the Air Vent

Because the world needs another opinion

Cloud Driven Temperature Variation

Posted by Jeff Id on June 3, 2010

Roy Spencer had what I consider an interesting post on a model with negative feedback – copied below.  He was able to demonstrate that warming of a magnitude we measure in a modeled climate system with strong negative feedbacks to CO2 rather than the strong positive feedbacks which lead to global doom IPCC style.  IOW, we don’t need aerosols and high CO2 forcing to achieve a match to what has been measured, we only need natural climate variation.  In addition, were this model to be more correct, it would indicate that CO2 cannot cause the kind of warming predicted by the other pro’s.  The second interesting point is that this negative feedback matches their interpretation of some recently obtained satellite cloud data.

———–

Millennial Climate Cycles Driven by Random Cloud Variations

June 2nd, 2010 by Roy W. Spencer, Ph. D.

I’ve been having an e-mail discussion with another researcher who publishes on the subject of climate feedbacks, and who remains unconvinced of my ideas regarding the ability of clouds to cause climate change. Since I am using the simple forcing-feedback model as evidence of my claims, I thought I would show some model results for a 1,000 year integration period.

What I want to demonstrate is one of the issues that is almost totally forgotten in the global warming debate: long-term climate changes can be caused by short-term random cloud variations.

The main reason this counter-intuitive mechanism is possible is that the large heat capacity of the ocean retains a memory of past temperature change, and so it experiences a “random-walk” like behavior. It is not a true random walk because the temperature excursions from the average climate state are somewhat constrained by the temperature-dependent emission of infrared radiation to space.

A 1,000 Year Model Run

The temperature variability in this model experiment is entirely driven by a 1,000 year time series of monthly random numbers, which is then smoothed with a 30-year filter to mimic multi-decadal variability in cloud cover.

I’ve run the model with a 700 m deep ocean, and strong negative feeedback (6 Watts per sq. meter of extra loss of energy to space per degree of warming, which is equivalent to only 0.5 deg. C of warming for a doubling of atmospheric CO2. This is what we observed in satellite data for month-to-month global average temperature variations.)

The first plot below shows the resulting global average radiative imbalance, which is a combination of (1) the random cloud forcing and (2) the radiative feedback upon any temperature change from that forcing. Note that the standard deviation of these variations over the 1,000 year model integration is only one-half of one percent of the average rate at which solar energy is absorbed by the Earth, which is about 240 Watts per sq. meter.

I also computed the average 10-year trends for all 10-year periods contained in the 1,000 year time series shown above, and got about the same value as NASA’s best radiation budget instrument (CERES) has observed from the Terra satellite for the ten-year period 2000 – 2010: about 1 Watt per sq. meter per decade. Thus, we have satellite evidence that the radiative imbalances seen above are not unrealistic.

The second plot shows the resulting temperature changes over the 1,000 year model run. Note that even though the time scale of the forcing is relatively short — 30 year smoothed monthly random numbers — the 700 m ocean layer can experience much longer time scale temperature changes.

In fact, if we think of this as the real temperature history for the last 1,000 years, we might even imagine a “Medieval Warm Period” 600 years before the end of the integration, with rapid global warming commencing in the last century.

Hmmm…sounds vaguely familiar.

The main point here is that random cloud variations in the climate system can cause climate change. You don’t need a change in solar irradiance, or any other external forcing mechanism.

The above plots also illustrate the danger in comparing things like sunspot activity (and its presumed modulation of cloud cover) to long-term temperature changes. As you can see, the temperature variations in the second plot look nothing like the global energy imbalance variations in the first plot. This is for two reasons: (1) forcing (global radiative imbalance) due to cloud variations is related to the time rate of change of temperature….not to the temperature per se; and (2) the ocean’s “memory” of previous forcing leads to much longer time scale temperature behavior than the short-term cloud forcing might have suggested.

The fact that climate change can be caused by seemingly random, short-term processes has been totally lost in the climate debate. I’m not sure why. Could it be that, if we were to admit the climate system can vary in unpredictable ways, there would be less room for our egos to cause climate change?


67 Responses to “Cloud Driven Temperature Variation”

  1. timetochooseagain said

    How realistic is the 30 year timescale for cloud behavior?

  2. steven Mosher said

    The fact that climate change can be caused by seemingly random, short-term processes has been totally lost in the climate debate. I’m not sure why. Could it be that, if we were to admit the climate system can vary in unpredictable ways, there would be less room for our egos to cause climate change?”

    Yup. Nobody, whose job it is to explain and predict things, likes to admit that certain things may be beyond prediction or understanding.

    I think for clarity, however, you need to make clear that on balance more C02 will make the planet warmer. I think your point here is more relevant to the intractable problem of parsing out the contribution of GHGs versus the variability the system is inherently capable of.

  3. DeWitt Payne said

    It is not a true random walk because the temperature excursions from the average climate state are somewhat constrained by the temperature-dependent emission of infrared radiation to space.

    For the unit root fans: Temperature is a near unit root process, not a pure unit root, because the bucket leaks and the leak rate is proportional to the height of water in the bucket. A unit root or random walk process can wander arbitrarily far from the origin over time. A near unit root process is bounded.

  4. DeWitt Payne said

    Along the lines of Koutsoyiannis, you could include a process linking cloud cover to temperature. Then, rather than driving it with random numbers, you just start it from some initial conditions. It’s possible that there is some set of initial conditions where the system is stable. However the system may not decay to that state, but just keep wandering around it.

  5. timetochooseagain said

    On some level it becomes a statistical problem. You need some sort of probability distribution to describe natural variability you can’t measure. With no reason to presume otherwise, the distribution of these variations would center around zero. As Mosher notes, this makes the problem of attribution difficult because it demands larger confidence intervals. It’s harder to reject the hypothesis of natural variability as the only important factor. As usual you can’t prove anything, at least not yet.

  6. David S said

    Mosh….which is likely to be an extremely complex exercise with both positive and negative feedbacks operating over different time scales, ie does the system amplify or dampen the effect of CO2, or even both at once in different ways, and at different speeds? A long long way away from the linear or even science-defying exponential trends shown in much alarmist material, and not something that the IPCC or RealClimate will find easy to explain to their respective audiences.
    Dr Spencer…presumably a different set of random numbers will generate a different overall temperature history, so although the system is capable of such variability, we do not know what actually happened and to what extent man made CO2 has distorted the natural cycles, or whether in fact current CO2 levels are within the range experienced in prehistoric times, so can be viewed as “natural”.

  7. Steve Fitzpatrick said

    As I commented over at Roy Spencer’s blog, it seems to me that it is a stretch to assume cloud cover changes could be random (or even mostly random), since changes in ocean surface temperature resulting from any change in cloud cover would for certain change the amount of evaporation and precipitation (these must be must be equal in total over any period longer than a few weeks). Rainfall and clouds are clearly connected, so it seems reasonable to expect feed-backs from cloud cover to make the variations in clouds not so random. A reasonable expectation (I think) is that increasing ocean temperature would increase tropical rainfall and tropical cloud cover.

  8. Mike H said

    Chaos theory seems to have eluded the climate scientists. Yes indeed.

    Slightly off topic but I have wondered, is it that Earth’s global climate behaves as a Lorentz attractor where periods of “snow ball” Earth are one of the Lorentz attractor poles and warm Earth periods are the other attractor pole? In either state, the global climate varies a great deal and as a result Earth’s climate sometimes switches poles. History tells us that snow ball earth is the stronger attractor.

  9. Steven Mosher said

    I hasten to add this. Just because an observed change is WITHIN the bounds of “natural variation” ( observed variation with no identified physical cause) does not mean that the observed change is the result of “natural variation.” For example, After a volcano erupts we note the cooling of the planet ( assuming certain conditions are met WRT locations, volumes of aerosols etc) The COOLING is within the bounds of “natural variability.” Yet, we are able to attribute the cooling to that physical cause. Noting that a change is within the “natural bounds.” ( WIllis and his Null hypothesis) is merely a lack of curiousity. Throwing one’s hands up and muttering “chaos” does free up the afternoon for golf, but it’s hardly a satisfactory answer. We probably won’t know the exact contribution that GHGs make and the ‘true’ sensitivity parameters. That said, blindly pumping substances into the atmosphere that we know have some effect, is not exactly prudent.

  10. Steven Mosher said

    david S.

    “which is likely to be an extremely complex exercise with both positive and negative feedbacks operating over different time scales, ie does the system amplify or dampen the effect of CO2, or even both at once in different ways, and at different speeds?”

    Stop asking tough questions

  11. Tim Clark said

    This might be the paper you were looking for regarding recent Arctic ice consensus.

    http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6VBC-4YKFMY0-2&_user=10&_coverDate=03%2F12%2F2010&_rdoc=1&_fmt=high&_orig=search&_sort=d&_docanchor=&view=c&_acct=C000050221&_version=1&_urlVersion=0&_userid

  12. timetochooseagain said

    9-You are completely correct. In fact, rather than throwing up our hands and playing golf, the discovery of the possibility for larger confidences intervals around our null hypothesis suggests that we should invest effort into narrowing them by figuring out ways to better constrain potential past cloud, and climate, variations.

    I always thought that when a problem proves intractable, you should come at it from a new direction. For example, instead of trying to determine the natural impacts on climate, if we constrain the anthropogenic impacts by determining the sensitivity, then the residuals of removing that signal would be the natural variations.

  13. timetochooseagain said

    12-“if we constrain the anthropogenic impacts by determining the sensitivity”

    I should add that constraining the anthropogenic impact will also require better determination of the forcings, in addition to the sensitivity.

  14. Steven Mosher said

    People mistake natural variation as an explanation, when in fact, it is the absence of an explanation and merely an observation. Nothing irks me more (hyperbole) than the claim that phenomena are explained by “natural variation.”

    I observed the natural variation of temperatures for several days now. There are wide swings between the low and the high that seem to be driven by the sun. A cloud passed by. It cooled. But the change I saw was smaller than the natural variation, hence, clouds do not cause cooling. cooling is explained by natural variation. there is ‘nothing’ to explain.My Null Hypothesis about the variation in daily temperatures is robust. Framed this way, we see how silly and disengenious such arguments are.

  15. DeWitt Payne said

    Re: Tim Clark (Jun 3 15:04),

    Your link was broken, at least for me. This one works for me.

    History of sea ice in the Arctic

    Quaternary Science Reviews

  16. timetochooseagain said

    14-The daily insolation cycle is known and and it’s impact can be accounted for. The natural cloud variations that could cause climate changes are NOT known and CAN’T be accounted for. So the situations are not really comparable. If I took the daily insolation cycle I can notice that there are effects which aren’t explained by it and can start looking at possible causes. But if those residuals are within the measurement error there may be no point.

  17. Kenneth Fritsch said

    That said, blindly pumping substances into the atmosphere that we know have some effect, is not exactly prudent.

    Steven Mosher, when you make the above statement, I found it a greatly oversimplified explantion of the real situation. We pump GHGs into the air because we have a relatively cheap source of energy that provides higher standards of livings for most citizens of the world or at least a potential for a higher standard given a relatively free society. That situation does not square with blindly pumping GHGs into the atmosphere. We could, on the other hand, blindly stop using a relatively cheaper source of energry because it – well it makes us all feel better and so much better with the source of cheap energy leaking out of the bottom of the Gulf of Mexico.

  18. Howard said

    Mosher says:

    We probably won’t know the exact contribution that GHGs make and the ‘true’ sensitivity parameters. That said, blindly pumping substances into the atmosphere that we know have some effect, is not exactly prudent.

    We don’t need to be exact to be useful. The earth is horseshoes and hand-grenades, not a Joint Strike Fighter ;)

    The appropriate order of magnitude of the contribution that GHGs make is done: radiation physics gives us 1.2’C for doubling. The current sign and order of magnitude of the sensitivity parameters (feedbacks??) are swag’ed from a back of a postage stamp calculation. Obviously feedback systems are where a large chunk of research needs to be focused. Also, determining the underlying driver of the natural variation is important to truly know where we are at and where we might be going with temperature departures.

    You are correct, blindly pumping CO2 MAY not be prudent. More importantly, considering abandonment of cheap energy that has doubled lifespans and ended real poverty in the West IS not prudent. Denial cheap energy to developing and third world nations is immoral. The point being is that we need not be in a blind hurry to end our CO2 pumping. We will have to burn a lot more coal to fuel the technological revolution to replace it.

    The current settled science on AGW with a wink and a nod to catastrophic tipping points is just as blind as the deniers who think they are smarter than Planck.

  19. timetochooseagain said

    17-I am reminded of a thought I brought up earlier to someone in another context. Saying we shouldn’t pump CO2 etc. into the atmosphere to avoid it’s effects is like saying we shouldn’t eat food, to end the obesity epidemic. Never mind the starvation epidemic to follow.

    It’s just not prudent to eat, given that we could get fat.

  20. Eric Steig said

    Spencer’s points about memory and random walk and all that are fine, and I’d agree that these basic points are frequently ignored in discussions about climate, not just in the blogosphere but also in the mainstream published literature. But it is an exaggeration to say this is ‘overlooked’. For one thing, coupled climate models incorporate this sort of system ‘memory’ (because they have an ocean, just like the real world). See e.g. Roe and Steig, 2004, J. of Climate, and Roe, 2008, Reviews of Geophysics for a discussions of red noise and climate memory.

    None of this changes the conclusions that the natural variability has likely been exceeded by the CO2 forcing, nor that it will certainly be exceeded by the time CO2 doubles, any less true.

  21. timetochooseagain said

    20-Dr. Steig, I am unaware of any GCM which has shown variation comparable to figure 2 in a control run. The models I have seen always seem to have mostly much shorter term variability. Are you aware of any contrary results?

  22. Carrick said

    Andrew:

    The models I have seen always seem to have mostly much shorter term variability.

    Is this what you meant to say?

    The climates have a lot of trouble reproducing short-period variability, not so much with longer period variability.

    In any case, I agree with Eric on his final points. I think much of the warming since 1980 is from CO2 forcings, and if we double atmospheric CO2 concentration, this will likely overwhelm “natural” variability. Though with the caveat that increasing CO2 may increase natural climatic variability (increased CO2 forcings could yield larger amplitudes for atmospheric-ocean oscillations, so it may be that CO2 will alway be “chasing” “natural” climate variability).

  23. Carrick said

    Ugh. Make that “the climate models have a lot of trouble…”

  24. gallopingcamel said

    After reading Roy Spencer’s preview to his paper on clouds (May 7th on his blog) I was full of expectations.

    This new mini-revelation is a bummer; I was hoping for much more.

    I guess we will have to wait for his paper to be published in the Journal of Geophysical Research. It will probably turn out to be underwhelming.

  25. timetochooseagain said

    22-“[models] have a lot of trouble reproducing short-period variability, not so much with longer period variability.”

    I was specifically referring to control runs, which include no external forcings. It is true that the timing and magnitude of the short term changes in models is often way off, which is to be expected I suppose given that those are “weather” and thus the initial state is crucial. But I wasn’t referring to the realism of these fluctuations with regard to observed variability. I was simply saying that, if you run a control simulation (no CO2 or “natural” or “anthropogenic” forcings whatsoever) the average climate over a period of more than about thirty years will be the same in those models century after simulated century, totally unlike Spencer’s figure 2. AR4 bragged that they had gotten rid of the tendency for earlier models to “drift” away from a firmly steady state over long time periods without forcing, and to some extent this was an improvement, since the earlier models often didn’t conserve mass over long periods of time, either. Still, how realistic is a climate which is completely unchanging without external forcing, except on timescales of a few years?

  26. Layman Lurker said

    #22 Carrick

    The climates have a lot of trouble reproducing short-period variability, not so much with longer period variability.

    I think there could be much argument on this relating to assumptions and constraints of model parameters.

  27. Yarmy said

    It would be nice to have some proxies for cloudiness over time, but I’m not aware of any. The only study I know of is Hans Neuberger’s ‘Climate in Art’ which studied several thousand paintings over the period 1400-1967. I can’t find the original, but this is from Brian Fagan’s book ‘The Little Ice Age':

    “His (Neuberger’s) statistical analysis revealed a slow increase in cloudiness between the the beginning of the 15th and mid-16th centuries, followed by a sudden jump in cloud cover. Low clouds (as opposed to fair-weather high clouds) increase sharply after 1550 but fall again after 1850. 18th and early 19th century summer artists regularly painted 50 to 75% cloud cover into their summer skies. The English landscape artist John Constable, born in Suffolk in 1776 and a highly successful painter of English country life, on average depicted almost 75% cloud cover. His contemporary JW Turner, who traveled widely painting cathedrals and English scenes, did roughly the same.

    After 1850, cloudiness tapers off slightly in Neuberger’s painting sample. But skies are never as blue as in earlier times, a phenomenon attributes to both the ‘hazy’ atmospheric effects caused by short brush strokes favoured by impressionists and to increased air pollution resulting from the Industrial Revolution, which diminished the blueness of European skies.”

  28. stupmy said

    If we cannot replicate or understand the many complex natural forcings, we can in no way attribute any warming to man. Attempting to do so is nothing but guess work. The current climate models have done little to progress our understanding of the climate and how it works.

    Unless we know the planets initial conditions to an unobtainable level of accuracy, we can never make any reliable climate forecast, due to the internal variability discussed above, and the long term thermal memory of the oceans, which can lead to 1000+ year trends in temperature.

    The fact that the IPCC models can look similar to the temperature record for only one short period of time is either that of luck or force fitting (I would argue the latter). By force fitting the data, they can then argue that natural variabiliy plays little role, but thats just circular reasoning.

  29. sod said

    most “sceptics” always forget the most important aspect of such “other” explanations of temperature rise:

    these random effects would be in ADDITION to the greenhouse effect, not instead of it.

    until Spencer has found serious evidence for current warming being caused by random cloud changes, and CO” not having a relevant effect, we have a pretty problematic result here:

    Spencer actually says: we should add up to 1°C for random cloud changes to all our climate models!

  30. Andrew said

    29-You obnviously haven’t got a firm grip on the hypothesis involved here. There is only so muc warming to be caused by anything in the observational record. You can’t just add in these effects, they potential contribute some fraction of the change, but if the change you get exceeds what’s observed, one of your effects is to large, or your missing some cooling influence.

    And again NOBODY (of consequence) DENIES THAT CO2 HAS SOME EFFECT! Why sod and others seem to stick to this point to debate on seems to only be explained by the fact that it is firm ground to stand on and forces their opponents to stand on strawmen.

  31. Laws of Nature said

    Hi Jeff,

    hmm .. after you were very nice with your explanations last time, I would like to shoot another question at you . . my apologies for not posting completly in the right topic:
    In http://geoplasma.spaces.live.com/blog/cns!C00F2616F39D0B2B!592.entry
    I saw the following paragraph:

    [..CO2 rf = f * ln([CO2]/[CO2]prein)/ln(2)
    where f= rf for CO2 doubling

    In further documentation according to the IPCC, the “Radiative Forcing” ÄF, in watts per square meter, due to additional carbon dioxide in the atmosphere, can be calculated from the formula:

    ÄF = 5.35 ln C/Co
    The value 5.35 in this equation and the term [CO2]prein in the generalized equation demonstrate that the forcing parameter is based on the 100ppmv increase from the preindustrial value of 280ppmv and the 0.6°C of measured temperature over the time period that this 100ppmv increase occurred.
    Further documentation in the IPCC reports states that the forcing of each watt/m2 raises the global temperature by 0.75°C + 0.25°C.
    The Nimbus 4 satellite measured the thermal radiation spectrum of the Earth in 1970, when the CO2 concentration was 325ppmv as measured at Mauna Loa. ..]

    Well the thing is, that I realized, that the math does not sum up!
    Only if you take the Nimbus data, which does not make much sense to me, you will get somethin like:
    0.6°C * 0.75W/m2/°C = 5.35 ln (325ppm/280ppm) [W/m2]

    Is this the formula which gave rise to the 3.7W/m2 for the doubling of CO2? (With actual number you will get quite lower forcings)

    Well doen’t that formula simply mean, that with ln(380/280)/ln(2)=43% we have seen quite some part of that effect already (if the doubling is even possible)

  32. KevinUK said

    Mosh

    “Noting that a change is within the “natural bounds.” ( WIllis and his Null hypothesis) is merely a lack of curiousity.”

    “People mistake natural variation as an explanation, when in fact, it is the absence of an explanation and merely an observation.”

    Just because someone is prepared to admit that they can’t fully explain and account for the observed effects of a phenomena like climate change does not mean that they are not curious, far from it! In my opinion it means that they are not sufficiently arrogant (as most climate scientists are) enough to try and pretend that they know the answer when in fact they (climate scientists) clearly don’t.

    Eric ‘East Antarctica is warming at an alarming rate’ Steig

    “None of this changes the conclusions that the natural variability has likely been exceeded by the CO2 forcing, nor that it will certainly be exceeded by the time CO2 doubles, any less true.”

    Other than the computer models used by arrogant climate scientists like yourself Eric, just exactly where is the evidence to support your statement that ‘the natural variability has likely been exceeded by the CO2 forcing, nor that it will certainly be exceeded by the time CO2 doubles’?

    Carrick

    “In any case, I agree with Eric on his final points. I think much of the warming since 1980 is from CO2 forcings, and if we double atmospheric CO2 concentration, this will likely overwhelm “natural” variability. Though with the caveat that increasing CO2 may increase natural climatic variability (increased CO2 forcings could yield larger amplitudes for atmospheric-ocean oscillations, so it may be that CO2 will alway be “chasing” “natural” climate variability).”

    So you think ‘much of the warming since 1980 is from CO2 forcings’? I don’t! I think some of it is but that the majority of it is due to wholely natural climatic variability as witnessed by the clear correlation between the 1910 to 1940 warming cycle followed by the 1940 to 1970 cooling cycle, followed by the 1970 to 2000 warming cycle followed by the 2000 to 2010 stasis to the PDO/NAO/AMO/ENSO oceanic positive and negative phases.

    I also think that the 20th century warming trend (the IPCC’s claimed 0.6C/century warming trend) is entirely a natural recovery from the nadir of the Little Ice Age which itself was preceded by a cooling trend from the zenith of the Medieval Warm Period. Both of these well documented climate periods were global and NOT regional as Michael Mann and his fellow ‘team’ members like Eric Steig would have us all believe. Finally Carick where is the evidence to support your statement that ‘increasing CO2 may increase natural climatic variability’?

    KevinUK

  33. sod said

    why don t you explain that “recovery from x” “forcing” to me, please?

  34. Sean said

    I may be mistaken but I think the point of what Roy Spenser is doing is not to try to make a comprehensive model of the climate. I think he is just trying to show that with his very simple model with relatively high frequency cloud variation and an ocean that carries away and redistributes heat that you can get a much lower frequency temperature profile with warm periods and cold periods that actually look like the paleo climate record. When you think about the basis of most climate alarmism, people see a correlation between the instrumental temperature record and the CO2 content then they create a model that shows CO2 drives the temperature. Then they go on to say they’ve looked at everything else and this is the only explanation. Roy is just saying there are natural vaiations that can produce the same results.

  35. Kenneth Fritsch said

    I think TTCA makes a valid point about the amplitude of the climate model control runs and Roy Spencer’s time series. Can we get a link to some typical climate model control runs (those with the ocean coupled)?

    Perhaps since Eric Steig brought the coupled models up he can provide some links for comparison with Spencer’s model.

  36. KevinUK said

    sod

    “why don t you explain that “recovery from x” “forcing” to me, please?”

    I’ll give you a big clue.

    Go to window, look out of it and you’ll hopefully see a large very bright glowing ball shaped object in the sky. Make sure you don’t look directly at it otherwise you might go blind. That’s the source of the of the ‘forcing from x’ thingy that I’m talking about.

    Now go and do some Googling on the PDO, NAO and ENSO and see what you can find out about what causes them to change phase. Even better why not send an email to Kevin ‘I did not say there was a relationship between hurricanes and global warming – honest!’ Trenberth and he can tell you all about the work he’s done on the PDO.

    >
    > The fact is that we can’t account for the lack of warming at the moment
    > and it is a travesty that we can’t. The CERES data published in the
    > August BAMS 09 supplement on 2008 shows there should be even more
    > warming: but the data are surely wrong. Our observing system is inadequate.
    >
    > That said there is a LOT of nonsense about the PDO. People like CPC are
    > tracking PDO on a monthly basis but it is highly correlated with ENSO.
    > Most of what they are seeing is the change in ENSO not real PDO. It
    > surely isn’t decadal. The PDO is already reversing with the switch to
    > El Nino. The PDO index became positive in September for first time
    > since Sept 2007. see
    > http://www.cpc.ncep.noaa.gov/products/GODAS/ocean_briefing_gif/global_ocean_monitoring_current.ppt
    >
    > Kevin

    Why not have a look at

    http://icecap.us/images/uploads/WashingtonPolicymakersaddress.pdf

    or

    http://www.esrl.noaa.gov/psd/people/gilbert.p.compo/CompoSardeshmukh2007a.pdf

    for starters and then have a look at

    http://www.appinsys.com/GlobalWarming/PDO_AMO.htm

    and if you want to learn something about the PDO and its effects on climate then why not start here

    http://www.atmos.washington.edu/~mantua/REPORTS/PDO/PDO_egec.htm

    and have a look at

    http://www.agu.org/pubs/crossref/2002/2002GL015191.shtml

    Then once you’ve finished your research why don’t you contact the UK Met Office and ask them just where exactly within HADCM3 are the equations that model the PDO, NAO, AMO and ENSO and ensure that they fully account for natural climatic variability. While you are at it ask them how they managed to over come the problem of having to apply ‘flux adjustements’ prior to their development of HADCM3. Just exactly what algorithm did they ass to HADCM2 that meantthat they no longer needed to apply ‘flux adjustements’ to keep their model numerically stable. I’m sure ‘Stottie’ and Myles will be very helpful.

    KevinUK

  37. Niels A Nielsen said

    Chapter 6 of the IPCC shows a figure with simulations using paleo data: http://www.ipcc.ch/publications_and_data/ar4/wg1/en/ch6s6-6-3.html

    The models that don’t use outdated solar irradiance data (eg González-Rouco et al., 2003) show essentially flat global temperatures over the last 20 centuries only interrupted by the occasional short vulcanically forced cooling period.

  38. Carrick said

    sod:

    we should add up to 1°C for random cloud changes to all our climate models

    I agree this is what it says.

    You can’t magically turn of CO2 radiative forcing just because you added another source of variability.

    You next a physics-based justification for dropping the CO2 forcings that doesn’t use bogus, made up theories on thermodynamic laws for example (apologies to people who think they are experts on thermodynamics without having any relevant physics/engineering training).

  39. Niels A Nielsen said

    Carrick #38
    >>You can’t magically turn of CO2 radiative forcing just because you added another source of variability.

    That is the strawman Andrew #30 mentioned.

    Another source of variablity undermines the attribution arguments used by the IPCC and often repeated by you. Why deny that?

  40. Curt said

    Mosh:

    I think what Dr. Spencer referred to as “natural variability” would be better termed as “internal variability” — that is, not requiring any change in “external” forcing to occur, whether from the sun, cosmic rays, or industrial CO2 emissions.

    Any course in differential equations or system dynamics will distinguish between the internally driven and externally driven components of the system response (although most courses will focus on linear equations and systems.)

    I think all that Dr. Spencer was trying to demonstrate here is that seemingly small and short-term variations in a system property such as cloud cover can plausibly lead to what most people would consider surprisingly large changes in another system property such as heat content/”average temperature”.

    Many people think that the possible results of these changes are too small to be of sufficient interest to investigate further into remote causes (although I doubt many would deny that these exist at all). Perhaps Dr. Spencer has shown that further investigation would be warranted.

  41. Kenneth Fritsch said

    I think this discussion is getting off point. I am certain that Roy Spencer and most who read and post here do not disagree that the straight CO2 physics based warming is valid. The uncertainty is with the feedbacks and this is the point of contention between warmists and skeptics. Does anyone think we would or should be concerned about the straight CO2 warming without feedback – and I even include the non doctrinaire warmists?

    What Spencer, I think, is attempting to show is that naturally caused variations in his model are relatively large and the discussion should center, in my view, on how does that compare to climate model controls, what does it infer about the validity of published temperature proxies and reconstructions going into the deeper past and, for that matter, about how well we have captured the recent (mostly instrumental period) for temperature changes and the attributions (besides CO2) attached to them.

  42. Andrew said

    35-If I’ve made a good point then something must be very wrong with the Universe! ;)

    39-Unfortunately it seems there are professional scientists who fall for that straw man, too (eg Dan Kirk-Davidoff, who ironically worked with Lindzen on some papers in the 90’s):

    http://www.drroyspencer.com/2010/06/millennial-climate-cycles-driven-by-random-cloud-variations/#comments

    40-You, and those other commenters I replied to, and anyone else who seems to “get” the point that I think we’ve all basically articulated, gets a Gold Star. Thank goodness, I was beginning to think it couldn’t be made understandable.

  43. Joe Born said

    I’m just a layman, but I’ve run through Dr. Spencer’s model myself, and I conclude from having done so that he’s merely saying the following.

    If you assume that, for whatever reason, the 30-year average net irradiation before temperature-feedback effects exhibits one or two watts per meter of standard deviation, and if you further make the simplifying assumption that all of that radiation imbalance is absorbed (and surrendered) by the oceans’ top 700 meters, then that top 700 meters’ spatial average temperature will exhibit secular variation of the magnitude his graph shows even if it is damped by an additional temperature-dependent outward radiation (“feedback”)of 6 W/m^2-K, which he assumed for his model runs.

    So, if this magnitude of secular temperature variation is comparable to what many scientists believe the earth exhibited over the last millennium–and if you think that you think the rolling 30-year-average net irradiation does indeed vary that much over the centuries–then it appears plausible that a significant amount of the global-average variation we have seen and will see has resulted from net-irradiation changes.

    His model also shows that such (to my layman’s mind) seemingly high net (positive and negative) irradiation actually is quite plausible if you accept that such a significant ocean depth contributes to the system’s thermal inertia.

  44. Jeff Id said

    #43, That’s how I took it too. It was a bit surprising to me but there you go, what we know can fill an entire thimble to the top, what we don’t can fill an ocean.

  45. Eric Steig said

    Guys,

    You are supposed to be skeptics, so how about being skeptical about Spencer’s model?
    There’s a pretty basic problem with it, which you ought to be able to find. I”ll give you a hint to get you started on your debunking:

    What’s the variance of the monthly-averaged radiation imbalance that is the input (forcing) in his model?

  46. timetochooseagain said

    45-I asked the question “How realistic is the 30 year timescale for cloud behavior?” which is, I believe, related to your objection. You are saying that the amount of variation in the radiative forcing from cloud variations is unrealistic. Why is it unrealistic?

    And I am sorry to see that you didn’t answer my question about whether climate model control runs can produce variability on long timescales. Still on the table.

  47. Tim said

    #45 – Eric

    And why would any uncertainities in the data be a problem unique to his model? All theories of climate change depend on noisy data.

    For example, most GCMs presume a jump in TSI in the first part of the 20th century in order to explain the warming during that period. However, more recent analyses of the TSI proxies claim that jump never occurred and that would mean those GCMs would no longer be able to produce a reasonable hindcast. It is possible that the hindcast could be fixed by tuning the aerosol forcings but doing that demonstrates that sceptics are right when they say the hindcast are a result of model tuning.

  48. timetochooseagain said

    47-Unfortunately the solar irradiance effect in the models is far too small for that to be true. But even with the solar forcing increase, the models don’t match the multidecadal trends very well until ~1950. The lack of that solar increase would just mean that the failure to capture the 1911-1941 warming, which was almost identical in rate to the 1979-2009 warming, is ever so slightly “worse than we thought”. And I do mean slightly, like .1% or something. TSI just doesn’t do much.

  49. Tim said

    #48 – timetochooseagain

    The solar effect is as large as aerosols in this study:

  50. timetochooseagain said

    49-Ah yes, the wiki cartoon. Hm, funny that when I look at GISS’s forcings, TSI is barely visible amid all the other tiny effects, totally dwarfed GHG’s:

    You can totally back TSI out from the total if you want, here:

    http://data.giss.nasa.gov/modelforce/RadF.txt

    It won’t make much difference at all. I don’t know what is up with the wiki cartoon but it makes very little sense.

  51. Tim said

    #50 – timetochooseagain

    That is my point. The modellers just fiddle with the aerosol forcings to get the answer they want after talking into account the stuff like solar. If new studies say solar effect is stronger – they reduce aerosols – and vise versa.

    No matter what new data we get the modellers will always claim that CO2 sensitivity is within the “consensus” range. Spencer is showing that others can play the same game and produce plausible models that explain the warming without a high sensitivity for CO2.

  52. timetochooseagain said

    50-Tim, I don’t disagree that models are tuned to agree with the data. Total solar irradiance just usually isn’t a major forcing used to tune the models behavior, since even Lean’s estimates were actually very small for climate forcing.

  53. Jeff Id said

    #45, I’m already skeptical of this model as I suspect Dr. Spencer is. I haven’t spend a lot of time with forcing, so to have any solid opinions at all will require probably months of foundation work. He does seem to believe in the negative feedback numbers to some degree though. I wonder what your thoughts are on the new cloud satellite data, have you had a chance to review it?

    I’ve still been messing around with your 60 stations, although pretty lazily. It’s difficult to match your data to the stations data by name only. I wrote some routines to match them to GHCN by name but they miss about 30% of the time so I spent about two hours verifying and manually matching to GHCN series and finished only the first 30.

    The method you used did select a large portion of urban vs rural, but since warming is supposed to be only since around 1978, I’m eventually going to try a bunch of random selections of 50+ year stations and see how they all match.

  54. Eric Steig said

    In response to a query about what GCMs do, I agree that it is unlikely that any GCM is gong to get the magnitude of variability in a control run seen in Spencer’s Figure 2. That’s kind of the point, isn’t it? On the other hand, there are plenty of models that produce the observed multi-decadal variability e.g. of the so-called “PDO”. So evidence that models don’t get the internal low frequency variability right is weak. That depends of course on what you think the real internal low frequency variability really is. Evidently, Spencer thinks it is much bigger than the IPCC does. That’s what he is illustrating here.

    That goes back to my hint on what’s wrong, for which no one seems to have bothered to do the ‘auditing’ yet.

    Let me try again: What do you have to do with the variations forcing in Spencer’s model to get the variation in SST he gets? To give you another hint: I can’t reproduce his results using his Excel model without undoing a few, shall we say ‘tricks’, in the code. ;)

  55. Tim said

    #54 – Eric

    This is not a classroom and there is no final exam that we have to pass. Being coy about the issues you believe you have found hinders communication.

  56. Jeff Id said

    #54, I don’t know your relationship with RC but if you wanted to post your results here I would take care of it.

    If there were games played by Dr. Spencer, I would be interested and a bit surprised. I’m sick to death of games, it takes too long to figure out and too much of my very limited time. You obviously have far more understanding of what is and was done.

  57. Layman Lurker said

    there are plenty of models that produce the observed multi-decadal variability e.g. of the so-called “PDO”

    Dr. Steig could you expand on this please? Do most GCM’s reproduce this variability? What differentiates models which mimic PDO variability from models that don’t?

  58. Andrew said

    What GCMs reproduce the decadal variability of the PDO? Most modelers I know of admit that a weakness of models is in their ability to reproduce complex patterns involving the timing and magnitude of these various “modes”. This is usual said to be due to their dependence of the initial state of the climate and choatic evolution. The low frequency variability of the GMST isn’t even captured in the models before ~1950.

  59. Layman Lurker said

    Here is the link to the code that Eric Steig is talking about:

    http://www.drroyspencer.com/Simple-climate-model-v1.0.xls

  60. Eric Steig said

    Tim: Au contraire. This is a classroom. The whole world is a classroom.

    If you discover the problems yourselves, you’ll understand them better. I’d be lead you to the answer, but I’m actually very very confident that Jeff (at the least) is perfectly capable of figuring it out. I’m challenging him — and his readers — to do so.

    Regarding Andrew’s question about the PDO, here is a paper on that, in Journal of Climate:

    http://journals.ametsoc.org/doi/abs/10.1175/1520-0442%282002%29015%3C0160%3APIAIEV%3E2.0.CO%3B2

    I’m afraid it’s subscription required (or go to your local University library), but the abstract conveys the key points well.

  61. Tim said

    Eric,

    There are people who have time/inclination to do the work themselves.
    Many readers do not but they do read arguments that others make and adjust opinions accordingly.
    If you think you can debunk Spencer’s claims then I would like to see it.
    I think it is mistake to depend on Jeff having time to do the digging if you wish to communicate your view.

  62. Kan said

    #60

    “If you discover the problems yourselves, you’ll understand them better. I’d be lead you to the answer, but I’m actually very very confident that Jeff (at the least) is perfectly capable of figuring it out. I’m challenging him — and his readers — to do so.”

    This is definitely in line with the Mann et. al. attitude.

  63. Andrew said

    60-Eric, your claim was regarding PDO, the paper in question seems to refer primarily to ENSO. The two phenomena are closely related, but the variability they exhibit is not the same for different timescales. Since Tropical climate seems to have been strongly constrained in geological and recent history to change relatively little, (although ENSO effects are often very large) it is hardly surprising that the low frequency variability would be harder to underestimate. The strong multidecadal patterns in the Mid-latitudes (Where the PDO phenomena is centered) appear to be much less well captured. Look at the Arctic for instance, where temperatures rose more than five times faster than the global mean between 1910 and 1940, cooled nine times as much from 1940 to 1970, and warmed only twice as much from 1970 to 2008, result in the 1990’s temperatures there being on par with the 30’s and 40’s. It wasn’t until the last decade that a few years started exceeding that earlier period’s peak.

  64. Eric Steig said

    Andrew, I’m not making any ‘claims’, I’m just answering your question about whether models capture decadal variability. Yes, at least some do, and yes, that paper I referenced is about the PDO.

    Note that I am not claiming that GCMs in general get the decadal variability right. I’m not sure they do, because GCMS are by no means my expertise, nor do I pay a great deal of attention to them (though I’m working more on this, these days, so stay tuned). The whole question of what the ‘natural variabilty’ is in the real world is a difficult question — indeed, one might say it is THE questions — because in the real world, such variability is contaminated by forcing (sun, volcanoes, CO2, etc.). Plus, we don’t really have enough data. We would actually need about 300 years of data to demonstrate that the PDO is anything other than ‘red noise’, and even that would only work if we could be confident there was no external forcing, which of course will never happen. That means we’re stuck with proxy data, and estimates of the forcing from proxy data, which folks reading this site aren’t too impressed with, I gather ;)

    Jeff: not meaning to score points here. I really do think it would be good for you — and your readers — to go through Spencer’s model and understand it. And I thought you might enjoy the challenge. The basic issue has to do with the magnitude of forcing, which is totally unrealistic. It’s hard to tell exactly what Spencer has done, because he smooths both the forcing and — as far as I can tell from a quick look — the results, and that leads to a pretty misleading picture. I think it is totally unrealistic, but you should look at this yourselves. Again, the question is simple: how much forcing does he need to apply to get the SST variability he claims he can get from clouds? And is that remotely realistic. I think the answer is very clearly no. But as I have repeatedly heard from your readers, why should you take my word for it?

    That said, I will write something up on RealClimate at some point, but like you, I’m busy with my day job.

  65. timetochooseagain said

    64-“And is that remotely realistic. I think the answer is very clearly no.”

    It isn’t really clear to the rest of us, I think. I do want to know if it is realistic. I want to know why you think it isn’t. So far I see no obvious reason to exclude it a priori. Let’s imagine I’m far too stupid to see what’s blindingly obvious. If it is as obvious as you say, and we don’t get it, then we may just be dumb enough that you will need to explain it to us.

  66. Joe Born said

    Since Spencer’s model is simply the linear first-order differential equation Cp(dT/dt)=TSI-lambda*T, where Cp is the heat capacity of the oceans’ top 700m and I interpret TSI to be the temperature-independent component of radiation imbalance, TSI in his spreadsheet cannot be his column-F numbers, which may be what Steig says are unrealistic, but rather the column-G numbers, which are what I take it Spencer is postulating radiation imbalance to be for the purpose of his post. The column-F numbers are just part of a kluge to get the type of column-G numbers he intends to postulate.

    But I’m guessing and I’m a layman. I have to confess that
    I wish he’d have someone translate his (characteristically elliptical) writing into something we great unwashed masses could understand.

  67. Joe Born said

    In my last comment I accused Dr. Spencer’s writing of being elliptical–after mine had just been so.

    For those who have not looked at his spreadsheet (and adjusted it by cranking up the column-F numbers’ variance to arrive at the temperature variation he shows in his graph, I should have stated explicitly why I inferred from his model’s defining differential equation that his column-F numbers are not what he’s using for “TSI.”

    The reason is that his column L, which is what his temperature graph depicts, is the integral of (if we ignore terms irrelevant to this discussion) the difference between column G and the feedback (lambda * T)-containing column M after multiplication by the number of seconds in a month and division by the heat capacity of the oceans’ top 700 m. So it numerically solves the model equation for the case where TSI is given by column G, not by column F.

Leave a Reply

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

WordPress.com Logo

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

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s

 
Follow

Get every new post delivered to your Inbox.

Join 140 other followers

%d bloggers like this: