RealClimate logo

Technical Note: Sorry for the recent unanticipated down-time, we had to perform some necessary updates. Please let us know if you have any problems.

Unforced variations 3

Filed under: — group @ 21 October 2010

Here’s an open thread for various climate science related discussions, to prevent more off-topic clutter everywhere else. We have some good posts coming up, but if you want to discuss something you read in the media, saw in a press release or just wanted to ask about, this is the time.

Some interesting things we’ve seen recently include discussions on the epistimology of climate modelling, Andy Dessler’s adventures in debate land and his new paper on water vapour trends, and a review of trends in the Columbia glacier. Have at it.

Addendum: Kevin McKinney has beaten us to the mention of this, but another recent article of importance is a thorough review of the state of knowledge of drought, past and future, by Dai.  The article is open access here.

573 Responses to “Unforced variations 3”

  1. 201
    Septic Matthew says:

    166, Jim in comment: The history of land and ecosystem “management” is littered with the residue of the unforseen consequences of similarly ill-conceived schemes which appeared, at the time, to be simple, “can’t fail” ideas. Ecosystems are exceedingly complex things that defy easy analysis or prediction, and make no mistake, you are talking about massive rearrangements to the earth system with almost entirely unknown consequences, logistical issues, feasibility questions etc.

    True enough. but it might be worth some tries, as part of a coordinated effort to prevent the end of civilization as we know it, e.g. to prevent the occurance of simultaneous crop failures in all agricultural regions that Barton Paul Levenson warns us of. “Massive rearrangements” are the solutions to global warming,

    [Response: They are?--Jim]

    and the complete consequences are not known to any of them.

    Reforestation and afforestation are among the originally listed “stabilization wedges” surveyed in the journal Science and reviewed from time to time. Only the occasional claim that they might be a panacea is objectionable.

    [Response: And the idea that you can essentially turn all the world's land into forest if you just move enough water around...Jim]

  2. 202
    Dean S. says:

    Regarding TTTM, models, risk to science…

    This is still ongoing argument only because “Tim” has a flwed definition of models, science, and the process of constructive debate…therfore it is simply wasting time as a distraction. As an easy hypocrite, I’ll waste a couple more lines…

    Models are a product of science, not the other way around. Flaws found with fundamentals of model cannot “risk” science, only belief in output of model for particular run. Science is never “at risk” from ANYTHING because it’s the nature of science to be defined by sum of knowledge, therefore new knowledge STRENGTHENS science and becomes science, doesn’t cause it to collapse, only redefines parts of it and encompasses the whole…like above references to difference between card house and jigsaw puzzle.

    Debate involves examining details of argument, defining key terms, considering validity of opposing viewpoints, eliminating logical fallacies, and UPDATING argument when such improvements have been made…rather than constantly returning to original raw argument even after it has been repeatedly shown to be based on false premises.

  3. 203
    Andy Gates says:

    @151, I think you’ve identified a catastrophic failure in TTT’s model of climate science.

    His ‘house of cards’ model predicts flaky, massively-collapsing knowledge regularly overturned by disturbance. Maybe that’s the impression that certain media sources (including the MSM’s “X was wrong!” headlines) gives.

    The ‘jigsaw’ model is more accurate, and more robust: it better predicts how whole revolutions in one area (“oh, this yellow is flowers, not a car”) can only improve the picture and don’t necessarily overturn the existing corners, pond and ducks. Even if you literally turn some upside down (“hey, this isn’t a second duck, it’s a reflection of the duck”) the rest isn’t necesssarily overturned, just informed (“I guess that means there’s got to be more water tiles…”)

    TTT, I humbly suggest that you update your model of how science works.

  4. 204
    t_p_hamilton says:

    Septic Matthew:”If someone has a comprehensive and well written exposition of how increased temps are supposed to lead to increased/unchanged/decreased cyclone intensities, I would like to have the link. Nothing I have found on my own to date includes all the factors.”

    Try Kerry Emanuel’s page. Depending on your level of understanding, pick the appropriate article. Short story: the Power Dissipation Index for Atlantic hurricanes correlates strongly with Atlantic sea surface temperature. This result has stood up for the last 5 years, and hence decreasingly likely to be incorrect as time moves on.

  5. 205

    162 Mine, and re moderator Jim,

    I said the albedo change was a one time event. I think you misread the actual written words.

    Yes, the effect of the albedo change is to create a higher rate of reflection that will continue to exist and have an ongoing effect. But the rate of reflection will not change over time. In other words, the reflection coefficient will be fixed according to the new albedo.

    As carbon dioxide is removed on an on-going basis, the impedance to upward radiation will be continuously decreased, and the rate of IR band energy transmission will continuously increase. In absence of other sources of CO2, the total atmospheric CO2 will continuously reduce as forest growth continues. This is what I meant by the term ‘integrative’ effect.

  6. 206
    Patrick 027 says:

    Re 168 Jim Bullis, Miastrada Co – if a particular method of CCS or otherwise reducing net CO2eq emissions efficiency is too expensive, another (clean energy, infrastructure, efficiency, agricultural practices, diet and lifestyle, peridotite, biochar) will be favored, if in an environment that taxes emissions and pays for sequestration at the same rate. Have you looked into the peridotite solution?

  7. 207

    CM 185,

    The troposphere and stratosphere make up 99% of the mass of Earth’s atmosphere. The mesosphere, just above those, is less than 1%. It doesn’t have much effect on the surface.

  8. 208
    Fred Moolten says:

    Since this is a free-wheeling blog post, I thought I’d introduce another issue here that deserves attention – tropospheric amplification (or the lack of it). Basic principles, including Clausius-Clapeyron, observed increases in tropospheric humidity, and increases in troposphere height all imply that tropospheric warming in the tropics should exceed the rate at the surface. Earlier observations conflicted severely with this expectation. As biases in the instrumental record have been corrected, observations and theory have grown closer, but it’s still not clear that the conflict has been adequately resolved. Most observers believe the ultimate resolution depends on correcting remaining putative cool biases in the troposphere (many authors) and/or warm biases at the surface (Pielke), but is there some additional phenomenon that also contributes to the disparity?

    The alleged absence of a “tropospheric hotspot” (it’s actually a very cold spot with an elevated warming rate, but “hotspot” is the common term used) is a standard argument of contrarians. It involves a nonsensical claim (a la Monckton) and a more legitimate one. The nonsensical version holds that tropospheric amplification is a “fingerprint” of CO2-mediated warming, and its absence proves the absence of AGW. Of course, the amplification is independent of the cause of warming. The more legitimate claim is that current climate sensitivity estimates require a strong water vapor feedback, and the failure of the troposphere to warm in accordance with the expected influence of tropospheric water is evidence that the feedback (and hence climate sensitivity) have been overestimated. (I would note parenthetically that amplification is more a function of the negative lapse-rate feedback due to vapor condensation than to the positive greenhouse gas feedback from increased water vapor, but of course, the two are linked by theory).

    My questions here are not intended to challenge either AGW or climate sensitivity values, which are well documented from multiple sources, but rather to inquire about current attempts to resolve the “hotspot” question. First, are there new data bearing on the issue?

    Second, although I don’t have the data in front of me, it is generally concluded that short term (e.g., one year) datasets show the expected amplification, but multidecadal analyses do not. This raises an interesting mathematical conundrum. If one serially adds short term amplifications to form a long term sum, they should continue to show amplification. If this has not happened, it suggests discontinuities in the record – e.g., either missing observations or downward jumps at various breakpoints, perhaps implicating changes in instrumentation. What is the evidence bearing on this? Comments would be welcome.

    [Response: There is a big confusion here. First off (as you rightly say), the hotspot is expected to occur due to any surface heating simply because there is a small modification to the moist adiabat as a function of temperature. This is well understood physics combining only a statically unstable column with the Clausius-Clapeyron equation. This has little or nothing to do with water vapour feedback however. You would still get the same pattern if water vapour had no impact on the greenhouse properties of the atmosphere. It does however have to do with the 'lapse rate feedback' (which is negative - i.e. it damps the increase of surface temperature driven by a specific forcing).

    To see this think about a system where the temperature rises are uniform with height, and the whole column adjusts to the forcing at the tropopause. Now think of a system where the temperature changes get larger as you go up. The net IR emission out of the system is a weighted sum of the contributions from different levels, and goes up at each level as a function of temperature anomaly. Compare the two situations where the surface temperature anomaly is the same. Which system has the higher IR emission? It should be clear that it is the one with the 'hotspot'. Thus the lapse-rate change allows the surface temperature change for a given forcing to be smaller!

    Now I don't think there is a 'missing hotspot' (for reasons we have gone over in previous posts), but if it isn't there, the implication is that it would imply a greater climate sensitivity than the current models show, not less. - gavin]

  9. 209

    182 Hank Roberts

    Your link does not work.

    I look for the relevance of the 600 trees per acre put forward as a reference. Basically, a 600 tree per acre situation would be a scrub second growth forest. Old growth forests probably are already around 50 to 100 trees per acre.

    A well managed forest project could start with 100 or more trees per acre, and as growth continued, some could be harvested for wood (for permanent wood products). Full growth could well involve 50 very large trees per acre.

    Restoring a forest would go slowly if one depended on nature to thin the scrub tree woods to allow for larger trees to grow.

  10. 210
  11. 211
    Didactylos says:

    I never know how to deal with people with “big” ideas. Big, impractical, impossible ideas, with holes you can sail an armada through.

    Do you smile, take their money and run? Do you point out the gaping holes? Do you waste time trying to bring the “vision” down to earth, and make it workable?

    In the past, I have mixed all these approaches, but impossible visions usually remain impossible. It usually depends on the visionary. Can they see the holes in their plan?

    Take Jim Bullis’s ideas. If I had an idea for an impossibly efficient car (and who hasn’t?) then it would remain strictly fantasy. I know what is needed to turn such an idea into reality (a lot of money, a lot of time, and a lot of people). I don’t think I would ever have the hubris to say “10 times more miles on a gallon” based on nothing but “aerodynamics”. Before I said something like that, I would want to go back to university, study engineering for a few years, then run some models. Even then, I would want expert advice, too.

    How do you deflate these ideas without killing that invaluable inventive mind?

    Jim, sometimes big dreams can work. But they have to be practical. SpaceShipTwo will in the end cost in excess of $300 million. Solar Impulse has cost an estimated $88 million. It’s a long, difficult road, and nobody is going to buy or license something which is only an idea.

  12. 212
    Hank Roberts says:
    Ecological Rest. 27(3):320-329 (2009); doi:10.3368/er.27.3.320

    Special Theme: Climate Change and Ecological Restoration
    Rethinking Conservation Practice in Light of Climate Change

    “Predicted changes in climate present unusual challenges to conservation planners, land managers, and restoration efforts directed toward preserving biodiversity. Successful organisms will respond to these changes by persisting in suitable microsites, adapting to novel conditions, or dispersing to new sites. We describe three general categories of strategies for restoring and managing natural systems in light of likely changes in future climate ….”

    Anyone newly interested in becoming involved in an area like creating forests could do worse than to start by reading, for example

  13. 213
    Patrick 027 says:

    Re 207 Barton Paul Levenson – that’s actually an understatement; the mesosphere is only about 0.1 % of the atmosphere’s mass (the stratopause is near 1 mb).

  14. 214
    Todd Albert says:

    Barton Paul Levenson 136: I’d love to see it. Send it to me at: todda AT me DOT com

  15. 215
    Bibasir says:

    I read the article about JC in Scientific American. She began her dialog with the skeptics, because their blogs had scathing attacks on her paper attributing an increase powerful tropical cyclones to global warming. The article does not say whether she modified her conclusions or whether she convinced the skeptics that her conclusions were correct. If neither of these happened, she not only wasted her time, but also became a shill.

    Curry also asserts that “scientists don’t even know the amount of warming a doubling of CO2 alone would cause.” Say what? Does she really believe that? I sure hate to think we will hear that comment echoed for the next decade.

  16. 216

    211 Didactylos,

    Such scoffing! As if you had read the material at the website limked to my name. It does not appear that you are inclined to read complete sets of words.

    True, it is not well organized, but if you take the time to leaf through the various pages, and actually read some of the stuff, maybe you will see the reason for the headline claims.

    Making a vision workable does certainly take time.

    And you say everyone has had an idea for an impossibly efficient car? Most who learned the laws of thermodynamics have been saved from this exercise. And these folks would not assert that they, or most people, had such an idea. But do tell us more about these ideas.

    As to the forestation idea, the hostility to such an elementary concept is quite amazing. I was hoping someone would have more to say about what China is actually doing in this regard, especially in light of the actual stated plan to do something like this.

    The factor of ten in efficiency for the car I am planning comes from the product of 2 and 5. The 2 comes from making the car half as wide using tandem seating, and the 5 comes from adapting the well know airship form that has a drag doefficient that is about a fifth that of most cars.

    The main impediment to making a car like this is the resistance of the market, represented by people who would rather live in global warmth than to change the way they ride in cars. These folks have a propensity to construct arguments about why global warming is a false threat. These folks are similar to those who construct arguments about why various solutions are impossible.

    Perhaps I should spell out the relationship of market resistance to an idea to availability of money to implement that idea.

  17. 217
    JiminMpls says:

    Just one word for Mr Bullis to ponder:


    If we really want to embark on some large scale reengineering of the planet, let’s start by restoring suburban areas to their wild state.

  18. 218

    206 Patrick 027

    I think you are right about how less expensive things have to emerge. Thus, I was led to look at new forests as such a ‘less expensive thing’.

    Efficiency clearly has potential, but not nearly enough at affordable costs, hence it does not get into the ‘favored’ category on a large enough scale. Biochar has real potential as an ancillary activity related to the forestation project.

    I will look again at peridotite, but of the chemical possibilities, the EPA study at:

    did not put that into a lead position as a CCS answer, even though there is some mention of similar things.

    Apparently you do not accept my point that an environment, where CCS is paid for by any mechanism at all, is an environment where our industrial economy will fail. That is exactly what I am trying to avoid.

    As to efficiency, there are some real limits to what can be done affordably. I have frequently mentioned household shifting from electricity to natural gas for heat using appliances. These include clothes dryers, cooking appliances, and even refrigerators and air conditioners (vapor diffusion). This has gained little traction, partly because there is no profit for the utility in this, and partly because our regulators do not understand the magnitude of heat loss in central power plants. Hence vapor diffusion cooling devices were excluded as inefficient devices, based on their local coefficient of performance only.
    But that is another idea; not really all that big.

    Cogeneration is a big efficiency answer though, maybe of silver bullet rank, where heat engines in motor vehicles are adapted to deliver discharged heat for household use on a distributed basis.

    My efficiency silver bullets are cars, trucks, hybrid wheels for trucks, and distributed cogeneration as mentioned. Add to this a massive new forestation project, and that might be enough silver bullets to get the job done.

  19. 219

    192 Hank Roberts

    Northern California redwoods take on a lot of moisture through there needles, as you say. However, the redwoods of Sequoia National Park seem to get along without doing that.

    Thanks for all the forestation references, which are loaded with background information. However, I am not talking about reforestation, though that is only because it is insufficient.

    The amount of world forestation that has been eliminated is enormous, and I would be delighted to have the land to work with, but the art of the possible tells me to look for affordable land.

    I honor the idea that natural state of things is highly desirable, but also note realistically that we lost that condition long ago. And the possibility to go back to the natural state has also been lost.

  20. 220

    193 TimTheToolMan

    If you are interested in discussing, I’m willing. You can reach me through the OSS site

    I have found that it is easier to explain through conversation in many cases. If you are really interested is getting into the logical constructs. Drop me a note.

  21. 221

    BTW since we are in unforced variations, i started my own little site and am allowing comments. My first discussion piece is

    In case anyone wants to play with the idea.

    okay, time for bed in my neck of the woods

  22. 222

    206 Patrick 027

    I got tripped up on peridotite. You must be referring to the deep rocks that some said could be used to react with CO2.

    The EPA cost analysis only covered what it would take to capture CO2 from the stack. So whether the CO2 simply stayed put in huge underground caverns or reacted with natural rock, it would seem to not matter.

    See the EPA analysis:

    (Much of it is incomprehensible, so far at least. Maybe others will glean more from it than I have yet to manage.)

  23. 223
    John E. Pearson says:

    TEST: {\mathcal E} = m c^3

    Does latex just work now? Is there an internet flavor being used so that one doesn’t have to wrap mathematical formulae with $ or other standard TeX equation offsetting commands? Anyway it is awesome that tex works on RC now!

  24. 224
    Fred Moolten says:

    To Gavin re my previous comment (208). I appreciate your response but it left my two main questions unanswered. Like you, I correlated tropospheric amplification with the lapse rate feedback, but I also mentioned that water vapor feedback and lapse rate feedback are linked via the increase in absolute humidity with preservation of relative humidity or close to it, so that the two phenomena are not independent. Is there a mechanism that would delink them and allow the water vapor effect to operate without a change in lapse rates toward the moist adiabat?

    [Response: I think so. The magnitude of the LR feedback depends on the fact that the rate of change of sat. vapour pressure with temperature is itself a function of temperature. If CC was linear instead of exponential, you'd still have a WV feedback but no LR feedback. ]

    My two earlier questions, in essence, were as follows. First, are there new data either reconciling observations more conclusively with theory, or alternatively identifying additional tropospheric mechanisms that would reduce the amplification? Second, and more specifically, how can short term amplification that has been reported be reconciled mathematically with a long term failure to see the same amplification. In particular, are there discontinuities or jumps in the record that could explain this?

    The main reason I bring this up is that I believe the way we answer some of these questions should ideally satisfy open-minded readers who are told that the lack of the “hotspot” invalidates current climate sensitivity estimates based on water vapor feedback. I don’t yet believe that these readers will be satisfied with what has been said so far even if the evidence for substantial climate sensitivity is otherwise compelling on the basis of multiple lines of evidence. I can guess that the issues will ultimately be resolved through further corrections to instrumental biases, but that remains a guess, and I’m not sure we can confidentally conclude it’s more than that.

    [Response: I can't tell you how this will be resolved over the next few years. There were clear inhomogeneities in the satellite records in the early 1990s which doesn't help, and radiosonde coverage remains poor in the tropics. I find it very difficult to think of any reason why the moist adiabat works fine for every timescale except decadal. AR5 might show some interesting sensitivities to ozone changes which are being widely incorporated into this generation of models, but I think that clarity will just come with longer records. - gavin]

  25. 225
    David B. Benson says:

    I’ll next post a revised version of

    GLobal warming, decade by decade.

    This is particularly relevant to a thread in which both the AMO and modeling have been themes. This simple, zero-dimensional, zero reservoir (or box) model manages to do quite well the explaining the 13 decades of GISTEMP. Some have called this model 8th grade or reshman. I wish it was so. Anyone introducing aspects of climatology to students in middle school or later is welcome to make use of this study.

    The revisions include fixing some minor typos, adding a reference (Tol & de Vos, 1998) suggested by James Annan and a mention of a slight variation which assumes a pre-existing linear trend, as in the cited paper. In addition, there are updated and additional links to other simple models, aspects of the AMO, etc.

  26. 226
    David B. Benson says:

    Global warming, decade by decade

    First look at
    to observe the general upward trend in global temperatures.
    From well understood physics, we suspect the logarithm of CO2 concentrations (lnCO2) explains most of the variation and indeed from a standard regression
    this is confirmed. Now we want to explain the decadally averaged GISTEMP global temperaure anomaly product (GTA) from the Arrhenius formula which (approximately) explains the warming due to atmospheric CO2 concentrations:
    Physics demands there must be some delay and one decade works well enough. So the formula as applied is, for each decade d,

    AE(d) = F*(lnCO2(d-1) – lnCO2(1870s)) – GTA(1880s)

    where the prior decade’s average lnCO2 provides the forcing with the assumption that the first decade, the 1880s, is unforced by CO2. The final term is the adjustment for the way GISTEMP anomalies are reported and there is a constant F to give the temperature change due to the forcing by lnCO2. This constant is traditionally reported for a doubling of CO2 concentrations, 2xCO2,

    OGTR for 2xCO2 = F*ln(2).

    OGTR stands for Observed GISTEMP Response and is estimated below to be 2.280 K for 2xCO2. Using just this, we have the following table in which the residuals are the differences between GTA and the AE estimate. The standard deviation of the residuals is sd= 0.052 K and the coefficient of determination, R^2=0.953, shows that almost all of the variance in the 13 decades of GISTEMP’s GTA is explained.

    The diffs are the differences lnCO2(d)-lnCO2(d-1). If these were all the same the additional forcing would be the same for every decade. This is approximately so before the 1940s. Notice the dramatic changes from the 1940s on.

    OGTR for 2xCO2 = 2.280 sd= 0.052 R^2= 0.953
    decade GTA — AE — residual diffs
    1880s -0.275 -0.275 +0.000 0.014
    1890s -0.254 -0.231 -0.023 0.007
    1900s -0.259 -0.206 -0.053 0.009
    1910s -0.276 -0.176 -0.100 0.013
    1920s -0.175 -0.135 -0.040 0.012
    1930s -0.043 -0.096 +0.053 0.014
    1940s +0.035 -0.051 +0.086 0.004
    1950s -0.020 -0.038 +0.018 0.009
    1960s -0.014 -0.010 -0.004 0.022
    1970s -0.001 +0.064 -0.065 0.033
    1980s +0.176 +0.173 +0.003 0.043
    1990s +0.313 +0.316 -0.003 0.042
    2000s +0.513 +0.454 +0.059 0.050

    All in all, quite decent for a simplified model based on the physics of the atmosphere plus the shallow ocean, but there is certainly some other effect(s) causing the temperatures to swing (wobble) more widely than can be explained by lnCO2 alone; there is the deep ocean. On the centennial scale of the instrumental record so far, the deep ocean is approximately just a heat sink, but one with a rate which varies on a multidecadal scale. Fortunately, there is a proxy for this internal variability, the Atlantic Multidecadal Oscillation (AMO):
    Although linearly detrended over 150 years, the AMO for the 13 decades of interest has an average of -0.014 which is removed for this decadal study. After that operation as well as decadal averaging, we have AMO(d) for each decade d. Our formula to account for internal variabilty, in addition to lnCO2, is

    AEP(d) = AE(d) + A*AMO(d)

    where A and F are jointly estimated by minimizing the standard deviation. Incidently, the AMO also will include some effects of lnCO2 (as this forcing is not linear in time) and also nonlinear portions of other forcings which affect the North Atlantic. Still, this works well to refine the estimate and better approximate GTA.

    OGTR for 2xCO2 = 2.280 A = 0.335 sd= 0.025 R^2= 0.991
    decade GTA — AEP – residual AMO
    1880s -0.275 -0.253 -0.022 +0.066
    1890s -0.254 -0.232 -0.022 -0.003
    1900s -0.259 -0.243 -0.016 -0.110
    1910s -0.276 -0.240 -0.036 -0.192
    1920s -0.175 -0.176 +0.001 -0.124
    1930s -0.043 -0.041 -0.002 +0.164
    1940s +0.035 -0.010 +0.045 +0.120
    1950s -0.020 +0.013 -0.033 +0.152
    1960s -0.014 -0.004 -0.010 +0.017
    1970s -0.001 -0.012 +0.011 -0.227
    1980s +0.176 +0.145 +0.031 -0.084
    1990s +0.313 +0.323 -0.010 +0.022
    2000s +0.513 +0.521 -0.008 +0.200
    2010s ??.??? +0.686

    Using lnCO2(2000s) and assuming the AMO does not change, there is a prediction for the 2010s of 0.686 +- 0.025 K, lots warmer.

    The autocorrelations of the resdiuals (not listed) show that essentially everthing is explained by AEP, in accordance with the value R^2=0.991.

    Summarizing, CO2 accounts for the centennial scale secular trend with some additional fluctuations due to internal variability.

    (1) The decadal delay in applying the forcing is a simplification of the two box model studied in
    from which I determine that about 11–13 years would be slightly better, but thought decadal averages would be more helpful in showing the essense of the climate response. A one decade delay in applying lnCO2 is better than using a two decade delay.
    (2) Attempts to remove the CO2 diffs from the AMO result in almost no change from the above. Using detrended NH SSTs instead of the AMO produced a somewhat inferior result.
    (3) The OGTR for 2xCO2 of 2.28 K is a transient repsponse roughly corresponding to a Charney equilibrium climate sensitivity (ECS) of about 3.3 K, by one rule of thumb. This just indicates the physics has not been too overly simplified and is not a serious estimate of ECS.
    (4) To the extent that the net of other forcings contributes to the secular trend, OGTR is overestimated. This effect is thought to be small as the IPCC AR4 report concludes, in essense, that the net of other forcings is near to zero.
    This was confimed by including two more parameters for a linear trend, as in [1]. The best fitting trend was 0.063 K/century, highly believable. But then the data becomes
    OGTR for 2xCO2 = 2.111 A = 0.333 sd= 0.022 R^2= 0.992
    so almost no more of the variance is explained, at the price of doubling the number of parameters being estimated. Without including the linear trend the adjusted R^2 is 0.9892 taking into account there are 13 data points and 2 parameters.With the linear trend the adjusted R^2 value is but 0.988 as there is in that case 4 parameters. The smaller value of OGTR obtained suggests that the actual value is strongly sensitive, but less than 10% so, to the assumptions of the study.
    (5) For more on attribution, see
    Attribution of 20th Century climate change to CO2

    (6) For more on the AMO, see

    Sources —
    CO2 concentrations:
    But also

    — David B. Benson

    [1] Tol, R.S.J. and A.F. de Vos (1998), ‘A Bayesian Statistical Analysis of the Enhanced Greenhouse Effect’, Climatic Change, 38, 87-112.

  27. 227
    Didactylos says:

    “The factor of ten in efficiency for the car I am planning comes from the product of 2 and 5. The 2 comes from making the car half as wide using tandem seating, and the 5 comes from adapting the well know airship form that has a drag doefficient that is about a fifth that of most cars.”

    See, Jim: this is why you don’t get taken seriously. I don’t want to derail RC onto a completely unrelated subject, but this is the same problem you have with trees. You simply don’t know how much you don’t know.

    I don’t know much about aerodynamics, but at least I know that I don’t know much about it. On the other hand, I know that you haven’t considered the very unstreamlined wheels of your car. I know that you haven’t considered that drag is only one factor that affects fuel efficiency. I know that you haven’t considered other more convenient aerodynamic shapes. I know that you haven’t considered that drag is proportional to speed squared, so is far less significant at lower (urban) speeds.

    I also know that your response to constructive criticism is to lash out and deny that there is a problem, so I imagine you will never get any further with your ideas.

    And that’s sad.

  28. 228
    Marty Singh says:

    Gavin, Re your reply to 208.

    The supposed missing hot- spot would affect the water-vapour feedback to the extent that the upper atmosphere would have a lower saturation vapour pressure and thus could ‘hold’ less water than if the changes were moist-adiabatic. Do you know if there is a simple calculation (or if anyone has done this with a radiation model) for which effect is larger?

    The calculation I would envision would be to take a Radiative-convective equilibrium profile and then increase the CO2 and change the temperature profile until the TOA fluxes are balanced again. The temperature change would be constrained by fixed relative humidity, and in one case to be moist adiabatic, and in another case to have the delta T be the same throughout the profile (with some allowances for what happens at the tropopause).

    I had at one point convinced myself, using simple emission level arguments, that the water-vapour feedback wins out, but I can’t for the life of me remember what I did.

    Of course the whole thing is a bit artificial, as there is no reason to suspect the upper-tropospheric relative humidity must remain constant, especially if the troposphere is not staying on a moist adiabat.


  29. 229
    Septic Matthew says:

    204 t_p_hamilton,

    Thank you.

  30. 230
    Septic Matthew says:

    201, comment by Jim: [Response: And the idea that you can essentially turn all the world's land into forest if you just move enough water around...Jim ]

    “essentially all the world’s land” is a straw man. No one has even claimed that “all” of Eritrea could be afforested with salt-tolerant mangroves, only enough to make an important improvement.

    Do you know of any solutions to global warming that do not entail “massive rearrangements”? The proposal to eliminate that 45% of US electricity that is generated from coal, and replace it with some other electricity, is a “massive rearrangement”, wouldn’t you agree?

  31. 231
    Patrick 027 says:

    Jim Bullis –

    adding to Didactylos’s 227, I had thought that rolling resistance was half of the total resistance at 55 mph; this can be reduced by reducing weight. Electric cars could have a motor for each wheel, which could dispense with some of the resistance in the transmission, though I’m not sure how much of a difference that would actually make.

    Of course, EVs, HEVs, and PHEVs don’t need to run the engine while idling, which is really a whole other aspect…

    If you want to submit your car to the market, fine with me. It would require a lifestyle change for people, though, just as some other ideas would. I like the option of riding in a car 2 by 2. I’d be willing to sacrifice that to save the planet, of course, but there are other things I could do instead, which I might prefer when it comes to my own lifestyle (every little bit helps, and it’s good to have options, and we may need to use many, but you don’t need ‘every’ little bit, just enough bits). I’m not hostile to your idea, but while your car may have some market share but I wouldn’t expect it to dominate anytime soon. I think you could change the shape of the car and still get some savings in aerodynamic drag without having to limit the width to fit one seat per row.

    How much of the heat generated by your car’s engine can be utilized?

    If your car’s efficiency can be so much improved, that apply to an PHEV or PEV with the same improvements.

    Of course, improvements to any energy conversion in use are to be welcomed.

    Wind, hydroelectric – not much of a waste heat issue with centralized or distant power. CSP, CPV, PV power plants – some heat might be wasted but you have to consider the trade-offs of different locations. CSP could be used for heat for some industrial processes and also with community heat systems. How efficient are community heating systems? There are also ways to reduce heating needs. Household furnaces might be made into cogeneration plants using TPV; there is also direct solar heating (passive solar – just need to make sure the windows and skylights are well-insulated) A lot of energy can come through a window. Rooftop PV can be made into cogeneration by heating (or preheating) water. Solar water heating is a great idea. Note use of cogeneration plants – in cars or otherwise – may have reduced usefulness in warmer climates, depending on the economics and volume of solar water heating and its possible pairing with PV.

    Some efficiency improvements really don’t add much net cost. Maybe see work by Stern.

    We don’t need to dot all the i’s and cross all the t’s – if we have good policy, the market will do a lot of the sorting out (I think it would be funny if not so sad that so much of the Laissez-faire crowd is so dead set against the use of the market to solve a problem!)

    The most efficient use of a mass of peridotite for CO2 sequestration may involve concentrated CO2, but I have wondered if crushing the rock and dispersing it could economically work to take CO2 from the air (thus divorcing it from CO2-producing activities) (this could also somewhat double as an albedo contribution, if the mineral dust is allowed to go windborn over the oceans, and then gets into the oceans, buffering the pH while taking up CO2); some time ago another commenter at RC posted some energy-use info about rock-crushing which looked promising. Presumably this could also be done with some effect with more common rocks, so long as there are Ca silicates involved – or even some other cations (?), or even in carbonates, that wouldn’t allow carbonate rocks to form but it would help buffer the pH and help the oceans take up more CO2 without needing to dissolve the corals and seafloor sediment, etc.)

    Now going off on a much less realistic solution: candy-wrappers and potato chip bags (commonly a high-albedo inside surface, at least in the visible portion of the spectrum) – how much area would they cover (and importantly, what is their LW emissivity, because a low LW emissivity could be counterproductive).

  32. 232

    227 Didactyolys

    Also in your comment, you clearly think urban speeds are a big deal.
    They are not. Most miles are not urban miles. If we are trying to reduce CO2, urban miles are not all that significant as a part of the picture.

    I am happy to let someone else fix the urban efficiency problem. That will be a small thing which will make people feel good, but will accomplish little.

    Also, the picture is never all that simple, but if one wants to say anything concisely, one has to ignore a lot of detail. The question is, is the big issue addressed.

  33. 233
    adelady says:

    There’s a bit of a difference with re-afforestation, new ‘forests’, using water and transporting water. My own initial gradualist approach is to the suburbs of major cities. Most of these areas are covering the best agricultural land – the reason the place was originally established as a good place to live. Therefore a prime area for tree planting as well as growing food.

    These cities all have “stormwater” systems. Why water that falls on roofs and other domestic hard surfaces should be treated as a waste disposal problem rather than a best use problem is an eternal mystery to me. (I admit I live in the driest state in the driest inhabited continent, but hey ho.) Redirecting water towards the growing area of one or several trees reduces the “waste management” aspect of rainfall and grows a carbon sequestration unit which might also produce food.

    I don’t know how many suburban blocks there are in USA, but putting 1 or 2 or 3 additional trees within the blocks or on the street verges would use “waste” water and achieve a result. Someone could do some numbers to work out the optimum number if 30 or 40 or 100 million trees is the target. But this approach is nowhere near what’s required even though it would be helpful.

    Reducing the use of geological carbon resources is the best way. And USA has more scope for making a big impact by efficiency improvements alone than practically any other country.

  34. 234
    Patrick 027 says:

    Re 208,224 Fred Moolten -

    - I think we can observe H2O vapor mixing ratios (a measure of specific humidity) independently of observing the ‘hot spot’. Including measurements from satellites.

    How might H2O vapor concentration increase without a hot spot? (PS I don’t really have a conclusion to answer your questiong, just a bunch of interesting information to consider:)

    1. first, consider an atmosphere with no precipitation from clouds; the air moves around, occasionally coming near a wet part of the surface and equilibrating it’s water vapor pressure with something proportional to the the equilibrium vapor pressure at that temperature, with some adjustment required by molecular diffusion and eddy mixing and the temperature gradient (that’s actually something I’d have to review again before I could get any more detailed). The total water content of the air would then tend to change with the equilibrium vapor pressure at the surface, with air aloft (above the cloud base) having some portion of that water in condensed form. In this alternate reality, the moist adiabatic lapse rate could be ‘the’ adiabatic lapse rate for the majority of the atmosphere (pseudoisentropic surfaces would become truly isentropic surfaces).

    2. Where does dry air come from? Consider a ‘conveyor belt’ model of overturning where air rises with H2O condensing and some precipitating out. It then reaches some height, moves laterally, and sinks (PS some more complex motions might be folded into such a depiction – for example, upper level air could gain cirrus clouds from deep convection, then mover around, sink a little and clear up, then rise a little with some condensation/deposition again, sink, as it moves through Rossby and gravity waves, or have clouds appear and disappear moving across lines of latitude with radiative heating and cooling). The remainging cloud water evaporates with adiabatic warming; as some was removed by precipitation, the cloud base in the sinking region is higher and the H2O vapor concentrations below that are smaller. If the efficiency of precipitation were increased or reduced (PS this is an aspect of a (so far as I know) speculative negative feedback proposed by Lindzen – of course you know what tends to happen with that category of phenomena, but it’s interesting to consider the physics involved anyway.), then there would be less or more water vapor in the regions of sinking air. What if the cloud base in the ascending region were raised or lowerd? Of course, that would (other things being equal) require a decrease or increase in relative humidity in the surface air underneath. But if the height of the cloud base were raised, that would reduce the ‘hot spot’ by having a greater thickness of air tend toward a dry rather than moist adiabat. Would the cloud base change if relative humidity underneath were fixed while the surface temperature was increased?

    (PS precipitation would actually slightly (and I mean slightly – you can ignore the liquid water heat capacity without much error) decrease the moist adiabatic lapse rate below the freezing level by removing some mass of liquid water, thus reducing the specific heat of the air parcel. However, then there is less latent heat available from freezing of that water (directly or via deposition of vapor onto ice crystals) – though that is roughly an order of magnitude smaller than the latent heat of condensation, per unit mass of water.)

    But the way water vapor is determined is more complex than that because there is mixing (cumulus clouds commonly entrain some dryer air, reducing the total water per unit air in the cloud), … (clouds can precipitate into layers of air beneath with subsequent evaporation).

    After evaporation is completed, dry sinking would follow a dry adiabat, except for radiational adjustment of temperature. Sinking is often much slower and over a much larger area, with radiative cooling being able to significantly affect temperature and even balance adiabatic warming (as in a steady-state Hadley cell) while moist updrafts are faster and concentrated and with radiation having essentially no role in changing the temperature within a cumulus cloud. But lapse rates larger than the moist adiabatic lapse rate can develop and remain until a trigger gets the moist convection going (PS although in the tropics I’ve gotten the impression that gravity waves tend to communicate the effects of moist convection so that larger regions adhere to a moist adiabat at the same vertical levels – which would make sense given a weak coriolis effect that would allow horizontal temperature variations to persist in the face of the tendency for thermally-direct overturning). This introduces some wiggle room.

    Meanwhile, an increase in the greenhouse effect (note, regional drying or reduced cloud cover, especially low cloud cover, would have the opposite effect) would tend to decrease the diurnal temperature range over land; since convective heating of the bulk of the troposphere tends to come somewhat preferentially from the warmer (and more humid, of course) times and places, reducing horizontal and temporal surface temperature variations might be expected to increase the average lapse rate near the surface (because the average lapse rate includes contributions from times and places with stable air, and such an evening-out of temperature could tend to bring the air in general closer to having neutral stability). In particular, radiative forcing or feedback, either LW or SW, at the surface, can warm the surface without an immediate convective adjustment through the bulk of the troposphere at places and times with stable air masses at the surface – such as at higher latitudes, particularly in winter (now remember the sea ice albedo feedback).

    Horizontally-large scale overturning of the troposphere is driven, and in part, organized, by differential heating both horizontally and vertically (it would become more sluggish without the heating below and cooling above, but can continue with horizontally differential heating alone, and can sustain a lapse rate that is stable to localized convection of all types). Reducing the horizontal differential heating would tend to slow this type of circulation, which would tend to allow a larger lapse rate to develop and favor more localized overturning.

    Meanwhile, if the sinking dry air spends a lot of time over dry land and a little time over a wet surface before rising to various heights with moist convection with precipitation, then the average humidity would be less, relative to the case where the sinking air gets humidified soon and waits a longer time before rising with moist convection and precipitation.

    Wind helps speed evaporation (eddy diffusion, or in the case of horizontal variations, dry air advection). On the other hand, still air near a wet surface could over time get closer to 100 % RH (but in a thinner layer/smaller region! However, if the supply of dry air to the boundary layer were slowed…). Meanwhile, molecular diffusion should occur faster at higher temperatures (right?), so RH should tend to increase with increasing temperatures, other things being equal, … but I’m guessing that would be a very small effect.

  35. 235
    Hank Roberts says:

    SM, forget global warming, focus on ocean pH change — it’s faster and simpler.
    If we lose use of the oceans, we won’t worry about the atmosphere.

  36. 236
    Gilles says:

    BPL : thank you for the very clear demonstration that you can get have a fairly well determination of the influence of CO2 on temperature , with a simple spreadsheet and basic statistical tests. But I have a question ; if the sensitivity is already fairly well determined with your simple statistical tests, at a ridiculous computer and human cost, what do you expect from much more detailed ( and enormously more expensive) heavy computer simulations ? which uncertainty could be much more reduced with them, since there is already hardly any , FAPP ?

  37. 237
    meteor says:

    thanks Gavin

    sorry, I misunderstood this sentence:

    “in experiments where atmospheric constituents (including water vapor, clouds, CO2, O3, N2O, CH4, CFCs, and aerosols) were added to or subtracted from an equilibrium atmosphere”

    I am not a good reader…

    [Response: This refers to the earlier set of experiments (also discussed in Schmidt et al, 2010) where each constituent was zeroed out in turn while everything else was held constant. Different kind of thing entirely. - gavin]

    But what is your opinion concerning the sharp rising of the cloud cover in the beginning of the experience in fig 2?

    [Response: Big expansion of low cloud cover. - gavin]

  38. 238
    CM says:

    BPL #207, Patrick #213 — thanks for that. Ought to have seen it myself.

  39. 239
    CM says:

    More plugins: This reCaptcha replacement could come in real handy.

  40. 240

    The switch gets thrown today on a small “closed-loop” energy and food project that actually seems to show some useful potential:

    It’ll be interesting to see if they’re able to pull off the aquaculture, hydroponics and algae production as planned. The only inputs are said to be a) sunlight, b) landfill gas (mostly methane), and c) some fish food.

  41. 241
    Didactylos says:

    Jim Bullis: if you want to contribute something these days, it usually has to be in the detail. Yes, there are completely novel ideas out there. But lower drag coefficients, tandem cars and afforestation are not among them.

    This doesn’t mean that there isn’t a huge amount you can do (in the detail, and in the practicalities of turning vision into reality).

    But detail means you can’t be a “big picture” guy. You can’t just walk in and make sweeping statements that are almost right.

    For example, you say: “Most miles are not urban miles.” That’s not true. In the US, 60% of driven miles are urban vs. rural [Federal Highway Administration].

    Please note that the people commenting on your approach are not deniers: we are all in favour of major action to combat global warming. Also, you’re not wrong – afforestation is generally a good thing, and aerodynamics really is an area which can have a big impact on fuel efficiency. Have a look at some solar vehicles – they use a superstreamlined form, and the wheels are barely protruding to maximise the advantage. They are long and thin. They genuinely have a tiny drag coefficient, and it helps a lot.

  42. 242
    Ray Ladbury says:

    What you get from more precise models is, of course, more precision. In particular, you narrow the confidence interval on CO2 sensitivity. Models have been instrumental in narrowing the 90% CL for sensitivity to 2-4.5 Kelvins/doubling. This is extremely important on the high side, as the adverse consequences of warming increase dramatically (exponentially, perhaps?) with sensitivity.

  43. 243
    Mike G says:

    @194 Hank,
    It’s not that fragmented forests attract less rain, but they lose more water.

    Google “edge effects”

  44. 244

    I was fortunate to attend Sam Harris’ talk yesterday at Caltech. The subject was his new book: “The Moral Landscape How Science Can Determine Values. It’s sad when I see scientists talking about how bad things can get re global warming. They get ridiculed and threatened. I think if we REALLY know how bad things can/perhaps will get we should be talking about it to the public. And many are thanks goodness. But then they will be referred to as crazy “alarmists”.

    I guess my point is I want to thank all of these scientists for attempting to get the information out there. I am a layman and while I am more educated than most laymen on this subject I am far from where many of you guys(and gals) are.

    Anyway, thank you all for having the passion that you do!


  45. 245
    Septic Matthew says:

    233, adelady:
    There’s a bit of a difference with re-afforestation, new ‘forests’, using water and transporting water. My own initial gradualist approach is to the suburbs of major cities. et seq.

    I agree.

    I think that the solution to global warming is some of everything that is useful, gradually implemented everywhere, over decades. Most of what is useful to solving global warming is valuable in other respects, and should be pursued even if global warming isn’t occurring. Harvesting more wastewater is one of the practices that should be adopted more widely, and used to irrigate varieties of species. This is increasingly common throughout California, to pick one place, and waste water is used to irrigate golf courses, palm groves, and avocado groves.

  46. 246
    Septic Matthew says:

    235, Hank Roberts: SM, forget global warming, focus on ocean pH change — it’s faster and simpler.

    I disagree for two reasons: (1) global warming has been prominently discussed, and many methods for reducing CO2 have already been promoted and started to reduce it; (2) pH reduction from 8.1 to 7.7 or 7.3 isn’t (pace Jane Lubchenko) going to dissolve chalk, much less animal skeletons.

    Rhetorically, the continual changing of the alarm from “cooling” to “overpopulation” to “resource constraints” to “unsustainability” to “warming” to “change” to “disruption” to “biodiversity” to “ocean pH” contributes to the perception that scientists are in a continual state of alarm about everything, or else simply misanthropic. Everything known to reduce anthropogenic CO2 is good for other reasons (as far as I am aware), and can be pursued to achieve multiple goals.

  47. 247
    Hank Roberts says:

    MikeG, here’s the cite and abstract again, as posted above. It’s not a restatement on edge effects; it’s a suggestion about looking at an additional possible effect.

  48. 248
    Hank Roberts says:

    > SM
    > gradually, over decades

    We know that won’t work. Why prefer it?

  49. 249
    Ray Minjares says:

    Can you help me (and us) understand the implications of the recent review article on semi-direct effects of aerosols?

    Here is the link:

  50. 250
    Hank Roberts says:

    PS for Mike G — not arguing against edge effects at all; those have many strong effects. Just noting there’s some possibility that the authors have a reasonable hypothesis to add.

    Here’s an earlier paper along the same line. It’s not a strong effect, but may emerge from statistics:
    Webb, Thomas J., F. Ian Woodward, Lee Hannah, and Kevin J. Gaston. 2005. FOREST COVER–RAINFALL RELATIONSHIPS IN A BIODIVERSITY HOTSPOT: THE ATLANTIC FOREST OF BRAZIL. Ecological Applications 15:1968–1983. [doi:10.1890/04-1675]
    “… significant positive relationships between tree cover and the number of rain days consistently emerge. The degree of forest fragmentation seems to influence this relationship, with patchier forests associated with fewer rain days; and tree cover also predicts interannual variability in rainfall.”

Switch to our mobile site