/?attachment_id=438

    Similar pattern in UAH TLT for NH, though not the same in tropics or SH, there is a strong and common 2.48y peak that is probably what it called QBO (quasi-biennial osc).

    http://climategrog.wordpress.com/?attachment_id=407

    2.44y is the strongest peak in the trade wind spectrum.

    http://climategrog.wordpress.com/?attachment_id=283

    Over the short ARGO sample of 6 years , it could be the alternation you noted is due to 2.4 rather than 2y.

    In view of the other global indications I would suggest what you noted is more like real than coin flip.

  361. AK “Here’s the Adjusted R^2 verification statistic for the linear regression. For the individual years this is ugly, with some years getting into negative territory at a fairly shallow depth. ”

    With phase lag getting close to a year in some cases this may not be surprising.

    Would this behaviour suggest a circa 2y variation below the thermocline that is alternately adding and then opposing the surface signal? In even numbered years it opposes, leading to neg corr and long phase lag.

  362. x1<-cos(x)
    x2<-sin(x)

    m<-lm(y ~ x1 + x2)

    I can never get my head around the arcane way model functions are specified in R , could someone translate please?

    AK, could you explain what “newtonian view” refers to in this context? Thanks.

  363. Tried to run the R code but even after a few simple syntax changes (linux netcdf package has differences) and over a hour of data download it now shits out with an incredibly useless error message that tells me nothing about where look. (remind me why I gave up wasting time on R years ago).

    Error in double(totvarsize) : vector size cannot be NA/NaN

    Not that that is any reflection of AK, I just don’t have the time piss debugging that kind of crap environment.

    The reason I tried to run the code was that AK’s graphs all stop in 2010 yet the data is “near real time” and should have complete years to 2012.

    AK, is there a reason for this or where those graphs you linked created a couple of years ago?

    Thanks.

  364. Greg Goodman says:
    September 25, 2013 at 10:59 pm

    Re: 6mo cycle

    You might be misreading the graph… I plotted 24 months by doubling up on the 12 month period. I find it easier to see patterns this way.

    Probably won’t have time today to answer other questions… maybe tonight.

  365. Actually Greg, I see what your talking about now. The wedge in the annual signal. Not sure what that is. There was only 6yrs of data available at the time and there’s two more now, so maybe it’s disappeared.

  366. Greg Goodman says:
    September 26, 2013 at 12:06 am
    “…could someone translate please?”
    ######
    What I’m doing here is fitting a sine curve using a linear model (lm). Here’s a good explanation:

    https://stat.ethz.ch/pipermail/r-help/2008-May/162879.html

    The “Newtonian View” is using simple exponential decay “exp(-t/tau)” (i.e. a one box model).

    When I first started looking at this, I ran into Newton’s Model of Cooling, so that explains the terminology.

  367. Greg Goodman says:
    September 26, 2013 at 2:22 am

    Re: Error

    If you send your code to climateadj at gmail etc, maybe I’ll have a look tonight. I think I downloaded the ARGO data earlier this year, so that should save me a step.

  368. AK, thanks very much for the clarifications. I’d guessed ‘Newton’ was something to do with a linear feedback model but wondered whether it implied more than that.

    It would be very interesting to see what the graphs look like now, with the extra two years of data. Whether what I read as 6mo and you called a notch disappears or becomes clearer.

    In your 2008 plots, between 600m and 900m, there’s more 6mo than 12mo . How about odd numbered years? There’s definitely some interesting information there.

    The linux library (called ncdf) does not recognise the collapse=FALSE so I had to remove it, I don’t know whether that will mess things up but the error message was so vague and did not even relate to code in your script, I’d have to start putting ‘print’ statements everywhere and digging into the library code to track it down and I don’t have time to mess around like that.

    re pipermail, good explanation , nice trick. hwvr, I can’t see this as being exactly what your code does:

    x<-(2*pi*(((0:11+.5)/12)+(phaseoffsetdays/365))) # cf pipermail (x+b)
    x1<-cos(x) # cf pipermail sin(b)*cos(x)
    x2<-sin(x) # cf pipermail cos(b)*sin(x)

    Maybe I’m being thick.

  369. AJ says: Re: Error
    If you send your code to climateadj at gmail etc, maybe I’ll have a look tonight. I think I downloaded the ARGO data earlier this year, so that should save me a step.

    It’s not my code, it’s what you provided. I just had to make a few syntax changes :

    library (ncdf)
    nc2<-open.ncdf("argo_2005-2012_grd.nc")

    # var.get becomes get.var:
    long<-get.var.ncdf(nc,longvar)

    I suspect the problem may this edit. Incompatible option
    # ptsamp<-var.get.nc(nc,tempvar,start,count,collapse=FALSE) – k
    ptsamp source (“AJ.R”)
    Error in double(totvarsize) : vector size cannot be NA/NaN

    Maybe they’ve changed a variable name. If your code still works on the data , I guess it’s an incompatibility in the linux ncdf lib for R.

    thanks.

  370. duh, WordPress dumped my formatting tags…

    library (ncdf)
    nc2<-open.ncdf("argo_2005-2012_grd.nc")
    
    # var.get becomes get.var , all occurances changed
    long<-get.var.ncdf(nc,longvar)
    
    I suspect the problem may be this edit. Incompatible option removed:
    # ptsamp<-var.get.nc(nc,tempvar,start,count,collapse=FALSE) - k
     ptsamp <-get.var.ncdf(nc,tempvar,start,count) - k
    

    when I try to run it , it seems to have an empty vector despite the massive files.

    ptsamp source (“AJ.R”)
    Error in double(totvarsize) : vector size cannot be NA/NaN
    
    19838290 -rw-r--r--  1 user users 4031083428 Jan 23  2013 argo_2005-2012_grd.nc
    19838292 -rw-r--r--  1 user users  503889588 Jan 23  2013 argo_CLIM_grd.nc
    
    nc1<-open.ncdf("argo_CLIM_grd.nc")
    nc2<-open.ncdf("argo_2005-2012_grd.nc")
    
    

    Maybe they’ve changed a variable name. If your code still works on the data , I guess it’s an incompatibility in the linux ncdf lib for R.

    thanks.

  371. jeez, whatever happened to the new preview function Anthony got with his $550 upgrade?!

    The R command was obviously just :
    source (“AJ.R”)

    • @Greg Goodman

      I sent the upgrade back to the factory for retooling, there were several problems with it, and I discovered that it didn’t merit the cost.

      I’m waiting to see if wordpress.com can solve the issues before reinstating it.

  372. Greg:
    The source at the bottom is actually four separate programs. I’ve separated them using a long string of X’s such as:

    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

    My apologies on not making this clear. I should have done a better job. I really didn’t think anyone would ever look at the code.

  373. “I discovered that it didn’t merit the cost. ”
    I did not comment at the time since it was your choice and assumed you had your reasons but it seems a lot to pay for very little.

  374. AJ: “Anyway, that’s all I have.”

    You couldn’t do any better than that? :-)

    Seriously, great work; it makes the data talk to me–which is no mean accomplishment.

    I see what you mean about about the trend plots’ suggesting overshoot. To me they look like a still shot of wave propagation both in the north-south direction and (slightly) from the surface downward. Or (sorry to perseverate), rather than a wave-equation-type phenomenon, it could (I think) be a diffusion phenomenon analogous to what Mr. Eschenbach’s tautochrone plot depicted. (I have no idea what the physical basis could be; I suspect that thermal conductivity is way too low to implicate conduction, which is what explains the tautochrones.)

    In an attempt to make a quick test of diffusion-type equations I tried to find the code I used to generate tautochrones of my own when Mr. Eschenbach raised the subject, but it appears I didn’t keep it. A substitute you may want to try is a lumped-parameter approximation to that (distributed-parameter) implementation: a multi-box system in which dy_i / dt = r * (y_{i-1} – 2*y_i + y_{i+1}), with dy_1/dt = -(a + r) * y_1 + r * y_2 + x and dy_I/dt = r * (y_{I-1} – y_I), the i’s representing different latitudes. That is, you could imagine the system’s being driven only at the equator and then watch the results propagate poleward. I’m guessing you could get to I = 9 or 10 before numeric issues would make your computer balk or the solution would take the computer too much time.

    Again, great work.

  375. Greg:
    Yes it’s probably collapsing the latitudinal dimension as the code is only selecting one latitude (45S). My code can select multiple latitudes and expects a 4D array (lon,lat,lev,time).

    You can probably get around this by inserting something like this after your get.var:

    if (length(dim(ptsamp)) == 3){
    dimlen = dim(ptsamp)
    ptsamp = array(ptsamp,,dim=c(dimlen[1],1,dimlen[2],dimlen[3]))
    }

  376. Me: “I’m guessing you could get to I = 9 or 10 before numeric issues would make your computer balk or the solution would take the computer too much time.”

    Yeah, big talker. It just took my machine over seven minutes for I = 9.

  377. Greg:

    Re: pipermail explanation of linear sine fitting

    I don’t think your thick. I remember it making my head hurt to translate what was described into workable code. I also had to make modifications for such things as hemispheres/equinoxes, lags extending out past 1/2 cycle, etc.

    I’d walk you through it, but it’s been a couple of years since I wrote this and I’m afraid it would hurt my head again.

    Seems to work though.

  378. Greg:
    Actually the dimensions should be in the variable “count”, so this should work to preserve the 4D array:
    if (length(dim(ptsamp)) != length(count)){
    ptsamp = array(ptsamp,,dim=count)
    }

  379. Thanks for the suggestions AJ. but it’s not even getting as far as running the first line. of code.

    Crappy interpreted languages that cannot even give an intelligible error message are more that my time’s worth. Jeezuz, I have the impression to be working with some early-80’s ROM based BASIC. Worse, at least they used to give a line number when they hit a problem.

    It’s more akin to programming in a µP directly in HEX.

    Let us know if you get a chance to update your plots with the more recent data. What you got in 2010 looked very interesting.

  380.  
    # start<-c(NA,startlat,startdepth,sm)
     start<-c(1,startlat,startdepth,sm)
    # count<-c(NA,countlat,countdepth,cm)
     count<-c(-1,countlat,countdepth,cm)
    
    

    Seems like the linux ncdf library implementation does not know “collapse” and won’t accept the NA in the argument list. ( -1 is the flag to read all and it req a specific valid index in start).

    also needed to replace win.graph
    dev.new(); #win.graph()

    but now I have your beautiful graphs . Thanks much for supplying code. ;)

    cy<-08 # number of years to get 2011,2012

    Now it gets interesting. Something notable happens between 300m and 400m in the last two years.

    In the R2 plot 2011 starts out like an even year but ramps back up from 300m and ends like the other odd years. There is an odd jump in the phase plot at this point.

    2012 crosses to neg R2 at 300m and stays there , basically similar to the other even years. The phase plot seems to wrap around 365 but otherwise looks orderly.

  381. Ah, seems you’ve done at the same time, fine. Any thought on what they show or a mod to take care of phase wrapping around?

  382. AJ. your comment on the 2008 plot in the initial graph set:
    “For comparison purposes, here’s a plot of the fitted curve to the actual signal for year 2008 data. Does this model underestimate the amplitude once the signal starts to become lost?”

    I’d say not. I think it appears to correctly shows the reduced 12mo amplitude but does not model the 6mo component.

    I’d suggest the 6mo is a rectified 12mo not a 6mo sine. That’s why it appeared as a “notch”.

    What I find rather surprising at first view is that this signal is so significant at depth in the extra-tropics , when the initial jump-to-conclusion would be to associate this with tropical insolation cycle.

    Now since this is less marked in the multi-year means, maybe it is not exactly 6mo and/or there’s a phase drift.

  383. Digging around in the depths is proving very interesting. Using the segment AJ has called program 2 in his revised code, linked above, confirms that the even numbered years are more interesting.

    It all seems to be happening below 300m with a clear pattern emerging at 1000m

    2008 I’ve already commented has a strong 6mo signal, the second harmonic. Also 2004 has strong 2nd and 3rd, and 2010 has strong 3rd. An this is just looking at one lat band at 42.5 S.

    It seems the depths are not as deep and still as we are lead to believe (or assume).

  384. Willis said:

    “Now add an argon atmosphere, which is transparent to infrared. The pressureheads claim that the pressure of that atmosphere will warm the surface.

    But if it does so, the surface will radiate more … and at that point it’s radiating more than it is absorbing, and that’s not possible on a continual basis.

    That’s the proof that no such mechanism based on pressure can heat the surface. It would violate the law of conservation of energy, constantly emitting more than it is receiving.”

    Consider pressure as a proxy for density which is a function of mass and gravity.
    The more dense the argon at the surface the more it will exchange energy with the surface by non radiative processes.

    The surface then has to radiate energy to space AND exchange energy with the argon.

    Thus the surface temperature can (indeed must) be higher without offending the laws of thermodynamics.

    It is true that what the surface receives from the argon over time matches what it passes back to the argon but nonetheless the surface temperature must be higher so that notwithstanding the argon/surface exchange it STILL has enough energy left over to radiate as much out at ToA as is received at ToA.

    Then consider that when the argon atmosphere expands or contracts the density at the surface decreases or increases and the amount of energy exchanged by non radiative processes will change too.

    That results in circulation changes so as not to upset ToA equilibrium.

    All composition variations can do is change the volume of the atmosphere which then adjusts the density at the surface which also adjusts the temperature at the surface with an equal and negative thermal response for zero net effect on surface temperature but instead a circulation adjustment.

  385. Greg Goodman says:
    September 26, 2013 at 9:12 pm

    Re: NA’s in longitudinal start/count fields

    Yep… That was a “feature” I corrected. I didn’t notice it because I was selecting all longitudes. At the time I may have been wondering what to do when the selection crosses the prime meridian, which still isn’t resolved. Should only matter if you’re looking specifically at the Southeast Atlantic, North Sea, etc.

    I was wondering if the win.graph function would work on linux. That was back in the “old days”. Nowadays I use the dev functions.

    As far as phase wrapping, my thought would be to check if point1 was in the 4th quarter and point2 was in the 1st, then put point2 in a fifth quarter. Same with point1 being in 5th quarter and point2 in 1st quarter. Kinda cludgy, but looks like it would work. There’s probably more elegant solutions using slopes.

    I think the amplitude is lost once the signal becomes noisy. I better metric would be to calculate the variance or stdev as that would give you the power across the full spectrum of frequencies:

    https://sites.google.com/site/climateadj/ocean_variance

    I wouldn’t be surprised if the 6mo tropical signal migrates poleward. I’d also bet Chladni patterns emerge.

    Good to see you find the digging interesting. Happy hunting.

  386. I’ve hacked AJ’s program2 to show the annual cycle at 1000m for each year.

    There is a clear 2y alternation between a mainly 12mo seasonal cycle and a more complex variation dominated by higher harmonics.

  387. BTW anyone explain what a “potential temperature” is ? I find lots of mentions of the term without an explanation. Straight temp is also there.

  388. Greg… now that I’ve thought about it for more than 2 seconds and have had my coffee, here is a way of handling the phase wrap. Note, I was measuring phase in days (assuming a 365 day year).

    If the difference between two years is outside of 365/2 days, add or subtract 365 days:

    lag = c(200, 250, 300, 350, 40, 90, 40, 350, 300)

    lagdiff = diff(lag)

    x = lagdiff (365/2)
    lagdiff[x] = lagdiff[x] – 365

    lagdiff = c(lag[1],lagdiff)

    lag = cumsum(lagdiff)

  389. Let’s try this again with .lt., .gt.

    lag = c(200, 250, 300, 350, 40, 90, 40, 350, 300)

    lagdiff = diff(lag)

    x = lagdiff .lt. -(365/2)
    lagdiff[x] = lagdiff[x] + 365
    x = lagdiff .gt. (365/2)
    lagdiff[x] = lagdiff[x] – 365

    lagdiff = c(lag[1],lagdiff)

    lag = cumsum(lagdiff)

  390. Greg… your plots still have a 1 in 128 chance of being a fluke… so I wouldn’t get too excited :)

    Unless you were looking for an alternating pattern that started off with heads instead of tails. Then the chances of a fluke would be 1 in 256 :)

  391. AJ, thanks for the code suggestion. I’ll try to test it later. It would be interesting to see the phase without the flip.

    I appreciate the 1:128 warning , what needs to be done here is to extract the data and do frequency analysis, not talk about annual flip-flop. That is enough to suggest there’s something worth investigating, it is not a result over 8 individual years.

    What is much more interesting is whether there is a clear presence of 2nd and 3rd harmonics. I see suggestions of this but of course it’s easy to be fooled just scanning a few plots by eye.

  392. Greg, I applied my change and it look better. Update plot and source as well:

    https://sites.google.com/site/climateadj/argo-sine-fitting-2012

     ############ avoid phase wrap-arounds
     
     bdiff = diff(b)
     
     x = bdiff < -(365/2)
     bdiff[x] = bdiff[x] + 365
     x = bdiff >= (365/2)
     bdiff[x] = bdiff[x] - 365
     
     bdiff = c(b[1],bdiff)
     
     b = cumsum(bdiff)
     
     ############
    

    Apparently code can be inserted into wordpress:

    http://en.support.wordpress.com/code/posting-source-code/

    Now I just gotta fix the longitude selection issue.

  393. So in summary you have to replace the getncsample and sinefit functions. The getncsample function now has a subfunction selectncsample that has to be added.

    Both programs on the page have been fixed.

  394. Many thanks for the improved code AJ. Being able to limit the sample to separate ocean basins is valuable.

    It does look like 2011 was an oddball. It stared like an even year above 300m then veers back to odd year pattern below 400m. I need to pull out complete time series rather than looking at 12m snippets but it’s interesting.

    thx.

  395. Something like this will get you a detrended timeseries:

    require(ncdf)
    require(abind)
    
    ##############################################################
    ############## INSERT FUNCTIONS getncsample and selectncsample
    ##############################################################
    
    nc<-open.ncdf("argo_2005-2012_grd.nc")
    
    #
    #### PARAMETERS ####
    # 
    
    sy<-01 # 1st year = 1
    cy<-08 # number of years
    
    t <- (1:(cy*12))/12
    
    ub<-500     # Upper Level Depth
    lb<-500   # Lower Level Depth - ARGO data goes to 2000M
    
    wb<-160     # western boundary
    eb<-280     # eastern boundary
    
    samplat<- -45
    
    nb<- samplat + 1
    sb<- samplat - 1
    
    #
    #### MAIN PROGRAM ####
    # 
    
    ncsource1 <- "argo"
    ptnc <- getncsample(nc, ncsource1, wb, eb, sb, nb, ub, lb, sy, cy)  
    
    #
    # Get rid of longitudes that have NA values.
    #
    
    ptlong <- apply(ptnc,1,function(x)all(!is.na(x)))
    ptlong <- which(ptlong==TRUE)
    ptsamp <- ptnc[ptlong,,,]
    
    ptsamp <- apply(ptsamp,3,mean)
    
    ptresid <- resid(lm(ptsamp ~ t))
    
    plot(t,ptresid,type='l')
    

    Anyway, my wife just got back from a road trip, so we got some “catching up” to do. Think I’ll sign out now :)

  396. Thanks for the code, should save some time. Fairly strong 4year pattern down at 1000m. Again more like rectified 8 year. At 500m more like a 4 year ramp.

    Look in again on Monday, I should have something clearer.
    Have a good weekend ;)

  397. I’ve dropped the deeper stuff, though it’s interesting short signals go that deep, amplitude is very small. However, variations around 75m-100m are very interesting compared to the surface.

    http://climategrog.wordpress.com/?attachment_id=535

    http://climategrog.wordpress.com/?attachment_id=534

    http://climategrog.wordpress.com/?attachment_id=533

    http://climategrog.wordpress.com/?attachment_id=532

    http://climategrog.wordpress.com/?attachment_id=531

    http://climategrog.wordpress.com/?attachment_id=530

    Often there is larger variation at these depths than at the surface. Also the strong annual peak seems to anti-correlate with these depths.

    Even as far north as 25N , at 200m (grey line) the annual cycle peaks as we would see in SH.
    At 25S similar suppression of the the surface peak and 200m showing peaks in Jan and June.

    These are all 160E to 280E so span trade wind / El Nino east-west variation.

    10N is near the ITCZ but still seems dominated by SH like timing.

    http://climategrog.wordpress.com/?attachment_id=533

    There does not seem to be much evidence of a downward propagating wave from the surface insolation. Often 75m is fairly flat as the surface pattern morphs to the lower annual cycle.

    This data really needs to be visualised as an animated 3D plot. I have seen this done but only for very short periods. IIRC that was depth of the thermocline over 2 years.

  398. Nice graphs Greg. Yes the variations below the thermocline are very small. I think the largish variations under the tropics, say between 50-150m, probably have something to do with the trade winds affecting the thermocline, but what do I know?

    It would be interesting to see an animated gif of your sample area, say above 200m, with each frame consisting of an image plot of monthly temperature by latitude and depth. Maybe I’ll try that over the coming days.

  399. Thanks AJ. I may play around with that code. It’s a useful way to visualise stuff.
    It’s interesting to see how the annual cycle is almost in anti-phase by the time it gets to 150m.

    I’ve been looking at my initial idea of frequency content. 10S and 10N are quite interesting.

    http://climategrog.wordpress.com/?attachment_id=538

    http://climategrog.wordpress.com/?attachment_id=540

    The latter is close to ITCZ and the frequencies marked correspond to QBO and 3.7 years: a prominent cycle I’ve found in a lot of SST data and trade winds and even arctic ice extent.

    There’s a whole chain of interlinked frequencies and interference patterns, which could be shadow chasing, false patterns or Tsonis type linked oscillators.

    The ARGO data is a little too short to have too much confidence in these longer periods but it’s interesting how the deeper water, especially around 150m, seems to tie in with surface phenomena with a global scope.

    Thanks again for the sample R code.

  400. [adding notify as well, doubt if the comments will be open much longer]

    btw… did you mean to label both of your plots 10N?

  401. http://climategrog.wordpress.com/?attachment_id=538

    -10N=10S ;)
    perhaps lat=-10 would have been better.

    That one which around the level of Nino1 region is the most interesting. It has a freq structure which is very consistent with depth. The second peak after annual is 2.85y = 34mo

    I think the ENSO pseudo cycle of “3 to 5 years” is likely to be an interference pattern of rather constant 3 AND 5 y oscillations. This is “3 year” part.

    Also of interest, 18mo and 6mo peaks at 10S. I would have expected a clearer 6mo component in the tropics due to insolation. It being weak at -10 and almost invisible at 10N

    Your animation also shows the latitude of ITCZ to be notably cooler at depths >=75m
    I thought ITCZ was an atmospheric phenomenon (shows what I know).

  402. I’ve been looking at this from a different angle. Similar plot to the last but only for all longitudes and down to 1000m, which in essence treats all the oceans as one. Values were normalized at each level.

    https://sites.google.com/site/climateadj/argo-animation-normalized

    As an analogy, if we consider the red bands emanating from the tropics as “lungs”, it looks to me like they are “inhaling”. That is, they are moving poleward, which you can sort of see when the animation recycles back to 2005. IIRC from a couple of years ago, the highest PTemp trends can be found on the poleward side of the “lungs” and the most negative on the tropical side.

    How long will WUWT leave this post open anyway?

  403. How long?: I sent you my email, didn’t you get it?

    I’m not really seeing the breathing. However, what is that source of deep heat around 15N ?! Maybe we’ve found where the missing heat is hiding ! LOL.

    seriously, looks like a top end of a deep heat source. Any variation in that is going to get integrated into OHC. That may help explain why deep oceans are heating while the surface isn’t.

    I’ll have to fiddle with a few parameters.

  404. I flipped the variables to get an W-E profile across Nino1 latitudes.
    (I upped the z rng to 32 to fill some white spots appearing in 2010.)

    Nothing too interesting on this rather ENSO neutral period but quite a marked E-W variation through the year.

    I guess La Nina years are when the blue stuff makes it to the surface.

  405. I think the E-W animation would be more useful done on anomalies. I can see that there is some inter-annual variation in the pattern but it’s rather swamped out. I don’t think there’s an anomaly field so that may be a bit of a drag.

    I don’t like anomalies but with such sort TS, filtering is problematic.

    d/dt is a kind of 1/f HP filter, so I may try using 1-d/dt as a low-pass.

    (-0.5,0,0.5) is a three point approx to diff.
    (0.5,1,-0.5) should provide a degree of LP.

    Never tried doing convolution in R though. Any suggestions on how that could be added?

Comments are closed.