Andy Lacis responds to Steve Koonin

credit : xkcd

credit : xkcd

I know Eli’s already posted Andy Lacis’s response to Steve Koonin on Judith Curry’s blog, but I thought it worth repeating. It’s a pretty impressive comment in terms of what it covers, so it’s worth reading in it’s own right. I do find myself amazed at what Steve Koonin has been willing to say. Ignoring that much of what he says suggests a woeful lack of understanding of the topic itself, that anyone of his supposed intellectual calibre would construct an argument that essentially goes “look, this number is small, nothing to worry about” is remarkable, and not in a good way. It’s one thing to suffer from hubris, but it’s hard to see why if one’s argument is so obviously silly. Maybe Eli’s right that the best description is beyond contempt.

Anyway, Andy Lacis’s comment is below (bolds mine).

Physicists should take the time to understand their physics better (Comment: some of us are trying 🙂 )

Only 1% to 2% . . . that may sound small and insignificant . . . but it isn’t.

It is well known that the normal human body temperature is about 310 K. Furthermore, it is also well known that a seemingly small change (up or down) in absolute body temperature by only 1% (3.1 K, or 5.6 F) would make one sicker than a dog, and, that a 2% change in body temperature (up or down by 6.2 K, or 11.2 F) will virtually guarantee a dead body. From this, it should be sufficiently clear that, when viewed in absolute energy terms, the viable margin between life and death in the Earth’s biosphere is remarkably narrow – so much so that a seemingly insignificant 1% to 2% change in the total energy of the global environment will invariably result in serious disruption of the established infrastructure of life in the biosphere.

There is no substitute for appealing directly to basic physics for physical insight and better understanding of the ongoing global warming problem. And I do recall one particular case in the 1970s (in which you might have participated) when the JASON group of physicists was tasked to weigh in on the then open question of radiative forcing due to doubled CO2. At that point in time, the JASONs had available the computational resources to calculate one of the earliest line-by-line radiative forcing determinations for doubled CO2. They found the downward flux change at the ground surface to be less than 1 W/m2, from which they erroneously concluded that the radiative forcing caused by the doubling of atmospheric CO2 was “not all that significant”.

While the JASON group’s radiative calculations were numerically on target, the JASONs were clearly mistaken in their interpretation of the calculated results. Radiative forcing takes place over the entire atmosphere, and not just at the ground surface. If they had to select a single point on the vertical profile that best describes the radiative forcing by CO2, they should have selected the tropopause point, where the instantaneous flux change due to doubled CO2 is nearly 5 W/m2 for a clear-sky atmosphere. Moreover, the JASONs did not take into account the additional radiative magnification that is invariably contributed by the longwave opacity from water vapor and cloud feedbacks, which are several times larger than the radiative forcing due to CO2 alone, and therefore should have been included in their analysis.

In simple terms, the basic essence of the global warming problem is best understood as a straightforward problem in global energy conservation (Comment : I like this, because this is precisely how I normally think of this issue), as was first noted by Joseph Fourier in 1824. Specifically, the global-mean surface temperature of the Earth is about 288 K, which implies that the Planck emission from the ground surface must be about 390 W/m2. Furthermore, the global-mean solar energy absorbed by the Earth is observed to be about 240 W/m2 (with about 100 W/m2 reflected directly back to space).

Given that the Earth should be in near-global energy balance, this implies that the Earth must radiate about 240 W/m2 of longwave energy out to space (as has also been verified by satellite measurements). Absent the greenhouse effect, the 240 W/m2 of absorbed solar energy can only support a surface temperature of 255 K. This “missing energy” circumstance led Joseph Fourier to conclude that there must be thermal heat energy radiated downward from the atmosphere to supply the additional heating of the ground surface.

The flux difference of 150 W/m2 between the 390 W/m2 emitted by the ground surface and the 240 W/m2 of LW flux going out to space at the top of the atmosphere is a direct measure of the strength of the terrestrial greenhouse effect. Greenhouse action builds up the surface-emitted flux to 390 W/m2 and creates the ensuing reduction by 150 W/m2 of the outgoing longwave flux to space – all accomplished by radiative energy transfer means (via sequential emission, absorption, and re-emission interactions).

Physicists should also appreciate the nature of the Clausius-Clapeyron relation, and the fact that it is exponential in temperature. Undisturbed, with a source of liquid water, the atmosphere is always striving to reach an equilibrium 100% relative humidity. In simple terms this means that the holding capacity of the atmosphere for water vapor doubles for every 10 K increase in atmospheric temperature. And, there is no doubt that water vapor is a very potent greenhouse gas.

Detailed radiative attribution calculations show explicitly that water vapor accounts for about 50% of the 150 W/m2 of greenhouse effect, and that longwave cloud opacity accounts for 25%. Both of these radiative effects are due to the climate system’s fast feedback processes. The remaining 25% of the greenhouse effect comes from the radiative forcings by the non-condensing greenhouse gases (which incidentally also act to sustain the terrestrial greenhouse effect at its present strength). Of the non-condensing greenhouse gas contributions, CO2 is by far the strongest contributor accounting for about 20% of the 150 W/m2 greenhouse effect, with the remaining 5% due to minor GHGs like CH4, N2O, O3, and CFCs.

A key point to keep in mind is that it is these non-condensing greenhouse gases that act as the principal radiative forcing agents of the climate system. Because of their thermodynamic, chemical, and radiative properties, CO2 and the minor GHGs are chemically slow-reacting with atmospheric lifetimes ranging from decades to many centuries. Once they are injected into the atmosphere these gases effectively remain there indefinitely by not condensing or precipitating at prevailing atmospheric temperatures as they continue to exert their radiative forcing.

Since CO2 is the strongest and most effective of these non-condensing radiative forcing gases, it then follows that CO2 can be identified as the principal LW control knob that governs the global climate of Earth. The fact is that the other forms of radiative climate forcing (e.g., changes in solar irradiance, surface albedo, and aerosol forcing) are small by comparison. This makes the case for recognizing CO2 as the principal climate control knob that much more compelling.

Atmospheric water vapor, on the other hand, has the role of principal fast feedback process in the climate system by condensing and precipitating from the atmosphere in response to changes in local meteorological conditions (constrained by the exponential temperature dependence of the Clausius-Clapeyron relation), meaning that the atmospheric distribution of water vapor (and clouds) can change rapidly on a time scale of hours and days in response to changing weather conditions.

Applied radiative forcings that heat (or cool) the atmosphere cause more (or less) water vapor to evaporate, which generates more (or less) longwave opacity, which then contributes more (or less) radiative greenhouse effect. Such changes in water vapor cause big changes in radiative heating or cooling, but the changes are limited in magnitude by how much change the water vapor undergoes in reaching its new equilibrium distribution.

Because of this, water vapor and clouds act to magnify the initial radiative perturbation, but cannot on their own initiative manufacture or impose a warming or cooling trend on global climate, even though they contribute more strongly to the atmospheric radiative structure than the radiative forcing gases that actually drive and control the global temperature trend.

The physics cause-and-effect nature of the global warming problem is not all that complicated. The basic “cause” component of global warming has been clearly identified and understood for many decades, and has been accurately quantified by precise measurements of atmospheric CO2 (e.g., the Keeling curve).

This is fully corroborated by the latest annual data report of fossil fuel extraction that now approaches 10 gigatons of carbon/yr (roughly equivalent to 10 cubic km of coal/yr, which when burned, adds about 5 ppm CO2 to the atmosphere, half of which remains there for many centuries). The radiative effects of CO2 are fully known from well-established understanding of greenhouse gas radiative properties and radiative transfer modeling of the atmospheric structure.

How can a physicist not comprehend that it is atmospheric CO2 that is the principal radiative forcing agent for the ongoing global warming? . . . and not be concerned that water vapor, as the climate system’s principal feedback agent, has an exponential dependence on temperature?

To be sure, there are other factors that contribute to climate change. But decades of measurements and analysis have shown that variations in solar irradiance, land use, aerosols, ozone, and other minor greenhouse gases, while making a contribution, are small by comparison to CO2.

Of greater interest is the “unforced” variability of the climate system on decadal time scales that arises from changes in ocean circulation patterns that are effectively un-influenced by changes in atmospheric radiative forcing. The deep ocean is a very large cold storage reservoir. An upwelling blob of cold ocean water can put a “pause” in the ongoing global warming, temporarily diverting the greenhouse “heat” to warming the ocean. But once that cold blob of ocean water has been warmed up to its equilibrium temperature, it is back to the business of continued global warming. And also note that the ocean cannot cause a decadal warming spurt – the deep ocean is colder than the surface biosphere, so it cannot be a source of heat.

Significantly, the key climate system components (water vapor, clouds, ocean) are not configured to respond to radiative and/or temperature perturbations on a sufficiently small enough incremental scale that would permit a monotonic approach to global energy balance equilibrium. Instead, there is always over-reaction such as when water vapor condenses en mass to produce storms, coupled with the similarly over-reactive responses by atmospheric and ocean dynamics to pressure-temperature and salinity differences, to produce the quasi-chaotic weather and the longer-term climate noise that characterizes the climate system.

Physicists should not be confused by these random-looking quasi-chaotic fluctuations about the local climate equilibrium point, and should instead focus more on the changing energy balance equilibrium point of the climate system. They should also pay attention to the geological record that points to an atmospheric CO2 level of 450 ppm as being incompatible with polar ice caps, a level that is expected to be reached by the end of this century. While it may take a thousand years for the polar ice to melt, the future course is being prepared for a 70 meter rise in sea level.

Advertisements
This entry was posted in Climate change, Climate sensitivity, ClimateBall, Judith Curry and tagged , , , , , , , . Bookmark the permalink.

279 Responses to Andy Lacis responds to Steve Koonin

  1. GSR says:

    CE, Rabbett and now ATTP.
    Each time I’ve read Andy Lacis’ piece it’s as though each site has somehow improved it over the last. When I first read it on CE I was very impressed. When I read it at Eli’s I was pretending that I had written it.
    By the time it got to ATTP’s iteration, I was delivering it off the top of my head to the George. C Marshall Institute.

  2. Yep, it’s a beauty -a model of clarity. I saved a copy to my reference library.

  3. John Hartz says:

    ATTP: Thanks for posting this. I am recommending that we repost on SkS.

  4. izen says:

    The Andrew Lacis post is certainly an exemplar of clear, elegant and accurate communication of the known science, but the absence of such statements is not the cause of the opposition to this known science.

    I recently posted that Dr Curry was annoyed by the APS draft statement because she had no more involvement in the process than suggesting names for the expert advisors to the POPA group. However I have since been reading, or at least skimming, the minutes of those expert meetings that both ATTP and Eli have linked to;-

    And find that Dr Curry was actually involved as an expert witness, along with Lindzen. From what little I have read so far she did not make a strong case for a statement other than that produced, she seems to have spent much of the time promoting the Stadium Wave and decrying the inadequate science in the summaries for policy makers, whilst ignoring the science in the WG1 section of the IPCC reports.

    Dr Curry must have been exposed to clear analysis of the issue like that of Andy Lacis, but that does not seem to dent her, or Koonin’s enthusiasm for minimising the probable impacts at least.

    The other impression I get from the transcripts of the expert meetings is a turf-war concern. The idea that climate change is a PHYSICS core problem is put forward several times with the role of chemical and biological sciences as mere bit players. The motive for the statement to some extent seems to have been a drive to get more funding for physics on the back of the claim that AGW is a pure physics problem.

    There is a justification for this. The classic line that physics models complex systems as a perfect sphere in a vacuum happens to be justified, and materially accurate for the Earth.

  5. izen,

    The motive for the statement to some extent seems to have been a drive to get more funding for physics on the back of the claim that AGW is a pure physics problem.

    Interesting point. Personally, I think the bits that are less well understood (well, by me, at least) are the impacts of climate change, rather than the physical basis for it. I’d like to see more about the impact to the biosphere, for example, rather than more being spent on trying to pin down fine details of the underlying physical processes.

  6. John Hartz says:

    izen: You seem to be ignoring ocean acidifcation in your closing statement..

  7. Eli Rabett says:

    Don’t forget to submit your #Kooninisms

  8. Willard says:

    Well, if we’re discussing clarity, I know an editor or two who would cringe on this one:

    Significantly, the key climate system components (water vapor, clouds, ocean) are not configured to respond to radiative and/or temperature perturbations on a sufficiently small enough incremental scale that would permit a monotonic approach to global energy balance equilibrium.

    The emphasized expression only works for Grrrowth talk.

  9. Given that the Earth should be in near-global energy balance, this implies that the Earth must radiate about 240 W/m2 of longwave energy out to space (as has also been verified by satellite measurements).

    Should? This is personification. The earth will emit energy based on the temperatures and the range of vertical profiles of temperature and longwave emitting gasses. If those emissions are less than received energy, energy will accumulate, temperatures and emissions will rise.

    Physicists should also appreciate the nature of the Clausius-Clapeyron relation, and the fact that it is exponential in temperature. Undisturbed, with a source of liquid water, the atmosphere is always striving to reach an equilibrium 100% relative humidity. In simple terms this means that the holding capacity of the atmosphere for water vapor doubles for every 10 K increase in atmospheric temperature. And, there is no doubt that water vapor is a very potent greenhouse gas.

    Striving? More personification. Conservation of mass means that for any rising motion, there is compensatory subsiding motion. Most of the rising motion takes place in narrow bands of cold fronts and the inter-tropical convergence zone. That means the majority of earths surface area incurs subsiding motion, which acts to reduce the relative humidity. Evidently, more of the atmosphere is striving to have 0% relative humidity as seen in this 500mb chart:

    Further, water vapor is radiatively active, does increase the opacity of the atmosphere in the IR, and does impose a net radiative gain for the lower troposphere. It also makes the top half of the troposphere incur a net loss of energy which leaves the water vapour feedback in question.

    Atmospheric water vapor, on the other hand, has the role of principal fast feedback process in the climate system by condensing and precipitating

    All this should comfort Lacis and everyone. Water vapour is by far the largest conceived feedback and is a fast feedback. Whatever the water vapour response, it is already part of the warming to date. And yet the warming to date is commensurate with about 1.6K per nominal doubling of CO2. If Lacis and everyone have faith in radiative models, they should be relieved since forcing rates are less than the low end scenarios.

  10. Joshua says:

    Not on twitter – but it seems to me the obvious one would be: Only a small percentage of scientists are unconcerned about CO2 emissions – therefore there’s no reason to pay any attention to their views

  11. Eddie,

    Should? This is personification. The earth will emit energy based on the temperatures and the range of vertical profiles of temperature and longwave emitting gasses. If those emissions are less than received energy, energy will accumulate, temperatures and emissions will rise.

    That’s the point. If you’re going to appear to disagree with someone, maybe you could say something that actually contradicts what they’re saying, rather than confirms it.

    And yet the warming to date is commensurate with about 1.6K per nominal doubling of CO2. If Lacis and everyone have faith in radiative models, they should be relieved since forcing rates are less than the low end scenarios.

    Only true if feedbacks are linear, internal variabilty has provided no cooling, aerosol forcing is indeed less negative than -1 W/m^2, etc. Going, “pheww, looks like it will all be fine” is only really sensible if it turns out that all the constraints that need to be met in order for that to be true, are actually correct.

    As I said to you before, you seem to be looking at this from a very optimistic perspective. That’s nice for you, but only worth doing if reality actually turns out as your optimistic perspective indicates.

  12. Willard says:

    > Should? This is personification.

    This would mean that counterfactuals like “if we accept the laws of physics” amount to personification.

    ***

    Please acknowledge that you were Lucifer, Eddie. I won’t ask twice. Instead, I will start to cite sources where you might have used your graphs.

    Thank you again for your concerns.

  13. Kevin ONeill says:

    Eddie, the word ‘should’ does not anthropomorphize a statement. One of should’s common definitions is to indicate what is probable. If you’re going to play parse-omatic at least try to be correct.

  14. Willard – yes of course.
    Restrictive logins now require a WordPress or some such, though not here at ATTP.
    Rather than maintain ‘Lucifer’ here and a WordPress elsewhere,
    I chose Eddie Turbulence as a hat tip to science rather than the devil’s advocacy of Lucifer.
    I have nothing to gain other than use of the WordPress blog that trends kind of forced me into.
    It should matter more about the veracity of what I write ( which I strive to reference and demonstrate with data ) rather than who I am, particularly in an anonymous forum such as this.

  15. Joshua says:

    Eddie –

    ==> “It should matter more about the veracity of what I write ( which I strive to reference and demonstrate with data ) rather than who I am, particularly in an anonymous forum such as this.”

    That seems fair enough. Hopefully you will be consistent with that approach also.

    ==> “I have nothing to gain other than use of the WordPress blog that trends kind of forced me into.

    Could you explain more about how “trends kind of forced [you] into” posting under more than one handle at this blog?

  16. Willard says:

    Thank you, Eddie.

    Bear in mind that pseudonymity is not the same thing as anonymity. A strict anonymity rule would forbid that any comment be undersigned. I would play by this rule, but then precedents indicate that this may lead to an “information cascade,” to borrow the ClimateBall’s word of the day.

    Also note that once you adopt a pseudonym, you are bound by it. This implies that you should publish everything under that name, at least insofar as it’s within the ClimateBall microcosm. I hope I don’t need to clarify what I mean here.

    Again, thank you for your concerns,

    W

  17. BBD says:

    Turbulent Lucifer

    Whatever the water vapour response, it is already part of the warming to date. And yet the warming to date is commensurate with about 1.6K per nominal doubling of CO2. If Lacis and everyone have faith in radiative models, they should be relieved since forcing rates are less than the low end scenarios.

    I think you are mistaken to say that WV feedback is fully expressed in current GAT because we are not at equilibrium to 400ppm CO2 yet.

  18. Mike M. says:

    I think this post is intellectually dishonest. Does Koonin say, in either his post on Climate Etc. or the original Journal article, that there is nothing to worry about? Does he say that anthropogenic effects are negligible? Or insignificant? If he does, please quote the words, because I can’t find it.

    What he does say, quite correctly, is that the effect is small. “Small” does not mean either “negligible” or “insignificant”. In the context he uses the word, it means “hard to detect”. He makes this quite clear in the Climate etc. post: “Since the climate system is highly variable on its own, that smallness sets a very high bar for confidently projecting the consequences of human influences” and “because life at the 1% level is rich, the models have to get many small phenomena right to confidently isolate and project the response to anthropogenic effects”. Both statements are spot on. And he clearly makes the case that the science is most definitely NOT settled as to quantifying anthropogenic effects.

    One can legitimately debate what to do in the face of such uncertainty. But that is not what ATTP does. He reads what he wants to into Koonin article and then condemns what he imagines is there. You can not have a reasoned debate when you refuse to listen to the other side.

  19. Mike M. says:

    At 1:39 p.m. Izzen wrote: “I recently posted that Dr Curry was annoyed by the APS draft statement … Dr Curry was actually involved as an expert witness, along with Lindzen. From what little I have read so far she did not make a strong case for a statement other than that produced”

    From the workshop transcript starting on page 517:
    “DR. CURRY: Personally, I don’t think the scientific societies should make statements about those kind of, what I would call public policy that is not related to the policy of science like we need more observing systems and things like that. … you should only speak to where your expertise is” … I am not even sure that APS has sufficient expertise on the climate issue to be making a statement at all.”

  20. MikeM,
    The title of Koonin’s post is “Are human influences on the climate really small?” His post is a list of reasons why the answer is yes.

    You can not have a reasoned debate when you refuse to listen to the other side.

    Well, you certainly can’t have one if one side won’t even accept a position that they almost certainly hold.

    I’ll add an extra point. This seems to be a standard tactic. Say/write something that then gets criticised. Then deny that what was said was what the critic claims was said, while it being patently obvious that it is what the critic claims was said. How else is one meant to interpret Koonin’s article other than as something like this

    “look, this number is small, nothing to worry about”

    his post is a list of reason why the human effect is small, can’t be measured, etc. Maybe he didn’t actually say “nothing to worry about”, but that’s why I said “essentially goes”. An alternative might be “it’s very small, we shouldn’t do anything yet”. Choose some suitable variant, but it is certainly an argument that essentially minimises the risks associated with AGW because – according to Koonin – they’re small.

    FWIW,

    I think this post is intellectually dishonest.

    I’m rather tired of being called intellectually dishonest by people who should probably consider their own honesty before making such claims about others.

  21. MikeM,
    You need to hone your ability to interpret what people mean when they write something. It appears to be somewhat lacking. Illustrating that Judith Curry did not support them writing a statement is not illustrating that she made a strong case for them saying something else.

  22. anoilman says:

    Mike M: The avalanche has started its too late for the pebbles to vote.

    Let me tell you what to do in the face of uncertainty. Stop shipment, admit mistake, contact customers, and fix it.

    That fact is that your stance (and Koonin’s) is pro-terraforming. Yet you don’t have a plan. You don’t have a definition of acceptable death rates, and damage caused. You don’t have a plan to fix it if gets broke. You don’t even know where you want to stop, and how you’re going to do that. You got nothing.

  23. Willard says:

    > Does Koonin say, in either his post on Climate Etc. or the original Journal article, that there is nothing to worry about?

    Neither does Koonin say anything about beating squirrels, i.e. there’s no need for Koonin to “say” what is being dogwhistled. His implicit premise is that unless we know more about what has profund implications, [inaudible]. Here’s the implicit premise’s at work:

    First, it means precision observations are required to see the climate response. Second, it means that natural variations can easily overwhelm human influences, at least on multidecadal scales (witness the current stasis in global mean surface temperature). And finally, because life at the 1% level is rich, the models have to get many small phenomena right to confidently isolate and project the response to anthropogenic effects.

    The implicit premise is conveyed by the question: why the hell would we need to confidently isolate and project the response to anthropogenic effects? The answer to this is the usual lukewarm playbook, e.g.:

    https://contrarianmatrix.wordpress.com/do-not-panic/

    https://contrarianmatrix.wordpress.com/do-no-harm/

    Cf. with what Koonin said in his op-ed

    But climate strategies beyond such “no regrets” efforts carry costs, risks and questions of effectiveness, so nonscientific factors inevitably enter the decision. These include our tolerance for risk and the priorities that we assign to economic development, poverty reduction, environmental quality, and intergenerational and geographical equity.

    http://www.wsj.com/articles/climate-science-is-not-settled-1411143565?tesla=y

    TL;DR — there’s no need to say anything when what you convey suffices.

    ***

    At the ClimateBall level, Koonin’s argument basically transposes the “but CO2 is a trace gas” (an ontological claim) into its epistemic form: “but CO2’s impact’s so small we barely can trace it.” Rhetorically speaking, his argument relies on the law of small numbers to operate. Andy Lacis’ comment adressses both Koonin’s argument and his rhetorical ploy.

    Speaking of which, MikeM introduces two other ones with his fall:

    You can not have a reasoned debate when you refuse to listen to the other side.

    The first part of that fall falsely presumes Andy Lacis’ does not reasonably question Koonin’s trick. The second part speaks of “the other side,” which introduces unnecessary manicheism.

    A fall to this comment, say about honesty, squirrels, false assumptions, and reasoned debate is left as an exercise to readers.

  24. BBD says:

    You can not have a reasoned debate when you refuse to listen to the other side.

    Enough to make a cat laugh.

  25. Lars Karlsson says:

    Actually, Mike M is (inadvertently) pointing to the core of Koonin’s error.

    Koonin: ” First, it means precision observations are required to see the climate response. Second, it means that natural variations can easily overwhelm human influences, at least on multidecadal scales (witness the current stasis in global mean surface temperature). And finally, because life at the 1% level is rich, the models have to get many small phenomena right to confidently isolate and project the response to anthropogenic effects.”

    But of course, the fact that Earth’s mean surface temperature is 288 K does not tell us much about how large natural variations are. It is not like Earth’s orbit is changing very much (not on decadal or centennial scales), or that the amount of light emitted by the sun varies a lot, like a flickering fluorescent lamp. But that would be required if we were going to have natural changes that are “not small” according to Koonin’s way of reasoning.

    And it doesn’t mean that temperature measurements have an margin of error of many degrees. That is obviously complete nonsense.

  26. anoilman says:

    Willard… It just ‘feels safe’ to Mike M.

  27. mwgrant says:

    aTTP

    The xkcd cartoon is double edged, but I will not force the issue. :O) (I do like the cartoon however. “When in doubt linearize” has proven to be a practical credo over the years.)

  28. Joshua says:

    Hmmm.

    I’m going to defend MikeM to a certain extent here. While there might be sufficient evidence to make assumptions that Koonin was making a rhetorical play hidden behind a screen of plausible deniability, we don’t really have to go there. To got there seems, to me, to rely a bit too much on judging motivations.

    IMO, the standard should be higher. Koonin obviously knows that the discussion about climate change is politicized. The underlying thesis of his WSJ article makes it plainly obvious that he was intending to address that politicization. Accordingly, as someone who presumably has enough knowledge about the world to be a very successful scientist, then he strikes me that he should be interested in directly addressing how an WSJ editorial might exacerbate existing tensions; i.e., be interpreted by many politicized readers on both sides of the climate wars – in this case, as an advocacy for not being concerned about the potential of risk from ACO2 emissions. This reminds me of when Stevie-Mac makes plausibly deniable comparisons between Mann and pedophiles, to the point where his supporters tell him that they think the comparison is counterproductive, only to read then other “skeptics” arguing that no such comparison was actually made. Am I to believe that Stevie-Mac couldn’t anticipate polarized reactions to a plausibly deniable comparison? Really? I don’t buy that, and as such, whether or not the reactions can be proven, there is an unfortunate lack of accountability for reactions that would be easily predictable.

    IMO – Kooniin had a choice when he wrote that editorial. He could have decided to not exacerbate the destructive political environment by being clear to address the easily predictable counter-arguments (and interpretation from some “skeptics” that he was supporting a dismissal of concern about ACOw emissions) that his editorial would stimulate. That would not have meant that he couldn’t have presented his views on the science, but that he could have expressed them more clearly to make it very, very difficult to interpret his argument as something other than saying “Don’t worry, be happy.” But he chose not to do that.

    In the end, I’m left with trying to figure out whether in contrast to his stated intent, he actually wanted to exacerbate the existing public policy tensions along preexisting cleavages. I tend to doubt that, and I’m not in a position to judge his motives anyway. In the end, I just chalk his editorial up to the unfortunate ability of some folks to rise above sameolsameol.

  29. Joshua says:

    Anders –

    You could just delete my 6:09 comment to save everyone the pain of reading my ramblings – but if you don’t, would you mind ending the bold after “….both sides of the climate wars”?

  30. Joshua,
    Fixed it. I’m still trying to work out why that is a defense of MikeM though 🙂

  31. Joshua says:

    Maybe a bit convoluted. But what get from what MikeM is saying (beyond the quibbling about technical issues) is that these discussions are problematic if you assume bad faith intent.

    Although, then, his introduction of “I think this post is intellectually dishonest.” would have to be seen as a touch unintentionally ironic. 🙂

    BTW, can someone explain to me the difference between being “intellectually dishonest” and lying? Is it supposed to me unintentionally biased? If so,then why don’t people just say “unintentionally biased” instead of referring to “dishonesty” at all?

  32. Joshua,

    that these discussions are problematic if you assume bad faith intent.

    Well, yes, I agree with that. I think they’re also problematic if someone throws in an “intellectually dishonest”, although maybe that the same as assuming bad faith intent.

    can someone explain to me the difference between being “intellectually dishonest” and lying?

    Until convinced otherwise, my default assumption is that when someone uses “intellectually dishonest” they’re implying that you’re lying, but don’t want to actually come out and say that. Of course, that may be wrong but it’s hard to see how anyone who describes someone else (or what they’ve written) using the term “dishonest” could really hope for it to be interpreted differently.

  33. mwgrant says:

    If so,then why don’t people just say “unintentionally biased…?”

    Because ‘mispeaking’ has not advanced that far yet?

  34. David Young says:

    I do think Koonin’s point is fundamentally right. It’s not the temperature that’s the problem but the energy fluxes. Take an airplane. The lift force keeping it in the air is O(1). The drag force retarding forward motion is O(0.01) at least at cruise conditions. In a numerical simulation, you cannot separately compute the drag without also computing the lift. Thus, the RELATIVE error in the drag will generally be 2 orders of magnitude larger than the relative error in the lift. But, its the drag that we really care about. The climate is analogous. We care about small changes in rather large energy fluxes. Those small changes are going to be hard to measure or compute.

  35. DY,
    What you’re really saying is that according to you it should be difficult to do these measurements. Those who actually work in the field disagree.

    We care about small changes in rather large energy fluxes.

    Except it’s not small. Koonin’s mistake is by making the comparison to the total flux on one side of the energy balance calculation. Also, he chooses to ignore feedbacks. The greenhouse effect produces a surface flux 150W/m^2 greater than would be the case in the absence of an atmospheric greenhouse effect. Our current position is a surface warming of just under 1K and a planetary energy imbalance of around 0.5W/m^2. The increase in temperature of just under 1K means an increase in outgoing surface flux of over 5W/m^2. The increase in downwelling longwavelength flux has been measured. It’s almost 8W/m^2. The difference is because we’ve also had an increase in sensible heat from the surface and latent heat. In others words, anthropogenic influences have already increased the greenhouse effect from about 150W/m^2 to 158W/m^2. I don’t call that small.

  36. I realised that what I said above is not quite right, because the latent heat and convective transport from the surface isn’t normally included in the surface greenhouse forcing. Still, we’ve amplified it from about 150W/m^2 to about 155W/m^2. Still not exactly small.

  37. David Young says:

    Yes, but any computational method that attempts to model energy flows will be far less accurate at computing the 5W/m2 than at computing the 150W/m2. This is because the governing equations are nonlinear so it is impossible to derive a nice equation for the delta that doesn’t involve the full quantity itself. This is first year graduate student numerical analysis.

    It seems to me that generally, people in the field acknowledge that these things, even temperature, are very difficult to measure and require very sophisticated mathematical techniques and many decades of research. Certainly, Trenberth acknowledged that the uncertainty in the energy fluxes are not small, as I recall order 5-10 W/m2.

    The thing about this is I don’t quite understand why people object to this obviously true assertion. Perhaps there is a more neutral way to say it that would you would find more palatable.

  38. DY,

    Certainly, Trenberth acknowledged that the uncertainty in the energy fluxes are not small, as I recall order 5-10 W/m2.

    The uncertainties in the satellite measurements of the TOA fluxes are that large. The uncertainties in the increase in ocean heat content are nowhere near that large and are consistent with a continued increase in energy.

    The thing about this is I don’t quite understand why people object to this obviously true assertion.

    What obviously true assertion? That if you set up a model that can reproduce the greenhouse effect, and then were to double CO2 so as to produce a change in forcing of 3.7W/m^2, that the errors in the model would be too large for the model to properly represent an increase in outgoing surface flux of about 16W/m^2? If so, the reason I would object is because it probably isn’t true.

  39. DY,

    This is because the governing equations are nonlinear so it is impossible to derive a nice equation for the delta that doesn’t involve the full quantity itself. This is first year graduate student numerical analysis.

    Ignoring the irritating “first-year graduate student” jibe, you should probably read what Isaac Held has to say. An important distinction between the underlying non-linearities and the impact of some kind of external forcing is that the system appears (both observationally and computationally) to respond linearly to these changes in forcings. One obvious reason is that if I apply a force to a non-linear system, it will respond to that force in a way that is governed by Newton’s 2nd law. It doesn’t matter how complicated the internal dynamics are, if I push it, it will move. A second point is consistent with the argument you’re almost trying to make. These changes in forcing are quite small relative to the most fundamental force – Solar insolation. Given that the anthropogenic forcings are always going to be small relative to the total solar forcing, that the system behaves linearly to these changes in external forcing is not a great surprise. It would be much more surprising if the response was highly non-linear. That’s not to suggest that the internal dynamics don’t remain complex, non-linear, and chaotic, simply that the overall (averaged) global response is likely to be linear.

  40. David Young says:

    Yes, I read what Held said and perhaps for small changes it is correct. But its a very top level assertion. We know a lot about this from simpler modeling problems.

    We do this all the time in aerodynamics too and its true that the lift force is roughy linear over a range of angles of attack. Linear methods can replicate that slope IF they are adjusted with actual test data. They are NOT predictive. Nonlinear methods are required to get even the intercept of the linear curve right. So one could make the odd assertion that the real complex nonlinear interactions “cancel out” or are irrelevant in a very “rough” and top level way. And for some things, that’s true or else linear methods would never be used. Except, there are equally important details such as pitching moment that are very wrong in linear methods. The total force is linear (over a limited range) but the details are completely wrong. Further, there are nonlinear phenomena that can cause real problems even in the “linear” range. We are writing a paper on this right now.

    But in any case, if you want to claim that the nonlinear details don’t matter in some way, then it seems to me that you MUST rely on real data which actually tells us the global behavior. Then you can calibrate your simple model and know its limits.

    This is perhaps relevant to the ice age forcing discussion we had earlier. To predict what happens in an ice age, you cannot just add albedo to the forcing, because its unknown a priori. You must somehow model or predict it and then you have the issue that its nonlinear. I personally doubt that we can be very good at determining albedo 25000 years ago in any case.

    So i guess I would say what a turbulence modeler said about our current eddy viscosity models, they are post-dictive and not pre-dictive. It seems to me predicting things in advance is vastly superior. Nonlinear details matter there unless you have a lot of data to constrain simple models.

  41. DY,
    I think the problem with your lift force analogy is that the lift force is not external. It exists because of the structure of the wing, the orientation of the wing, and on the speed. If you were to simply apply an upward force, you could be pretty certain of what would happen. The point about anthropogenic forcings is that they are genuinely external. They don’t depend on the internal dynamics of the system. I still think that you aren’t recognising what is meant by an external forcing.

    But in any case, if you want to claim that the nonlinear details don’t matter in some way, then it seems to me that you MUST rely on real data which actually tells us the global behavior.

    I didn’t say they don’t matter. I’m suggesting that they don’t particularly matter when it comes to understanding the global, average response to some change in external forcing (or, rather, that there isn’t any evidence to suggest that they matter in some significant way).

    I believe, for example, that the seasonal response is an example of a broadly linear response to external changes. If it weren’t the case, we would probably expect the seasonal variations to be so complex as to not have a particularly good idea of what to expect – on average – over the next decade or so. However, we do, and typically the seasons do behave as expected. Clearly there are variations, but they are not large enough to produce something like snow in summer at sea level in the UK.

  42. BBD says:

    DY

    This is perhaps relevant to the ice age forcing discussion we had earlier. To predict what happens in an ice age, you cannot just add albedo to the forcing, because its unknown a priori. You must somehow model or predict it and then you have the issue that its nonlinear. I personally doubt that we can be very good at determining albedo 25000 years ago in any case.

    Uncertainty over LGM albedo can buy you +/- 1W^m2 but that’s not really enough, is it?

    🙂

    Hansen & Sato (2012):

    The altered boundary conditions that maintained the climate change between these two periods had to be changes on Earth’s surface and changes of longlived atmospheric constituents, because the incoming solar energy does not change much in 20,000 years. Changes of long-lived GHGs are known accurately for the past 800,000 years from Antarctic ice core data (Luthi et al. 2008; Loulergue et al. 2008). Climate forcings due to GHG and surface albedo changes between the LGM and Holocene were approximately 3 and 3.5 W/m2, respectively, with largest uncertainty (1 W/m2) in the surface change (ice sheet area, vegetation distribution, shoreline movement) due to uncertainty in ice sheet sizes (Hansen et al. 1984; Hewitt and Mitchell 1997).

  43. Willard says:

    > So i guess I would say what a turbulence modeler said about our current eddy viscosity models, they are post-dictive and not pre-dictive. It seems to me predicting things in advance is vastly superior. Nonlinear details matter there unless you have a lot of data to constrain simple models.

    Here’s how I’d rephrase this into an argument, David Young:

    (1) Climate models are post-dictive.
    (2) Predictive models would be better.
    (3) We need to include nonlinear details.

    For the argument to work, it seems to me that (2) requires:

    (4) We can create climate models with predictive power.

    I’m not sure what warrants this claim. Also, this claim is irreconciliable with the usual circularity argument regarding climate modelling. In other words, one does not simply argue that the Mordor modulz are both unfalsifiable and falsified.

    ***

    Also, (3) needs to be complemented with the following:

    (5) Unless we have predictive models, […]

    How would you fill behind the “[…]”?

    The problem here is that (3) is incomplete: we need to include nonlinear details for what exactly? I suppose it’s to make the modulz predictive. This in turn would imply that climate models that include some nonlinear magic could be predictive.

    Is this what you’re claiming?

    Many thanks!

  44. David Young says:

    ATTP, We can argue about whether the engine propulsion force is an “external” forcing or internal to the system. In a wind tunnel, there is certainly an external forcing. The lift is a response to that.

    Perhaps another point supporting Koonin’s observation is that we seem to have been unable to reduce the range of ECS since the Charney report and in fact it got a little bigger in AR5. That tells me that even if we think the response is linear for small changes, we don’t know the slope and the slope is what’s really important. However, even in the linear range, there could be surprises.

    I personally favor intermediate complexity models that are constrained by data. That’s based on my experience over the years. Basically what happens is that “colorful” looking and complex simulations sell better.

    We know the sign of seasonal responses and roughly the magnitude but local conditions are critical. Just compare Seattle and Minneapolis. The seasonal variation in Seattle is O(20F). In Minnespolis, I’d guess at least O(60F).

    Willard,

    Thanks for your thoughtful response (seriously). I do agree that it is an open question about whether we can build predictive GCM’s. The complexity is so high that constraining things with real data becomes a task too big for the human mind. Boundary layers are hard enough.

    I don’t believe its possible to adequately include nonlinearities in a fully resolved GCM. The range of scales is too great. But I do believe that simpler models may be pretty good if we have enough data. Predictive? Hard to say, but probably better than GCM’s. If we are in the “linear range” and we may be until 2100, these simple models might be surprisingly good. I know its technical, but our last August paper makes this case for boundary layer flows, along with appropriate caveats too.

    I’m not even sure GCM’s can do an adequate job with tropical convection, which is a critical detail for the water vapor feedback. Just my opinion. Your mileage will vary. One goal of mine is to imitate Nic Lewis or Nic Stokes after I retire. Maybe I can answer your question then.

  45. BBD says:

    DY

    I look forward to your monograph reconciling your concerns about models with palaeoclimate behaviour.

  46. DY,

    We can argue about whether the engine propulsion force is an “external” forcing or internal to the system. In a wind tunnel, there is certainly an external forcing. The lift is a response to that.

    Well now you’ve gone and changed your analogy. I don’t remember you mentioning propulsion before. Sure, I’d agree that propulsion is external, lift is not really. Any chance you could acknowledge my point, rather than suggesting that we were discussing something different to what we actually were? It’s not the most constructive manner in which to undertake these discussions.

    Perhaps another point supporting Koonin’s observation is that we seem to have been unable to reduce the range of ECS since the Charney report and in fact it got a little bigger in AR5.

    That isn’t Koonin’s point.

    I do agree that it is an open question about whether we can build predictive GCM’s.

    I don’t think that is what Willard was suggesting. Multiple planets and a time machine might be necessary if we are to develop genuinely predictive GCMs. My guess is that that is unlikely.

  47. Willard says:

    > Maybe I can answer your question then.

    That’s my question in the sense that I’m the one who asked it, David Young. However, I think the question is all yours. Until the question get answered or even resolved, what are we to conclude about the requirement of modulz to be predictive, i.e. (2)? While everyone can agree that having predictive modulz would be better, I’m not sure the same kind of agreement can be reached with burdening climate science with such a demand. Which means that we’re still left with some dots to be filled after (5), i.e.

    (5) Unless we have predictive models, […]

    I just don’t see any reason to require predictive models if teh modulz we have are clearly not that accurate.

    ***

    Good enough retrodictivity might be good enough to give us an idea of the ballpark we’re looking for. Non-model based approaches such as Vaughan Pratt’s seems good enough for me. Even a brain-dead argument such as this one seems good enough to me. To quote myself:

    > I don’t recall paying any dues to the American Oracular Society […]

    I thought it was a simple consequence of a master argument that should go like this:

    (P1) AGW is real, i.e. our addition of CO2 adds more warming to the globe.

    (P2) Dumping CO2 like there’s no tomorrow will trigger more of P1.

    (P3) More P1 will create significant disruptions in ecological systems, social systems, security and human health are likely to occur, &c.

    (P4) We ought to stop P3 from happening.

    (C) We ought to stop dumping CO2 like there’s no tomorrow.

    There’s no need for oracles to substantiate P3. There’s no need to have a full-blown version of P3 for the argument to hold. “More warming will eventually bring adverse changes to the American way of life” should suffice.

    http://rabett.blogspot.com/2015/04/anybunny-want-to-send-message-to-aps.html?showComment=1428545123231#c8868475130878044937

    If you have anything against that argument, comments welcome. I seek the most simple argument regarding AGW.

  48. David Young says:

    Willard, Your argument is a policy argument and depends on people’s values. I’m not going to enter that arena here. What I know is that change is going to happen due either to natural or anthropogenic causes and we had better get busy and start adapting. Human civilization has faced many very severe challenges and the progress has been very uneven.

    I’ve said many times that one problem in the climate debate is the blurring between the policy and science debates. It’s almost as bad as 19th century politics. My experience in fluid dynamics is actually a lot more positive. Most of the best people there are not involved in policy arguments.

    I do think we can narrow the range of ECS as we get more data and use simple models. I think Nic Lewis’ method is a lot better than Vaughan’s because Nic has very detailed arguments for every decision and has tried to include more effects. Certainly his (and Annan’s) point about uniform priors seems to have been accepted even though without any admission of error. Perhaps that will come 10 years from now after the guilty have moved on.

  49. Brandon Gates says:

    Willard,

    In other words, one does not simply argue both that the Morder modulz are both unfalsifiable and falsified.

    Falsifiability has been topical for me over at WHUTTFJH [1]: http://wattsupwiththat.com/2015/04/05/agreeing-to-disagree/#comment-1898509

    Obvious bait, but I was feeling game. A ruckus royale ensued as I don’t suffer disingenuity gladly. As well, I find it particularly galling when the sum total of the opposition’s argument is: Even if the hypothesis you have not proved falsifiable really is, it has been falsified by observation.

    My head —> ‘sploded.

    I should particularly like to point to various contentions by Terry Oldberg who asserts that he has “proven” the AGW hypothesis unfalsifiable; his main thesis being that since equilibrium temperature is not observable, any theory purporting to predict it cannot be falsified.

    In a previous tangle with him, I invoked Heisenberg’s uncertainty inequality, which I figured should have sent his confident declarations into a smouldering heap of logical fail, but it bounced right off and was countered by: what’s the relevance?

    A comment or two from the literati in this forum would restore some faith that I’m not wildly off-base, and that there is, in fact, some intelligent life on this rock worth being concerned with saving from themselves. A fine point or two on how I might otherwise have replied would also be welcome, esp. any glaring errors I ought not be making.

    “Just give it up” might actually find positive reception.

    ————————

    [1] Trans: Whut the eff just happened ….

  50. David Young says:

    ATTP, The aerodynamic problem is always posed in terms of an external forcing, even if only gravity in a glide. I apologize for leaving that out. Lift can only originate if there is an external forcing of some kind.

    The ECS uncertainty is an “illustration” of Koonin’s point that the effects are small compared to the total forcings and thus are hard to compute or estimate at least it seems so to me.

    Yes, if we agree that predictive GCM’s are unlikely, then we should jointly advocate a shift in funding to more realistic lines of research. I believe there are plenty of much simpler problems to work on, such as tropical convection. I think I understand what is happening here. It’s the colorful and complex looking simulations syndrome coupled with the prejudice that if we just included all scales we would get “the” answer. I have found that this view is very common especially among decision makers. It can be dangerous.

  51. David Young says:

    BBD, Here is what Lewis says AR5 says about that:

    “The first three bullet points of Slide 3 reiterate what the IPCC fifth assessment WG1 report (AR5) said in Chapters 10 and 12 about palaeoclimate ECS estimates and those based on short timescales or non-greenhouse gas (GHG) forcings, and its implicit conclusion that estimates based on multidecadal warming during the instrumental period (since about 1850) were likely to prove most reliable and provide the narrowest uncertainty bounds.”

    I have no intention of getting into paleoclimate as it seems to me an even more ill posed problem than the GCM problem.

  52. Willard says:

    Come now, David Young. These talking points don’t fill up the dots at the end of:

    (5) Unless we have predictive models, […]

    Talking about adaptation, the fact/value dichotomy, narrowing the range of ECS, and the usual names dropped won’t help fill up these dots. While it is tempting to follow you on these squirrels (Nic’s track record on error admission is quite low, and he’s up to his neck in policy), I’d rather not play ClimateBall when you really could help me improve the lukewarm playbook or the simplest argument regarding AGW.

    Speaking of which, I rather like Snarkrate’s version:

    We know with 100% certainty that CO2 is a greenhouse gas and that greenhouse gasses raise the planet’s temperature. We have good reasons to think that this rise in temperature will raise sea levels and perhaps lead to more flooding events and more drought in some regions.

    So, given that we have established a credible threat, rather than taking comfort in the assertion that the models are not reliable, shouldn’t you instead be saying “Holy fuck! The models are unreliable, and therefor we cannot bound the risks due to climate change!

    It would seem to me that a lukewarmer would be very worried if they thought the models didn’t work. Why do you instead find this a source of sanguinity?

    http://rabett.blogspot.com/2015/04/anybunny-want-to-send-message-to-aps.html?showComment=1428700259252#c1377301935247975229

    You may notice that Russell also tried “but teh modulz” against it. It does not seem to work. However, to make sure, I’d like to know what’s under these dots in (5).

    Thank you for letting me express my seriousness for a change,

    W

  53. BBD says:

    And then there’s palaeoclimate:

    Veron (2008)

    All the nonsense in the blogosphere doesn’t add up to a single mass extinction, never mind five.

  54. Joshua says:

    ==> “It would seem to me that a lukewarmer would be very worried if they thought the models didn’t work. ”

    It’s interesting how many times I’ve seen Anders raise that question and yet (perhaps I’m biased?) I can’t recall seeing an answer.

  55. Brandon Gates says:

    David Young,

    What I know is that change is going to happen due either to natural or anthropogenic causes and we had better get busy and start adapting.

    Ok, to what future state?

    Your argument is a policy argument and depends on people’s values. I’m not going to enter that arena here.

    You predicted wrongly: read the latter part of the first quote block, starting with “we had better”.

  56. Brandon Gates says:

    Joshua,

    Perhaps I am biased as well, but that seems a defining characteristic of the position. For possible exceptions, I think Mosher might actually apply. You guys track him far better, however.

  57. BBD says:

    You guys track him [Mosher][ far better, however.

    He has yet to clarify what he means by calling himself a lukewarmer despite repeated requests. That’s naughty and he will be asked again until this is sorted out.

  58. izen says:

    @- Mike M.

    Thank you for finding and pointing out another of Dr Curry’s contributions to the internal discussion by experts within the APS about how to respond if asked where physicists stand on AGW. –

    From the workshop transcript starting on page 517:
    “DR. CURRY: Personally, I don’t think the scientific societies should make statements about those kind of, what I would call public policy that is not related to the policy of science like we need more observing systems and things like that. … you should only speak to where your expertise is” … I am not even sure that APS has sufficient expertise on the climate issue to be making a statement at all.”

    The irony of a member of the APS asked as an expert who clearly considers their own expertise to be sufficient to make a statement on this subject to Congress and the George C Marshall institute, advising the APS it lacks the expertise to make a statement, is not lost on me.

    However I suspect that the underlying context of Dr Curry’s argument is not really that the APS lacks the expertise to formulate an answer to the question; “What do physicists think about climate change”.
    It is that as an associative, rather than regulatory, society it lack the epistemological authority to make a general statement about what the science of physics, as practiced by it members, has concluded about this issue. I am guessing that in Dr Currys view it is acceptable for individuals to make a statement, based on their personal expertise, about AGW (at least if its the minority view!) but it is not acceptable for a society of physicists to state what the general view of its member physicists might be if it was asked.
    That would smack of turning the majority view into an authoritarian central edict about what view ALL physicists SHOULD hold about the science of AGW.

    I think this is also one of the reasons the Cook13 paper and its ilk attract so much antipathy. A measure of the communal position of many scientists gives no accurate measure of the validity of the science (?!), but becomes a club of conformity. The measured majority becomes a reason to ask- ‘what the hell is wrong’ with the 2% of scientists and/or science papers that DO dissent from the mainstream view of AGW.

  59. David Young says:

    Willard, We seem to be back to the usual disagreements about values and we will not be able to resolve these on a blog. Remember Bertrand Russell?? And I had such hopes at the beginning of staying out of this thicket.

    I have faith in science and mankind and think that we will come in our own muddled way to a way through the anthropocene, despite many challenges. My gut feeling is that AGW is a problem but probably a less severe one than many people say, just because of the strong politization of the issue and the nasty nature of the debate even for scientists. I believe that in 10 or 20 years, we will have a lot better handle on ECS just because of better data assuming that we direct our resources in the right ways. I actually regard the huge imbalance of scientific resources between GCM’s and data gathering, theoretical understanding, etc. to be a far bigger threat to progress. Facts and data will help us resolve our differences, sarcastic cartoons will not achieve anything but make you feel falsely virtuous.

  60. Joshua says:

    ==> “For possible exceptions, I think Mosher might actually apply.”

    My Magic 8-Ball has gone from “Outlook not so good” to “Reply hazy, try again”

  61. Joshua says:

    David –

    ==> ” I actually regard the huge imbalance of scientific resources between GCM’s and data gathering, theoretical understanding, etc. to be a far bigger threat to progress. ”

    A far bigger threat than what to progress on what?

  62. Brandon Gates says:

    BBD,

    I have been following the conversation about Mosh’s positioning statements with some interest. My opinion is that he’s in a political bind and has to be circumspect. That’s me being diplomatic, slippery is more the term. I think one should first be clear with whom it is they ultimately stand, then make arguments for/against any and all. That’s not just ethical, but I think more effective. Oh look, I have just expressed “concerns”. Dammit.

    To his credit he shoves a fair amount of the WHUTTers’ nonsense down their throats, one of those topics is the necessity of models. I don’t think I’ve seen him actually endorse the current CMIP5 crop … I’d have to check.

    Philosophical day here, my navel-lint is interesting as ever.

    Joshua,

    I’m learning to appreciate your 8-ball, it agrees with mine an awful lot.

  63. BBD says:

    DY

    Facts and data will help us resolve our differences

    So why shy away from acknowledging the data?

    Veron (2008)

  64. Willard says:

    OK, David. Let’s follow through this one:

    I believe that in 10 or 20 years, we will have a lot better handle on ECS just because of better data assuming that we direct our resources in the right ways.

    Nevermind the “right ways,” which sounds awfully normative – to what avail do we need to have a lot better handle on ECS?

    The actual range looks awfully like the one we had 30 years ago.

  65. Joshua says:

    izen –

    ==> “The irony of a member of the APS asked as an expert who clearly considers their own expertise to be sufficient to make a statement on this subject to Congress and the George C Marshall institute, advising the APS it lacks the expertise to make a statement, is not lost on me.

    But you are missing the most important point there. Others are activists while Judith is not. And Judith is an expert yet she knows that others are not.

    I’ve asked Judith what criteria she uses to determine expertise, and in response she explained that she is an expert on expertise – so I guess I should just be convinced by her appeal to self-authority.

  66. BBD says:

    Brandon G

    I think one should first be clear with whom it is they ultimately stand, then make arguments for/against any and all. That’s not just ethical, but I think more effective.

    Physics vs libertarian physics takes no prisoners. Mosh is headed for the pit and he has nobody but himself to blame.

  67. Brandon Gates says:

    BBD, I find myself uncharacteristically bereft of any substantive objections.

  68. David Young says:

    Yes, Willard, AR5 has essentially the same range as the Charney report. But there is a lot of research going on right now, a lot of it based on data. Getting rid of uniform priors was a step forward, after a nasty fight about it, like it seems everything in climate science. Certainly, the credible observationally based estimates have been decreasing. GCM’s are probably hopeless for this.

    You of course know why we need to know the answer, I’m sure.

    I would observe that Bertrand Russell was very “normative” in his thinking.

  69. russellseitz says:

    The wiley Willard fails to note that Snarkrates is wagging a finger at his own straw man. Please read my replies at the same url.

    https://andthentheresphysics.wordpress.com/2015/04/10/andy-lacis-responds-to-steve-koonin/#comments

    The intractible policy problem goes beyond scientific uncertainty: it’s the disparity of time scales , human and geophysical.

  70. Willard says:

    GCMs do many other things than to estimate ECS, David Young. Just like the previous “objective Bayesian” word placement, “observationally-based estimate” is mostly a marketing effort. This time, it engineers dissent among engineer-minded interpretative experts.

    (Speaking of which, Joshua reminded me to add another box: expert expertise.)

    This is the third time you refuse to fill up the dots. The “observationally-based estimate” rests on these dots. So be it. Let’s return to Isaac instead:

    DR. HELD: The models look pretty linear. The observed seasonal cycle, that looks linear. Even if in the Ice Age times, things look pretty linear. We don’t know that much about it. So, why should I assume that things are, gee, the anthropogenic CO2 pulse is going to interact in some exotic way with internal modes of variability? Well, it’s conceivable. But I am not convinced. I don’t think that is particularly relevant.

    DR. KOONIN: But to come back to my earlier hobbyhorse, that means that the sensitivity you determined to, let’s say, CO2 from the last 30 years, you should use in extrapolating out of next century?

    DR. HELD: Yes, I don’t think there is much evidence that there is much secular variation in sensitivity.

    DR. LINDZEN: But I think this is important. For instance, when I presented the simple analysis, I was assuming it was all due to anthropogenic. Sensitivity is a separate question. And I think in conflating the two issues, we are confusing things.

    DR. HELD: I was trying to separate them here. I don’t think there is so much a collection of sensitivity as you are saying. I just think if you want internal variability to be important, you have to be in a low-sensitivity model by definition. And then you are going to have the heat going in the wrong direction. It’s just so basic to me, I don’t see why we talk about it.

    DR. KOONIN: Some of us haven’t spent 30 years.

    DR. HELD: You asked about our models getting better. I actually don’t hink this is a big issue for this group. It’s a hard problem and if you go into chapter 9, there is a frequently asked question. All the chapters have these things, and one of them is are models getting better? And this is a figure [next page] I took from the answer to that frequently asked question showing precipitation correlation with observations, CMIP2, CMIP3, CMIP5.

    DR. KOONIN: As we have discussed, the correlation coefficients depend on what frequency band you are looking in?

    DR. HELD: This is a spatial correlation, nothing to do with time, just space. One thing that has happened is that poor models disappear. You say how do models get selected? People get embarrassed at zeroth order. People put models in. They look really bad when they put them into these databases. They just drop out and you end up with better models just by public relations. It’s harder to say that the best models are getting better.

    The last sentences emphasized seem to describe quite well most of the auditing sciences.

  71. Joshua says:

    Cross-posted at Climate Etc.

    Uh Oh. I’m in big trouble now!

    So a way of spotting trolls early in their online careers and preventing their worst excesses would be a valuable tool.

    Today, Justin Cheng at Stanford University in California and a few pals say they have created just such a tool by analyzing the behavior of trolls on several well-known websites and creating an algorithm that can accurately spot them after as few as 10 posts. They say their technique should be of high practical importance to the people who maintain online communities.

    My days are numbered:

    http://www.technologyreview.com/view/536621/how-a-troll-spotting-algorithm-learned-its-anti-antisocial-trade/

  72. Willard says:

    > Please read my replies at the same url.

    Let’s quote them.

    The first is:

    Why don’t you head up to the Pnyx and warn the Athenians about the clear and present danger of the 100 years war.

    http://rabett.blogspot.com/2015/04/anybunny-want-to-send-message-to-aps.html?showComment=1428695318901#c6291118800512469266

    The second:

    The National Interest April 1990, pp 54-62
    Foreign Affairs letters , April 2009
    Earth’s Future December 2013, pp 43-47

    http://rabett.blogspot.com/2015/04/anybunny-want-to-send-message-to-aps.html?showComment=1428701329208#c4872680377136810932

    ***

    That we don’t know when we ought to stop dumping CO2 like there’s no tomorrow doesn’t contradict that we ought.

  73. John Hartz says:

    Meanwhile, back in the real world…

    A new study bolsters the case that a period of much faster global warming may be imminent, if not already beginning. The study, published Wednesday in Geophysical Research Letters, uses climate records gleaned from coral reefs in the South Pacific to recreate sea surface temperatures and ocean heat content dating back to 1791. The corals examined were from Fiji, Tonga and Rarotonga.

    Information from the coral reef core samples reveals how ocean surface temperatures have varied over time in the South Pacific, along with how the uptake and release of upper ocean heat content has varied over time, as well. The insights they provide, together with other recent research, carry important implications for how global warming may play out during the next two decades or so.

    Rapid global warming may be coming sooner than you think by Andrew Freedman, Mashable, Apr 9, 2015

  74. Willard says:

    > As well, I find it particularly galling when the sum total of the opposition’s argument is: Even if the hypothesis you have not proved falsifiable really is, it has been falsified by observation.

    Just caught your comment, BG.

    Here’s my favorite death thread about falsification:

    https://ourchangingclimate.wordpress.com/2014/02/17/is-climate-science-falsifiable/

    Good ClimateBall exchange, BTW.

  75. russellseitz says:

    Bad Willard!

    You left out four of my comments in that thread preceding the one you style : “The first is:”:

    No ellipsis , please.

  76. Willard says:

    > You left out four of my comments in that thread preceding the one you style : “The first is:”:

    Your “please read my replies at the same url” was preceded by this:

    The wiley Willard fails to note that Snarkrates is wagging a finger at his own straw man.

    https://andthentheresphysics.wordpress.com/2015/04/10/andy-lacis-responds-to-steve-koonin/#comment-52906

    I don’t think there’s any other response to Snarkrates.

    ***

    Even if Russell (no, not that Russell, David) only provided three incomplete citations, let’s produce his master argument:

    P1. Physics is neither ecology, social history,international security or medicine .

    P2. Parameter value uncertainties as elementary as doubling sensitivity render it physically impossible to physically predict the quantitative evolution of climate on decadal scales .

    P3. the dynamic system in question is complex , chaotic, internally variable and posessed of thermal mass and inertia on milennial scales

    C since GCM’s are not physical systems, many contemporary assertions as to how the system will respond to radiative forcing are more oracular than physical.

    http://rabett.blogspot.com/2015/04/anybunny-want-to-send-message-to-aps.html?showComment=1428556367905#c2285387123698002718

    In response to his “the dynamic system in question is complex , chaotic, internally variable and posessed of thermal mass and inertia on milennial scales,” here’s what I said:

    Yet adding up CO2 warms it up, adding more to it will warm it up, we know enough Greek mythology to know what happens when the elements are not playing for the home team, and we ought to stop adding CO2.

    I guess this would be something like a counterpoint:

    Until the model of the big badly modeled gooey thing shown in deep blue underneath the atmosphere catches up in realism, policy belief systems predicated on GCM’s will remain metaphysical as appeals to distant equilibria – or other folk’s physical intuition.

    http://rabett.blogspot.com/2015/04/anybunny-want-to-send-message-to-aps.html?showComment=1428615205049#c8763686666859309562

    Then Snarkrates took over, and Russell admitted that it was “about time,” which conflates a when-question with an if-question.

    One obvious problem with this counterpoint is that “policy belief systems” do not appear in the premises of the argument, and only echoes the “oracular” remark.

    ***

    Now, compare Russell’s (P2) with David’s (4):

    (4) We can create climate models with predictive power.

    (P2) Parameter value uncertainties as elementary as doubling sensitivity render it physically impossible to physically predict the quantitative evolution of climate on decadal scales .

    Reconciling the two claims may require some work.

    ***

    Finally, speaking of ellipsis, compare Russell’s counterpoint with David’s (5)

    (5) Unless we have predictive models, […]

    (Counterpoint) Until the model of the big badly modeled gooey thing shown in deep blue underneath the atmosphere catches up in realism, policy belief systems predicated on GCM’s will remain metaphysical as appeals to distant equilibria – or other folk’s physical intuition.

    Reconciling these two claims may require less work.

    ***

    On the basis of Russell’s argument and counterpoint, one might suspect that David’s ellipsis contains more norms than he’s willing to admit.

  77. Larry says:

    I will confess that I have not yet read much beyond the part I quote below but the issue I raise does not seem to be addressed (as it must) nearby.

    “From this, it should be sufficiently clear that, when viewed in absolute energy terms, the viable margin between life and death in the Earth’s biosphere is remarkably narrow – so much so that a seemingly insignificant 1% to 2% change in the total energy of the global environment will invariably result in serious disruption of the established infrastructure of life in the biosphere.”

    You are asking me to ignore the fact that we are talking about a system within a system, where the inner system is designed (Designed?) to survive in a wide variety of outer-system environments–it is why we cah survive in the current-day Antarctica and in equatorial Africa.

  78. Vaughan Pratt says:

    @DY: I think Nic Lewis’ method is a lot better than Vaughan’s because Nic has very detailed arguments for every decision and has tried to include more effects.

    This raises the interesting question of whether the more detailed an argument the better. In 1910 Russell (yes, that Russell, Willard) and Whitehead gave a very detailed argument for 1 + 1 = 2, occupying many pages. Elsewhere I have given an argument for the same theorem based on Roman in place of Arabic numerals and the interpretation of concatenation as addition. (Exercise: come up with that argument yourself.) According to David’s criterion, Russell and Whitehead’s much more detailed method would be a lot better than mine.

    As to whether more effects are better, again David would apparently prefer an argument based on a hundred effects over one based on only ten. Right now I’m working on explaining “modern global climate”, defined as five octaves of the temporal spectrum counting down (in frequency) from five-year periods to 160 years, in terms of only three effects, on the ground that if three effects adequately explain the climate at that resolution, then wouldn’t Occam’s Razor militate against additional effects? (Anyone wielding a razor is clearly not mitigating but militating.)

    I have two objections to long proofs when an equally sound short one suffices.

    1. Long proofs unduly burden their audience. For both writer and reader, short proofs are a great labor saving device.

    2. Long proofs offer many more hiding places for lacunae. How does David guard against Nic squirreling away lacunae in his detailed arguments?

  79. Brandon Gates says:

    Willard,

    Here’s my favorite death thread about falsification:

    I’ve skimmed the article, not dug into it nor the thread itself. But that’s jackpot, thank you. In a way I’m glad I didn’t have it prior to the furball I waded into on WUWT.

    Good ClimateBall exchange, BTW.

    Why thanks. It’s started up again. I offered Willis the carrot or the stick. He chose the stick.

    A point of negative feedback from you would be most welcome if you have the time or inclination. I have to have screwed something up, and the idjuts aren’t helpful on that score … everything I say is wrong in their eyes. Cheers.

  80. Hyperactive Hydrologist says:

    Surely the baseline metric for change should be the temperature range that sustain the current life on Earth. Also the rate of change needs to be considered and should be maintained at a level that allows all species to adapt, this should include the rate of change of ocean acidification.

  81. Larry,

    You are asking me to ignore the fact that we are talking about a system within a system, where the inner system is designed (Designed?) to survive in a wide variety of outer-system environments–it is why we cah survive in the current-day Antarctica and in equatorial Africa.

    Designed? I’m certainly not asking you to ignore that we can survive in Antarctica and in equatorial Africa. I might like to know why you think that’s relevant.

  82. HH,
    That would seem pretty obvious to me. Others seem to think that illustrating that life could survive, or has survived at some point in the past when conditions were similar to what we might expect in the future is sufficient to “rock on!”

  83. Brandon Gates says:

    Joshua,

    So a way of spotting trolls early in their online careers and preventing their worst excesses would be a valuable tool.

    Twice I’ve seen a certain poster on a certain contrarian blog I frequent be first responder to comments mentioning bots being deployed by the alarmist troll brigade to astroturf “sceptic” discussion fora. I figger one more data point will make my hypothesis plausible …

  84. John Hartz says:

    Speaking of the rate of ocean acidification…

    University of Edinburgh researchers warn that the carbon emissions that drove a mass extinction event some 252 million years ago were released at a rate similar to today.

    Oceans Facing Carbon Rates Which Spurred Mass Die-Off 250 Million Years Ago by Sarah Lazare, Common Dreams, Apr 10, 2015

  85. Eli Rabett says:

    Hey, what happens at Rabett Run stays at Rabett Run. Bad enough ATTP pinched the post, but the comments too?

  86. John Hartz says:

    Eli: On the other hand, imitation is the sincerest form of flattery.

  87. Nice to see the Bear awake from his hibernation and eviscerate Koonin and Curry with his extensive first-order physics argument.

    I agree with mwgrant that the XKCD cartoon is a double-edged sword. I have the most respect for those climate scientists that came up through the physics, astronomy, and geophysics ranks, such as the Bear (Lacis), James Hanson, Raymond Pierrehumbert, Isaac Held, etc and even Carl Sagan. These are the guys that can actually state premises like those suggested in the cartoon — many of those assertions would cause climate scientists that came up through the meteorology ranks to blow steam out of their ears. The key to them is that the climate is a complex system and that only people such as Curry have sufficient old-school knowledge to make predictions.

    But then you have the issue of David Young who flings FOO all of the proceedings. Certainly he is a physicist or mechanical engineer with extensive experience in hydrodynamics, but that may only mean that he has lost sight of the the forest for the trees. As you can tell, the sword cuts both ways.

    I have a lot more to say about this, but refer to a comment I put up on Baez’s Azimuth blog yesterday:
    http://johncarlosbaez.wordpress.com/2014/06/24/el-nino-project-part-2/#comment-65699

    Talk about a few mathematicians and physicists that are experiencing the XKCD panel !

  88. Actually, Eli, according to Koonin, if I only steal 1% to 2% of your posts and comments, it’s really small.

  89. I agree with Willard that Vaughan Pratt’s model is not really a model, just a smart use of filtering and cycle fitting. One trick I picked up from Vaughan is the triple-running mean filter.

    And you have to admire how Vaughan drops by here and buries FOO Young in another variation of the first-order physics argument.

    I also recall Lacis pummeling Young and others over the role of GCMs in the infamous CO2 control knob thread on CE a few years ago:
    http://judithcurry.com/2011/10/11/co2-control-knob-discussion-thread/

  90. russellseitz says:

    Eli, Willard’s problem is less snatching the comments than not reading the references-

    Stoat has kindly linked to the first , A War Against Fire , at

    http://scienceblogs.com/stoat/2015/04/09/all-of-this-will-soon-be-moot-anyway/

  91. Willard says:

    Your argument, dear Russell, is invalid.

    Source: The Internet.

  92. Brandon Gates says:

    JH: I have said the same to the WUTTers when they indicate I am acting like an arse.

  93. BBD says:

    Nice to see that John Hartz is on the ball and RTFRing.

    There’s more to CO2 than climate sensitivity and there’s more to the palaeoclimate data than an easy way to show that lowball central sensitivity estimates are implausible.

  94. JCH says:

    BBD, or others – what do you think of Nic Lewis’ reliance upon the AMO? I think it’s Judy junk.

  95. BBD says:

    JCH

    Have you come across Tamino’s post on the AMO?

    He shows that:

    Variations in the forced signal do leak into the AMO definition. Correlations with the AMO index do alias effects of global warming.

  96. Vaughan Pratt says:

    @WHT: I agree with Willard that Vaughan Pratt’s model is not really a model, just a smart use of filtering and cycle fitting. One trick I picked up from Vaughan is the triple-running mean filter.

    My AGU 2014 poster (Dec. 16, 2014) replaced the 150-year sawtooth cycle in my AGU 2012 poster directly with Length of Day data. This January I developed a slightly more detailed physical model of the impact of LOD on surface temperature based on a hypothesis about how it influences magma leakage (the amount of leakage adds a parameter), which fitted HadCRUT4 much better. In this hypothesis the cause of the LOD fluctuations is as per the second paragraph of the rightmost column of my AGU 2012 poster, namely fluctuations in the angular velocity of the Earth’s inner core but now with stronger support from Hide et al (2000) (1st reference in my 2014 poster).

    The only cycle now left in my model is the 20-21 year Hale or magnetic solar cycle. (Its second harmonic as the non-magnetic solar cycle doesn’t appear to add enough to accuracy of modeling of multidecadal climate to be worth including, the same conclusion Loehle and Scafetta reached in 2011 with their four-component model.) The only role for that cycle is to account for the pause, only visible when HadCRUT4 is smoothed to a running mean of 10 years or less. With 20-year smoothing (the basis for IPCC’s definition of Transient Climate Response and all that really matters when projecting to 2100) it can be dispensed with. My model is then based on only two effects, rising CO2 and varying LOD, requiring a grand total of three parameters to achieve an excellent fit to HadCRUT4—one vertical-scale parameter for each of CO2 and LOD, and the third for magma leakage.

    Regarding filters, as Pekka Pirilla astutely noted immediately after my post on CE back in December 2012, my triple running mean filter F3 was so close to a Gaussian as to raise the question of why not just use a Gaussian (or a 21-point binomial filter, a Finite Impulse Response (FIR) filter which is even closer to a Gaussian than F3). My only reason at the time was that F3 completely kills the frequency targeted by F1, its first component. But the equivalent Gaussian kills almost all of it, and the difference is pretty much completely masked by the considerable noise in the climate signal.

    So these days I use a Gaussian when I want a really smooth result, and otherwise a box filter because (a) a running mean filter is accessible to a wider audience than either a Gaussian filter or F3, (b) FWIW it completely kills the target frequency when there is one, and (c) it is narrower than either the equivalent 21-point binomial filter or truncated Gaussian (truncated because technically a Gaussian is an Infinite Impulse Response or IIR filter) or F3. (Some physicists actually find the output of a Gaussian filter suspiciously smooth and are reassured when they can see the leakage from a box filter, presumably because that’s what they’re accustomed to seeing—“How I learned to stop worrying about the leakage and love the box.”) This is all at the elementary end of wavelet theory.

  97. Steven Mosher says:

    “the viable margin between life and death in the Earth’s biosphere is remarkably narrow – so much so that a seemingly insignificant 1% to 2% change in the total energy of the global environment will invariably result in serious disruption of the established infrastructure of life in the biosphere.”

    Huh.
    The viable margin is “narrow!
    ? “. “life” has managed quite well despite large changes.
    “serious disruption ” hmm. Not sure if he is a luck warmer with a comment like that.
    ” infrastructure ” of life? Wtf

    I thought his piece was great. But this part was lame.

  98. Rob Nicholls says:

    BBD, in the distant past you said “All the nonsense in the blogosphere doesn’t add up to a single mass extinction, never mind five,” which seems to me a useful summary of something important. Thanks for linking to Veron 2008. And thanks to John Hartz for highlighting the article about the paper on ocean acidification and the huge extinction event at the end of the Permian.

    This is truly alarming stuff (although I don’t know how these papers fit into the context of other published literature on the subject), and I’ve wondered a bit recently about why the possibility of mass extinction in relation to AGW / Ocean acidification isn’t something that I’ve seen discussed more. I get the impression talking about mass extinction is seen by many as alarmist hyperbole, but at a simple level, given that previous abrupt major changes in the environment seem to have sometimes led to mass extinctions, and given that we’re doing something fairly abrupt and major to the environment with AGW, I would think the evidence that AGW won’t cause a mass extinction would have to be very, very strong in order to dismiss the possibility (and it’s difficult for me to imagine that any such evidence would be very, very strong). I’m pretty ignorant about this; I’ll try and delve a bit deeper into IPCC AR5 and see what’s there about this, but any useful pointers to further papers / summaries of the evidence etc would be much appreciated.

  99. Steven,

    I thought his piece was great. But this part was lame.

    How wide do you think it is? Bear in mind that Andy Lacis was responding to Koonin’s construct of measuring everything relative to a surface temperature of 288K. Relative to 288K, the margin is probably +- 5% (about 15K). Therefore, by this metric 1% to 2% would indeed – I think – result in serious distruption of the established infrastructure of life in the biosphere.

    Of course, if Koonin had measured everything relative to a more sensible metric (33K, for example), then the viable width would indeed seem wider.

  100. > “life” has managed quite well despite large changes.

    To give one example:

    Tardigrades can survive in extreme environments. For example, they can withstand temperatures from just above absolute zero to well above the boiling point of water (100°C), pressures about six times greater than those found in the deepest ocean trenches, ionizing radiation at doses hundreds of times higher than the lethal dose for a human, and the vacuum of outer space. They can go without food or water for more than 10 years, drying out to the point where they are 3% or less water, only to rehydrate, forage, and reproduce.

    http://en.wikipedia.org/wiki/Tardigrade

    That’s very small water.

  101. Rob Nicholls says:

    I have a great deal of respect for tardigrades / water bears and it’s nice to think that we might be incapable of wiping them out.

  102. BBD says:

    Rob Nicholls

    There’s lots, but for an overview, try Hönisch et al. (2012) The geological record of ocean acidification.

  103. BBD says:

    Rob N

    There’s an excellent, short, very readable book by Peter Ward called Under A Green Sky about the relatively recent discovery of the link between tectonically-forced CO2 warming events and mass extinctions. Highly recommended.

  104. I vaguely remember Eli having a post where he included figures from one of the IPCC reports that indicated mass extinction events under some extreme scenarios. I can’t quite find the link though.

  105. BBD says:

    SkS recently reviewed the way in which the Chixulub impact is no longer considered the single event that triggered the extinction of the dinosaurs. The role of tectonic CO2 forcing and its ecological impacts is re-evaluated.

  106. BBD says:

    ATTP

    It’s increasingly clear from the palaeoclimate evidence like the just-published (Clarkson et al. 2015) is that the rate of forcing really matters. I’ll quote the abstract for the thread:

    Ocean acidification triggered by Siberian Trap volcanism was a possible kill mechanism for the Permo-Triassic Boundary mass extinction, but direct evidence for an acidification event is lacking. We present a high-resolution seawater pH record across this interval, using boron isotope data combined with a quantitative modeling approach. In the latest Permian, increased ocean alkalinity primed the Earth system with a low level of atmospheric CO2 and a high ocean buffering capacity. The first phase of extinction was coincident with a slow injection of carbon into the atmosphere, and ocean pH remained stable. During the second extinction pulse, however, a rapid and large injection of carbon caused an abrupt acidification event that drove the preferential loss of heavily calcified marine biota.

  107. Vaughan Pratt says:

    @BBD: [Tamino] shows that: Variations in the forced signal do leak into the AMO definition. Correlations with the AMO index do alias effects of global warming.

    (That was January 30, 2011.) Tamino didn’t show anything, he merely quoted Wikipedia to the effect that defining the AMO as the result of linearly detrending N. Atlantic SST has the effect of incorporating the nonlinear part of global warming into the AMO thus defined. This is obvious.

    A month earlier (December 2010) at

    http://thue.stanford.edu/killamo.pdf

    I had proposed a way round the detrending problem that exploited the fact that the fluctuations people were associating with the AMO, along with all other faster fluctuations, were on a shorter time scale than global warming and that one could therefore remove those fluctuations with a suitable low-pass filter, thereby leaving just the (visibly nonlinear) global warming signal. As the name “killamo” was intended to suggest, this was not intended as a way of defining the AMO but of eliminating it from the climate signal (though of course one could then define the AMO by detrending N. Atlantic SST by the global warming signal as suggested in the first paragraph of Robert Way’s response to Tamino’s post).

    When Frank O’Dwyer drew Tamino’s attention to my killamo method he responded “He hasn’t removed the impact of AMO, he’s just removed the fluctuations on a 65-year timescale.” Given that no one was expecting the AMO to include fluctuations on a significantly longer timescale, and that what remained was a great fit to the expected contribution of CO2, I didn’t understand Tamino’s objection and suggested that we might be talking at cross purposes. But Tamino seemed not to mind talking at cross purposes, saying “I think your theories have no merit. I respect your civility. But no, I don’t wish to argue with you about it” and kicked me off his blog. And he dismissed Robert Way’s suggestion with “Good luck to those who try to find new ways to remove the global warming signal from N.Atl SST” (which was the opposite of what I was aiming for, namely removing all short-period events up to and including anything that could be considered part of “the AMO” whatever that meant so as to leave a signal that correlated beautifully with the expected CO2 warming).

  108. Vaughan,
    Isn’t the point that Tamino is getting at that if you simply remove all variability from that temperature record, you’re essentially assuming that the external forcing contribution has no variability on any timescale less than 65 years. However, this may not be correct as the external forcings do show variability. Therefore, what you’ve extracted may not be the full climate signal. Similarly, what you removed may not be purely an internally driven component.

  109. JCH says:

    VP – I remember it well. I might have been the one who suggested you visit his blog.

    I’m all for killing the AMO. Can we also burn the AMO literature on the AMO’s pyre?

  110. Eli Rabett says:

    Richard Telford spotted something related in another lame manuscript by Lüdecke We may hope and pray that Zorita, or whomever is the editor tracks back to quantpaleo, where Richard flashes more teeth.

  111. Brandon Gates says:

    ATTP,

    Therefore, what you’ve extracted may not be the full climate signal. Similarly, what you removed may not be purely an internally driven component.

    Inspired by old Tamino posts, I’ve done a bit of very amateurish kitchen-sink regressions not only on the major ocean/atmosphere indices but things like CERES: http://ceres.larc.nasa.gov/order_data.php

    Obviously those data are used by pros for stuff, but my experience is that it’s mostly papers on one thing or another. I don’t find any widely-published, routinely updated metrics based on them like I do for things like ENSO or AMO. Is it not conceptually possible to start with a gridded product like GISTemp, HADCRUT4 and/or BEST, combine that with gridded products from ARGO, CERES EBAF and whatnot, then produce a regularly-updated and published temperature time series with internal variability and external forcings isolated?

    I’ve seen examples here and there that such a thing is being discussed. Pielke, Sr. on RC stumping for an AGW index in joules for instance. It also occurs to me that not having data going far enough back in time might be a deal-breaker. Do you, or anyone else here, have any thoughts or pointers do a discussion elsewhere about such a thing?

    JCH,

    I’m all for killing the AMO.

    I was going to ask you: and replace it with what? Then I read up, saw ATTP’s comment, and remembered that I have had some potentially grandiose thoughts on that question myself.

  112. JCH says:

    I’m think I’m sort of with VP.

    If you want about the same things, you can replace with the GISS Land and Ocean Temperature index.

    But to really answer your question, it should be replaced with the PDO, or better yet, with whatever Tsonis used to produce his graph through 1980 to 1985.

  113. Vaughan Pratt says:

    @JCH: I might have been the one who suggested you visit [Tamino’s] blog.

    It was Willard on Feb. 21, 2011. He wrote,

    “While cleaning up my Downloads, I stumbled upon your Killamo document.

    While searching for the meaning of the filename, as I really like it,
    I stumbled upon:

    http://tamino.wordpress.com/2011/01/30/amo/#comment-47666

    Since you have already been judged, would you mind presenting yourself
    so that you get properly quartered?”

    With regard to the quartering, I grossly underestimated Willard’s clairvoyance, and four days later rose to his bait imagining that a little logic would straighten out this obvious misunderstanding. Silly me. What I failed to reckon with is that Tamino is a statistician, not a logician (and seemingly not equipped with much more than freshman physics plus experience with variable star data and climate data).

    It was also Willard who tipped me off to David Young’s comparison with Nic Lewis’s modeling in the present thread.

    He’s done that sort of thing several times. Some day I will be strong and ignore Willard’s helpful tips.

  114. Steven Mosher says:

    Attp.

    I think that’s just the weakest part of his otherwise excellent piece. Instead of “life” I’d substitute civilized human life..not as snappy of course… Hmm and avoid if possible words like narrow range.. Avoid the wobbly language of narrow and wide..
    …. We are approaching the boundary..

    Just nit picking cause Internet.

    On the substance… It’s quite good..

  115. JCH says:

    Willard works in mysterious ways.

    I don’t see a comparison at all between killamo and what Nic Lewis is doing.

  116. Vaughan Pratt says:

    @ATTP: Therefore, what you’ve extracted may not be the full climate signal. Similarly, what you removed may not be purely an internally driven component.

    These are both excellent points, which I expected to be addressing when I showed up at Tamino’s blog. What I didn’t expect was to be treated like a criminal who was not even allowed the courtesy of defending himself. Tamino has what it takes to rise to greatness in a totalitarian society. This is even further from real climate science than the denial community gets.

    My first two AGU posters, in respectively 2011 and 2012, made the first point as follows:

    2011 (Conclusions panel): “1. The Arrhenius-Hofmann law is sufficient to account on its own
    for any trend in global temperature that could be considered correlated with either human population or advances in technology. Our analysis is unable to distinguish influence of aerosols from influence of greenhouse gases.”

    While there is always the possibility of a natural trend closely tracking these human ones, until someone proposes a plausible natural mechanism for such I consider that possibility sufficiently unlikely as to be worth ignoring.

    (Uncertainties panel): I supplemented the foregoing with “4. Analyses of this kind are in general not unique, and alternative analyses may lead to different conclusions. One approach to judging the relative merits of any such competing conclusions is in terms of the quality of the analysis methodologies supporting those conclusions as they come to hand.”

    2012: (Third column): “Whether SAW describes a single phenomenon or several is an excellent question, see our speculation on the right. What we can say is that the three curves in the figure to the upper left give one plausible account of multidecadal climate, MUL, since 1850. This in turn leads to the analysis shown at lower left, namely HadCRUT3 = MUL + SOL + DEC. We are unaware of other analyses claiming millikelvin precision.” (Today I would say “explains” rather than “describes”, using “explain” in its statistical rather than physical sense.)

    So yes, what I extracted may indeed not be “the full climate signal.” I’ve always been aware of this possibility, I have never claimed anything to the contrary, and I’ve always been careful to point this out in any situation calling for care.

    Regarding the possibility that I may have removed a component not internally driven, how could that impact the theory that anthropogenic CO2, ACO2, is the root cause of modern global warming? ACO2 has risen far too steadily for any aspect of it to have been inadvertently removed by my 65-year filter. Slightly attenuated perhaps but there are effective and accurate techniques for correcting for that.

  117. Vaughan Pratt says:

    @JCH: I don’t see a comparison at all between killamo and what Nic Lewis is doing.

    Indeed. Killamo indicates a TCR higher than Nic is comfortable with. Nic prefers complicated models because it’s easier to fit them to the lower TCR’s he favors. I prefer simpler models for the dual reason: it’s harder to incorporate/hide one’s biases in a simple model.

  118. Vaughan Pratt says:

    @JCH: I don’t see a comparison at all between killamo and what Nic Lewis is doing.

    Sorry, I overlooked the context of what Willard has been tipping me off about. As far as Willard’s tips go I don’t believe he intended any comparison, nor did I see one when I first responded. It wasn’t until your remark about comparisons that it even occurred to me to attempt to connect those particular dots. As you can see the Australian in me doesn’t mince words when I connect dots.

  119. Brandon Gates says:

    JCH,

    I’m think I’m sort of with VP.

    Gah. I think I sort of didn’t read upthread far enough and should have.

  120. Vaughan Pratt says:

    @BG: I was going to ask you: and replace [the AMO] with what?

    My last three AGU presentations (2012 poster, 2013 talk, 2014 poster) have all proposed replacing it with quasiperiodic interferences between Earth’s relatively solid inner core and mantle, mediated by the inviscid (low-viscosity) outer core. The details have improved every year, including in the recent months since AGU 2014 (December 16), with concomitant improvements in the quality of fit to HadCRUT4.

    These details explain (a) why the “oscillation” is not really an oscillation in the usual sense but merely events with a characteristic period of a few decades, and (b) why such events have been observed over the past 8000 years by e.g. Knudsen et al (Nature 2011) in the form of “a quasi-persistent ~55- to 70-year AMO”, which is exactly what my theory predicts, over vastly more than 8000 years to boot.

  121. oneuniverse says:

    Andrew Lacis wrote:

    And also note that the ocean cannot cause a decadal warming spurt – the deep ocean is colder than the surface biosphere, so it cannot be a source of heat.

    I’m not sure what argument Dr. Lacis is making here – I assume he’s not emulating the Skydragons, some of whom make a similar argument that the lower temperature of the atmosphere precludes it from exerting a warming influence on the surface. To be clear, changes in the net flow of energy between ocean and atmosphere can alter the temperature of the atmosphere both upwards and downwards. The direction of the net flow doesn’t have to change for the temperature of the atmosphere to change.

    Chen and Tung 2014 (“Varying planetary heat sink led to global-warming slowdown and acceleration”) identify 1980-2000 as a period of accelerated global surface warming, due to less heat being sequestered in the ocean. This is not in agreement with Dr. Lacis’ assertion.

    [..] water vapor, as the climate system’s principal feedback agent, has an exponential dependence on temperature [..]

    The saturation water vapor pressure has an exponential dependence on temperature, but the water vapor pressure itself does not neccessarily follow suit. (At least, it’s not a result derivable from basic physics). Relative humidity, defined as vapor pressure / saturation vapor pressure, has stayed fairly constant over the observational period, but it’s not clear whether this is a persistent feature of the climate system (eg. Simmons 2010 : “These results based on the observational record appear to be at variance with a prevailing modeling view summarized in the IPCC’s Fourth Assessment Report. There, Randall et al. [2007] conclude that humidity in the planetary boundary layer is controlled by strong coupling with the surface and describe a broad-scale quasi-unchanged relative humidity response to forcing by increased greenhouse gases as uncontroversial.”)

  122. Willard says:

    > As far as Willard’s tips go I don’t believe he intended any comparison, nor did I see one when I first responded.

    It would be tough to compare a quest for an explanation with what looks like a PR venture all the way down. My point in recalling Vaughan’s quest is to show that one can very well be skeptical of GCMs and live a fulfilling life, as long as one embraces simplicity. One could even argue that GCMs are not models in the same sense as Vaughan’s, which is why I affectionately call them modulz. This term is borrowed from PDA, who now tries to live a life away from any intoxication, including ClimateBall.

    I am quite confident in relating most of the quarrels over sensitivity matters to PR ventures for two reasons. First, its media coverage overplays its relative importance tremendously. Quotes upon requests, as I need to sleep. Second, I have yet to see a convincing argument to the effect that GCMs would need to be predictive, or that we need to refine them for prudent policy making.

    Auditors just recently realized that sensitivity was an emergent property, which was almost a side-effect of what is being explored. Again, quotes on request. To make matters even worse, all these splendid results will get fed into economical models! This is Mad Men territory.

    No wonder Matt King Coal is so addicted to these ClimateBall episodes.

    ***

    As far as I can tell, GCMs belongs to a quasi-empirical machinery to check how various implementations of how the theory behave. Not unlike a telescope, but instead of looking at an image from the past (we tend to forget that we look at the past), we try to “look” into a slice of possible futures. That’s how Backgammon and Scrabble engines work, but on a quite smaller scale.

    I should end by paying lip service to my mysterious ways, but I’ve ran out of steam for the day.

    Good night,

    W

  123. Vaughan Pratt says:

    @W: I have yet to see a convincing argument to the effect that GCMs would need to be predictive, or that we need to refine them for prudent policy making.

    It’s been my impression that the more parameters in a model, the better it can be fitted to past data and the lower its predictive skill for future data.

    It would be nice to support or refute this impression with some gedanken experiments on manageably sized fictional-but-plausible complex universes and a few relatively simple models thereof with a variety of number of parameters.

  124. Vaughan Pratt says:

    Ok, I’m mystified. What word or part thereof caused my comment of April 12, 6:27 am, to be put into moderation?

    Or did I merely submit too many comments?

  125. Vaughan Pratt said:


    This January I developed a slightly more detailed physical model of the impact of LOD on surface temperature based on a hypothesis about how it influences magma leakage (the amount of leakage adds a parameter), which fitted HadCRUT4 much better.

    As you probably know, Jean Dickey and colleagues at JPL have postulated the attribution of long-term temperature variation to a mechanism related to LOD changes. I do buy into the idea as well, but you won’t see much support for that particular idea on this particular blog. IMO, it is a little too far out for consensus science 🙂

  126. Vaughan Pratt says:

    @WHT: you won’t see much support for that particular idea on this particular blog. IMO, it is a little too far out for consensus science 🙂

    The odd thing is that everyone I’ve spoken with about this has had the same reaction: it sounds fine to them but good luck getting any consensus on it.

    Obviously I need a larger sample. 🙂

  127. izen says:

    Please forgive a comment from ignorance, which will reveal how little scientific insight I have into this matter, I make it in the hope someone can explain why it is probably irrelevant and ignorant.
    I cannot think of a causative mechanism that flows from LOD to climate.
    But I can see causation flowing from climate changes to the LOD because of the changes in sea level, local and global, altering the angular momentum and frictional interaction of the oceans.
    Can anyone clarify this ?

  128. I do recall now how Vaughan got under Tamino’s skin a while back. I would say that as far as climate science is concerned, physics + logic beats statistics. There is just too much strong deterministic behavior for having to invoke statistics to bolster an argument. In particular, I will ignore anyone that tries to push ENSO behavior as a red noise phenomena. That’s a pure deterministic signal governed by the boundary conditions of known forcings.

  129. Vaughan Pratt says:

    @izen: I cannot think of a causative mechanism that flows from LOD to climate.

    Decreasing LOD results in increased oblateness of Earth. Several mechanisms have been proposed based on that distortion, such as heating by friction from bending, increase in thermal conductivity of rocks due to stress, pulling apart of tectonic plates at ocean ridges., etc.

    The mechanism I currently favor does not involve changes in oblateness but rather magma being propelled by increased centrifugal force through ocean ridges and other entry points, suggested to me by Columbia’s Ryan Abernathey during a brief conversation with him and Scripp’s Lynne Talley at AGU 2014. Since the magma starts out a thousand degrees hotter than when it arrives at the ocean floor as pillow lava etc., it doesn’t take much flow of magma over a thirty year period to add enough joules to raise the oceanic mixed layer by a tenth of a degree during that period. This flux is tiny compared to fluxes between the ocean and the atmosphere, whence any cessation of that flow will in due course result in a return to the earlier temperature.

    But I can see causation flowing from climate changes to the LOD because of the changes in sea level, local and global, altering the angular momentum and frictional interaction of the oceans.

    Angular momentum is conserved and hence can’t be altered. What you presumably mean is altering the moment of inertia.

    To increase the LOD by 4 ms by increasing Earth’s moment of inertia would require piling up a ring of elevated water at the equator 100 m high and some large number of kilometers wide (I did the calculation at CE a while back but can’t find it right now, it’s not difficult). This is way beyond anything that happened to the ocean during the 4 ms increase in LOD between 1880 and 1910.

  130. Brandon Gates says:

    Willard,

    Second, I have yet to see a convincing argument to the effect that GCMs would need to be predictive, or that we need to refine them for prudent policy making.

    I have yet to see a convincing argument that CO2 has any effect whatsoever on climate.

    You’re welcome.

    Vaughan,

    Thank you for the readings. Much of it is going whoosh, but from your other comments here I’m beginning to get somewhat an idea of what it is you propose. I’m intrigued.

  131. Brandon Gates says:

    Web,

    In particular, I will ignore anyone that tries to push ENSO behavior as a red noise phenomena. That’s a pure deterministic signal governed by the boundary conditions of known forcings.

    Eyahhh, what I’ve noticed is that there’s some slop about using words like random when what’s really meant is, “we don’t know enough of the forcings but think we have a handle on the boundary conditions, so we punted and modelled it stochastically”.

  132. oneuniverse,

    Andrew Lacis wrote:

    And also note that the ocean cannot cause a decadal warming spurt – the deep ocean is colder than the surface biosphere, so it cannot be a source of heat.

    I’m not sure what argument Dr. Lacis is making here

    What I think he is pointing out is that although there is a lot of energy in the deep ocean, it’s much colder than the surface. Water is also incompressible. Therefore if you want to transfer energy from the deep ocean to the surface, you have to do it via the bulk motion of water itself. However, if you replace 1 cubic metre of water at the surface, with 1 cubic metre of water from the deep ocean, you have to cool the surface. There is no physically plausible way in which you can bring energy from the deep ocean in a way that would increase the temperature of the surface.

  133. VP,

    As you can see the Australian in me doesn’t mince words when I connect dots.

    Yes, the South African in me has a similar problem.

    Some day I will be strong and ignore Willard’s helpful tips.

    Me too, although typically Willard’s helpful tips are quite helpful (if you can work out what it is, at least 🙂 )

  134. A Lacis says:

    It is basic physics that explains how things work (global warming, being one example, among all other things). At least that is message that I was trying to convey in my recent and earlier postings at Climate Etc, and before that, at Andy Revkin’s NY Times blog. Unfortunately, for whatever their reasons may be, far too many people there appear to be singularly and overly focused in their predisposition to remain “skeptical” about nearly every aspect of global warming and climate change, and thus, they are finding it increasingly more difficult to view the climate problem objectively.

    If this were any normal problem in science, the decades of study and analysis that have been already expended would by now have reached a consensus understanding (and there is consensus, for all those who have been engaged in this study). But global warming is no ordinary back-water science problem. Important financial interests are at stake here. And therein lies the problem. For there are those who don’t really want to see a consensus understanding of the global warming problem since any understanding of the impending climate problem by the public (and heaven forbid, the policy makers) is perceived to be detrimental to the financial interests of all those so inclined.

    Climate uncertainties are real of course, and a good working understanding of the natural variability of the climate system is important, but also not going to come any time soon. Furthermore, most uncertainties can cut in all directions. Consequently, the fact that the global energy balance is being shifted away from its long-established equilibrium point by the ongoing increase in atmospheric CO2, does not necessarily mean that the current (long-established) natural variability of the climate system will remain unaffected.

    Whether the prolonged drought that is occurring in California can, or cannot, be linked directly or indirectly to ongoing global warming, is a topic on which current climate models cannot yet be fully definitive. Thus, as part of the substantial realm of uncertainties, there could well be unanticipated climate disruptions the likes of which have not been seen in recorded history. For anyone who would claim to have some understanding of the terrestrial climate system, to blandly state that nothing significant can be happening with climate, is simply not being responsible.

    What we do know and understand quite clearly is that: (1) the concentration of atmospheric CO2 is increasing now by more than 2 ppm/yr (being precisely measured, from about 280 ppm in 1850 to the current 400 ppm); (2) atmospheric water vapor operates according to the Clausius-Clapeyron relation; (3) both atmospheric CO2 and water vapor are strong absorbers of thermal radiation (and their absorption coefficients are accurately known, and have been verified by numerous laboratory measurements); (4) solar irradiance is holding steady (aside from its small 0.1% sunspot cycle variability); (5) energy is conserved (whether or not we can, or cannot, verify that by direct measurement); and furthermore, we know enough basic mathematics and radiative transfer physics to calculate the radiative effects that spring forth from this basic information.

    This provides the foundation for understanding how and why the atmospheric greenhouse effect keeps the global surface temperature some 33 K warmer than its direct Planck equivalent of 255 K, given the absorbed solar radiation. Since radiation is the only means for energy transfer in and out from the climate system, computing the change in global surface temperature that is caused by the increase in atmospheric CO2 is not something that is plagued by any significant uncertainty (other than some uncertainty in the rate of heat transport into the ocean and its effect on the time scale of ice melting, climate processes that are more difficult to model).

    For those inclined to think that climate models have only been good for climate hind-casting, there was the successful climate forecast made by Hansen et al using the GISS climate GCM predicting the global cooling and stratospheric warming (that was later verified by subsequent measurements) following the 1991 Pinatubo volcanic eruption.

    There is plenty of ammunition to go around for making arguments as to what, or if, anything should be done to counteract the ongoing global warming, who it is that might have been, or is now, responsible for creating this problem in the first place, and who will determine now what countermeasures will be necessary, followed by how, why, and when, or if, they are going to be implemented. But agreeing that there is indeed an ongoing global warming problem should by now be abundantly clear to everyone from the available facts and physics, and should therefore not be any part of the ongoing climate arguments.

    Other topics within this growing climate discord, are the hare-brained geoengineering schemes that are proposed as potential remedies (akin to medieval medical practices of purging and blood-letting) to generate volcanic-type aerosols in the stratosphere, to form synthetic clouds from sea water, or even to place scattering particles out in space at the L1 Lagrange point, all designed to reflect some significant fraction of the incident solar radiation back to space, and in this way counteract the growing strength of the greenhouse effect. One could also attempt to pump cold water up from the deep ocean to buy some time. The ocean is after all the cold temperature storage reservoir of the planet with a great deal of heat capacity.

    Unfortunately, all of these geoengineering schemes are inherently unsustainable losing propositions that fail to address the real problem while allowing the climate hole only to be dug ever deeper. They all require enormous outlay of energy and material in order to produce their required radiative effect, which has to be indefinitely sustained, and continuously increased, to keep up with the growing strength of the greenhouse effect, as atmospheric CO2 continues to increase.

  135. Andy,
    Thanks. I think you’ve essentially summarised my motivating for starting this blog (although, I’m not sure I was thinking quite that clearly when I did so).

  136. JCH says:

    The latest one I read was to pipe heat away from the ocean surface to the abyssal ocean. The hiatus forever.

  137. Vaughan on LOD


    The odd thing is that everyone I’ve spoken with about this has had the same reaction: it sounds fine to them but good luck getting any consensus on it.

    No one has any issue attributing causation via correlation of the finer temporal variations of LOD with changes in the atmospheric wind patterns. And since wind is correlated to pressure and pressure to temperature via thermodynamics, the fundamental correlation of LOD to temperature exists.

    Look at this chart put together by Chao

    One sees that the finest resolution in LOD variations is attributed to tidal variations, as tides subtly change the moment of inertia of the earth. There is no doubting this as the correlation and the physics agree.

    At the next resolution one can see the effects of El Nino as that phenomenon induces wind, pressure, and temperature changes in the earth’s atmosphere. There is no doubt as to the attribution here as well.

    But at the largest scale, the attribution is more muddled. In the figure, Chao attributes the LOD to variations in the core. And as Vaughan says, any variations in the internal core moment of inertia would bring on enormous amounts of friction and that this friction would be dissipated partly as heat. It could also be due to shifts in the ocean circulation, although the first-order-calculated strength is weak.

    I use the LOD as a multiple linear regression factor in my CSALT model of global temperature because with a lag of a few years it improves the correlation markedly. Yet, there are many naysayers on this attribution.

  138. Willard says:

    > I have yet to see a convincing argument that CO2 has any effect whatsoever on climate.

    Before the one Andy posted less than an hour after this claim, Brandon Gates, I rather liked NG’s:

    [S]cience doesn’t work by making predictions about future events, for the most part; it makes predictions about observable aspects of the world, things detectable in the present. The amount of trust scientists place in climate models, for example, depends on their ability to simulate relevant aspects of the past and present world. The amount of trust the public places in climate science should depend on the weight of evidence in the past and present world, which is enormous.

    Observation: analyses of global surface temperatures indicate a long-term warming trend.

    Hypothesis: the surface of the Earth is warmer than in the past.

    Testable prediction: phenomena sensitive to Earth’s surface temperature will reflect that increase.

    Results: satellite temperature measurements show similar warming; most glaciers are shrinking; lakes and rivers are freezing later and thawing sooner; oceans are expanding; plant and animal communities are mostly moving poleward.

    Conclusion: the Earth’s surface has been warming.

    So why do many people not believe the Earth’s surface has been warming, and what further evidence or predictions would convince them?

    Observation: Tyndall gas concentrations are increasing in the atmosphere.

    Hypothesis: The rate of increase of such gases is sufficient to cause global temperatures to rise by a couple of degrees by the middle of the next 21st century.

    Testable prediction: A substantial portion of temperature changes so far should be quantitatively attributable to Tyndall gases.

    Results: Spectral radiance emitted to space consistent with Tyndall gas concentrations (confirms ability to calculate radiative forcing); magnitude of Tyndall gas radiative forcing larger than that of all other known forcing agents; observed temperature changes similar in magnitude to those estimated from forcings (confirms ballpark estimates of climate sensitivity); observed pattern of temperature changes match Tyndall gas pattern better than that of all other known forcing agents.

    Conclusion: Anthropogenic global warming is real and significant.

    Based on these scientific predictions that have come true, even the most skeptical scientifically-literate individual ought to be able to conclude that the balance of evidence favors the reality of anthropogenic global warming. So why do many people not believe that anthropogenic global warming is real, and what further evidence or correct predictions would convince them?

    http://blog.chron.com/climateabyss/2011/08/roger-pielke-jr-s-inkblot/

    This formulation has the advantage of being formulated in falsifiable terms. Empirical evidence that it’s good is that the falsificationist at Bart’s (in the death thread I cited earlier) refused to bite.

    There’s also Alley’s detective story that I rather like:

    The interesting bit for your question, Brandon, starts at around 9:00.

    TL;DW — Snowball earths.

    ***

    Perhaps the main reason why I have yet to see a convincing argument regarding the need for modulz predictivity for policy making is that it requires a normative framework. Another reason is that it would bring us closer to what may compromise dominant economic interests. Hence the scientific distractions like the usual sensitivity matters.

    So yeah, I do believe that most of the Climateball episodes using “but science” arguments are mostly squirrel chasing, and that I’m far from being that meta in my own quest.

  139. Vaughan said:
    Vaughan said:


    To increase the LOD by 4 ms by increasing Earth’s moment of inertia would require piling up a ring of elevated water at the equator 100 m high and some large number of kilometers wide (I did the calculation at CE a while back but can’t find it right now, it’s not difficult). This is way beyond anything that happened to the ocean during the 4 ms increase in LOD between 1880 and 1910.

    I tried to do a similar calculation last year and documented on this blog post:
    http://contextearth.com/2014/03/04/decadal-temperature-variations-and-lod/

    This is the geometry that I was considering:

    Fig 1: As a premise for Length of Day (LOD) variations we consider that the rotational moment of inertia changes along the planetary surface. If a band of water positioned along the equator shifts to higher latitude, the rotational moment of inertia decreases and the rotational velocity increases, thus shortening the length of day.

    The general idea is that if global ocean circulation patterns change, these will get reflected in change of LOD. However, the general consensus is that these are far too subtle to make an impact on the observed LOD variations. Again, the favored attribution is to changes in the Earth’s core moment of inertia.

    However, either way, via frictional heat dissipation or cold-water overturning, the jury is still out whether to assign a causal interpretation to decadal temperature variations. The non-answer is to filter AMO or PDO and call that the multi-decadal variation … well, duh, that really helps don’t it? 🙂

  140. izen says:

    @- Vaughn Pratt
    “To increase the LOD by 4 ms by increasing Earth’s moment of inertia would require piling up a ring of elevated water at the equator 100 m high and some large number of kilometers wide (I did the calculation at CE a while back but can’t find it right now, it’s not difficult). This is way beyond anything that happened to the ocean during the 4 ms increase in LOD between 1880 and 1910.”

    I provisionally accept your claim and that of Web Hub that the required magnitude of oceanic shift is to great.
    Which leaves me with the puzzle of the 2004 Tsunami reportedly shortening LOD by 2.7ms.
    It certainly was not 100m high although it was many kilometres wide.

  141. Brandon said:


    Eyahhh, what I’ve noticed is that there’s some slop about using words like random when what’s really meant is, “we don’t know enough of the forcings but think we have a handle on the boundary conditions, so we punted and modelled it stochastically”.

    To paraphrase the former guv, attribution of a behavior such as ENSO to chaos or non-specific stochastic randomness is a crutch for weak-minded people.

    There really is little doubt that strong boundary conditions (i.e. forcing factors) are guiding the equatorial Pacific sloshing behavior, and a simple wave-based differential equation captures the behavior quite adequately:

  142. izen,
    I think it was microseconds, not milliseconds.


  143. … the required magnitude of oceanic shift is to great.

    Yet if such a shift in the ocean’s spatial distribution of water did occur, can you imagine the impact this would have on the global temperature? That said, consider that “something” is causing the multidecadal change in LOD and then consider what kind of heat dissipation due to friction, magma, etc would accompany this shifting of volume. Realistically, accounting for +/- 0.1C over the last 100 years is all that we need to consider.

    This topic may be more in the realm of geophysicists than climate scientists working GCMs.

  144. Steven Mosher says:

    yes, willard, I have noted that many heads explode when you tell them that you dont need models to set policy, and further that you can use bad models to set policy.

    perfecting and optimizing policies? ah different question.

    I would make the side point that modelers (some at least) can get sucked into the dream of perfecting policy. It can be, at least in my experience, part of the problem.
    .
    I face this . My models (utterly different field) suck. Well, I can tell you which direction to move ( up or down) but can’t tell you exactly how much.. Of course the models pretend to know how much..
    Move 14.67 units.. but.. it really is only skilfull enough to say “move up fool !!” so there is
    guess work,, and more data.. and more work.. in search of a less worse answer. But today
    folks need an answer so they use the model. Smart people know the limits and lower their expectations.
    Nevertheless there are folks who over believe in the model. They can cause the most heartache.
    Look, My pigs are barefaced pigs. I hate when guys try to put lipstick on them.

    what do climate models say: Less emission please.!! well duh, we knew that long ago.
    one could argue that pretending to be tools of policy optimization, was a bad move.
    however, now folks are committed to that main line .. it’s hard to go back and start with d4.

  145. Brandon Gates says:

    Willard,

    This formulation has the advantage of being formulated in falsifiable terms.

    With an emphasis on past observation, which is where I think it should be. I have made the argument many times that we don’t need teh modulz (I so want to steal that, but won’t) at all address AGW since we can see all we need to see in the rear view mirror. That argument draws a depressing number of blank stares. I hadn’t thought of it in normative terms as you mention … I think I like … yes I do like where that leads. Interesting, thanks.

    Empirical evidence that it’s good is that the falsificationist at Bart’s (in the death thread I cited earlier) refused to bite.

    Which I’ve yet to read in full, and feel remiss for it. It’s exactly the sort of thing I need to read more of because it’s a common argument, and one which I think is most accessible for me to argue as a lay person.

    So yeah, I do believe that most of the Climateball episodes using “but science” arguments are mostly squirrel chasing, and that I’m far from being that meta in my own quest.

    I’m not sure I’ve yet to understand your particular use of meta, but I sense it has shades of playing the game for the sake of the game. Which IS me in spades — has to be, otherwise I get discouraged. For various reasons, esp. including lack of technical depth, I’m constantly (hyper-)vigilant of my tactical footing out in the wild. Constructions such as “I haven’t seen compelling evidence of x” are the sort of argument which get hammered flat by pro-rhetoricians as well as more honest brokers with any facility in formal logic. I try to use, “I don’t find evidence y for x compelling because of a, b, c.” Invariably that will be answered with a “whuttabbout q” or “look, squirrel”, but as that happens either way, I prefer to be on my toes rather than my heels.

    Parahprasing R. Lee Ermey, you have your games, I have mine. One of mine is watching for exposed flanks and gaps in the line.

    Thanks for your feedback and for increasing the size of my reading and viewing list. Cheers.

  146. Brandon Gates says:

    Andy,

    Thank you for your clear and concise argument above. I’ve added it to my quotable expert bin, same as your comment over at CE. My deepest thanks for your actual work. Best regards.

  147. Rob Nicholls says:

    Many thanks BBD for the info. Hönisch et al. (2012) is really interesting, and seems to give a flavour of some of the complexities and difficulties involved in making inferences about past changes in ocean chemistry. The last paragraph says “the current rate of (mainly fossil fuel) CO2 release stands out as capable of driving a combination and magnitude of ocean geochemical changes potentially unparalleled in at least the last ~300 My of Earth history, raising the possibility that we are entering an unknown territory of marine ecosystem change.” Quite scary, given that the last 300 Million years includes the utterly devastating Permian-Triassic mass extinction.

    I read the Skeptical Science review on the mass extinction event at the end of the Cretaceous, which I thought was awesome.

    It occurs to me sometimes that, quite apart from AGW and ocean acidification, a lot of ecosystems and species are being put under a lot of strain by other aspects of human activity; I would guess that this would increase the chances of significant extinctions / mass extinction due to AGW/ocean acidification under BAU (unless we’re very lukey/lucky.)

  148. David Young says:

    Vaughan,

    1. There is a climate audit post where Nic Lewis is interacting with people and it would be enlightening if you went there and discussed your method and particularly why you think its better than Nic’s which is really I believe the standard one used for at least a decade in the field. Of course, yours could still be better.

    2. The examples you seek of more complex and “accurate” model failures are all over science and well known. In high Reynolds’ number turbulent flow modeling there are integral boundary layer models, Eddy viscosity models for Reynolds’ averaged Navier-Stokes, and Reynolds’ stress models. In principle, RSM’s have the potential to accurately model essentially arbitrary viscous stress fields, whereas eddy viscosity models use isotropic viscosity. IBL and eddy viscosity have problems but are very widely used, whereas RSM’s are occasionally trotted out as “more physics” and some cherry picked cases shown where they are “more accurate” – but RSM’s are simply not used at all because there are too many parameters and too little data. You have to look very hard in the literature to find this talked about honestly however. There are a lot of reasons why people like to believe that “including all the physics” will give better results. Like a lot of common prejudices (and marketing tools) this one is certainly only true in special circumstances.

  149. BBD says:

    Rob N

    Very glad to be of help.

    Those who blithely claim that a large, rapid increase in CO2 emissions will have little or no ecological impact are ignoring an awful lot of palaeoclimate evidence to the contrary.

  150. BBD says:

    DY

    why you think its better than Nic’s which is really I believe the standard one used for at least a decade in the field.

    The fact that NL’s method produces results incompatible with palaeoclimate behaviour suggests that it is deeply flawed.

    You really should think about this, rather than deny it. The quality of your understanding will be deepened accordingly.

  151. David Young says:

    BBD, My understanding will be deepened when Vaughan publishes his results and/or subjects them to criticism and comparison by others in the field.

  152. BBD says:

    DY

    Much though I applaud Vaughan’s efforts, I don’t agree that they – and they alone – are required to understand that lowball sensitivity estimates are implausible.

  153. mwgrant says:

    Re: Steven Mosher @ April 12, 2015 at 4:16 pm

    Them’s the facts, pure and simple.

  154. JCH says:

    The PDO appears to influence the direction of the global mean surface temperature.

    AMO lovers, meet your aimlessly wandering bride.

  155. DY,

    There is a climate audit post where Nic Lewis is interacting with people and it would be enlightening if you went there and discussed your method and particularly why you think its better than Nic’s which is really I believe the standard one used for at least a decade in the field. Of course, yours could still be better.

    WHT, said something that – IMO – illustrates why trying to interact on CA is generally pointless,

    as far as climate science is concerned, physics + logic beats statistics.

    I don’t have an issue with people delving into the details of some analysis to try and understand what was done, or trying to improve some analysis. However, at the end of the day, this is a physics problem and applying complicated statistics doesn’t mean that the result makes any sense, or finding some technical statistical fault with an analysis doesn’t mean the result is completely wrong. The day that Steve McIntyre and Nic Lewis recognise this, is the day I’ll take them more seriously. I’m not holding my breath.

  156. Willard says:

    > There are a lot of reasons why people like to believe that “including all the physics” will give better results.

    One reason is that an important goal in the modulz’ arms race is to come up with one all batteries included. That is, including all physics is the best result to some extent. Or perhaps not, since the best modulz of our planet would be a planet, preferably the very same planet.

    I’d settle for earth holodecks. Billions upon billions of them.

  157. Brandon Gates says:

    Web,

    There really is little doubt that strong boundary conditions (i.e. forcing factors) are guiding the equatorial Pacific sloshing behavior, and a simple wave-based differential equation captures the behavior quite adequately:

    I muddled terms in my original comment, didn’t I. Let’s see if I’ve got it sorted. The physics view of boundary conditions is that they’re set by applying known parameters to a differential equation. The more stats-minded (like me) see “boundary” and think “observed minima/maxima”.

    What I accept absolutely because it makes the most sense to me is that ocean/atmospheric couplings are deterministic, meaning that a stochastic model with some autocorrelation thrown in to look like the real thing explains absolutely butkus. In that sense, I think “random” means “stuff we can’t explain causally, but we can tell you what it looks like”.

    My question to you now is, how does one apply a model such as you’ve shown to an AOGCM, if at all?

  158. Unfortunately, for whatever their reasons may be, far too many people there appear to be singularly and overly focused in their predisposition to remain “skeptical” about nearly every aspect of global warming and climate change, and thus, they are finding it increasingly more difficult to view the climate problem objectively.

    Dr Lacis,
    From the IPCC, Hansen’s 1988 testimony, and observations, I compute the following trends:
    ———————————————————————
    MODEL: 4.2C/century, (through 2100), IPCC5 (RCP8.5)
    MODEL: 4.0C/century, (through 2100), IPCC4 ‘High Scenario’
    MODEL: 3.2C/century ( since 1979 ), Hansen A
    MODEL: 2.8C/century ( since 1979 ), Hansen B
    MODEL: 2.0C/century, (‘next few decades’), IPCC4
    MODEL: 1.9C/century ( since 1979 ), Hansen C
    MODEL: 1.8C/century, (through 2100), IPCC4 ‘Low Scenario’
    ———————————————————————
    Observed: 1.6C/century (since 1979), NASA GISS
    Observed: 1.5C/century (since 1979), NCDC
    Observed: 1.4C/century (since 1979), UAH MSU LT
    Observed: 1.3C/century (since 1979), RSS MSU LT
    Observed: 1.3C/century (since 1979), RATPAC-B 850 millibars
    Observed: 1.2C/century (since 1979), RATPAC-B 500 millibars
    MODEL: 1.0C/century, (through 2100), IPCC5 (RCP2.6)
    Observed: 1.0C/century, (since 1979), RATPAC-B 300 millibars
    Observed: 0.8C/century (since 1979 ), RSS MSU MT
    Observed: 0.5C/century (since 1979 ), UAH MSU MT
    ———————————————————————
    Denies Global Warming: 0.0C/century
    ———————————————————————

    To be sure, there is warming.
    However, that warming is between zero and the low end predictions.
    It appears there is subjectivity from those imagining rates of warming
    less than as well as greater than what we observe.

    Whether the prolonged drought that is occurring in California can, or cannot, be linked directly or indirectly to ongoing global warming, is a topic on which current climate models cannot yet be fully definitive. Thus, as part of the substantial realm of uncertainties, there could well be unanticipated climate disruptions the likes of which have not been seen in recorded history. For anyone who would claim to have some understanding of the terrestrial climate system, to blandly state that nothing significant can be happening with climate, is simply not being responsible.

    Many significant things can, have, and will occur with the climate. However, they apparently don’t have anything to do with global energy imbalance or global temperature. The California drought is due to a lack of precipitation. Global temperature tells you nothing about that. The fact that the general circulation has numerous quasi stable states which sometimes fluctuate does describe the lack of precipitation. Those who study the governing equations of motion of the atmosphere know that there is no term for what the global average temperature is – because it is not relevant. Are you proposing such and equation here? It is displeasing for some to say that weather is chaotic and we can’t predict it, but we do know why it is chaotic and such chaos will exist for all temperatures as long as we have an atmosphere, surface friction and a rotating, revolving, sunlit earth. Some want to have chaos both ways:
    1. climate models can’t predict circulation, but that doesn’t matter because everything will average out.
    2. climate models can’t predict circulation, and that means somehow temperature will change circulation in drastic ways.

    What we do know and understand quite clearly is that:
    (1) the concentration of atmospheric CO2 is increasing now by more than 2 ppm/yr (being precisely measured, from about 280 ppm in 1850 to the current 400 ppm);

    Yes, though because of the logarithmic forcing response, an exponential increase in greenhouse gasses is necessary just to maintain even a constant rate of warming.

    (2) atmospheric water vapor operates according to the Clausius-Clapeyron relation;
    That is incorrect, at least by omission. Clausius-Clapeyron determines the saturation point, not the content. The motion of the atmosphere largely determines what the water vapor content is for a given location. Indeed, as I displayed in the graphic above, most of the 500 millibar atmosphere has a water vapor content closer to zero than to saturation.

    (3) both atmospheric CO2 and water vapor are strong absorbers of thermal radiation (and their absorption coefficients are accurately known, and have been verified by numerous laboratory measurements);
    Yes, of course. And strong absorbers are strong emitters.
    And adding CO2 makes the top 10% of the troposphere lose, not gain radiative energy.
    And adding H2O makes the top 75% or so of the troposphere lose, not gain radiative energy.

    The rates of warming have been relatively low ( comparable to the warming of 1910 through 1945 )

    That’s as it should be – the forcing rates have been relatively low ( peaking around the early 1980s and stable at the ‘B1’ scenario since.

    And the CO2 emissions are declining for most of the developed world, now even China.

    Modest warming is irrelevant.

  159. Eddie,
    We’ve been through this before. Can we not do it again? Hansen’s model had an ECS of 4.2K, above the now accepted range.

    And the CO2 emissions are declining for most of the developed world, now even China.

    Good, but that doesn’t change that continuing to increase (or even maintain) our emissions carries risks.

    I will say that I do find your certainty about things like the California drought irritating. If you want to engage in a scientific discussion you have to impose the same level of uncertainty on your views as you expect of others.

  160. David Young says:


    The examples you seek of more complex and “accurate” model failures are all over science and well known. In high Reynolds’ number turbulent flow modeling there are integral boundary layer models, Eddy viscosity models for Reynolds’ averaged Navier-Stokes, and Reynolds’ stress models. …..
    ” ….. and on and on and on

    #WHUT kind of FOO is this ? That is quite a fallacious argument to bring up difficulties in your own field and think they apply to climate. It’s not our problem that you have problems buddy.

    The fact is that something as complex and erratic as ENSO actually follows macro laws of physics subject to forcing conditions that is not much more complicated than the wave equation.

    I assume that you do know how to solve a plain differential equation, eh?

  161. Vaughan Pratt says:

    @WHT: And as Vaughan says, any variations in the internal core moment of inertia would bring on enormous amounts of friction.

    Yes, that was my view in 2012. I was expanding on this when I realized the length had gotten out of hand, so I’ve moved my reply to a sort of write-up at

    http://clim.stanford.edu/lod.html

    I’ll add some clarifying figures to it shortly, and then try to turn it into something more publishable.

    I use the LOD as a multiple linear regression factor in my CSALT model of global temperature because with a lag of a few years it improves the correlation markedly.

    Yes indeed. My current model does something similar but more physically motivated than just a delay, more details in the write-up.

  162. David Young says:

    ATTP, Yes to some extent. Statistics by itself doesn’t help much. But statisticians can be very helpful on the team. I am working with 2 right now and they are very bright and have provided some very good insights. The problem in climate science has been that this has very rarely been the case. We can list uniform priors as a prime example where honest involvement of statisticians could have avoided some obvious errors and inflated estimates for sensitivity. I know you don’t want to talk about Mann, but that is another example where statistician involvement at the onset could certainly have saved a lot of wasted energy.

    However, Nic is I think well versed in the math and arguments for these simple models and what he is discussing goes very far beyond statistics. I really doubt Otto and Allan wouldn’t have included him in their paper if he was not competent.

    What I really think is needed here is for people like Nic and James Annan to get together and try to reconcile paleoclimate with recent observationally based estimates. I personally find it a stretch to think we can know albedo from the deep past for example.

  163. David Young says:

    Webby Webby, will you ever change? Nick Stokes just sent me a reference for boundary layer calculations in GCM’s and it uses eddy viscosity models just like “our” CFD does. Of course as Nick has said, its really all largely the same fundamentals. This is not hard and ignorance does not justify sarcasm and expressions of lack of personal emotional control.

  164. DY,

    But statisticians can be very helpful on the team.

    Yes, I agree and nothing I’ve said has suggested otherwise.

    The problem in climate science has been that this has very rarely been the case. We can list uniform priors as a prime example where honest involvement of statisticians could have avoided some obvious errors and inflated estimates for sensitivity.

    Rubbish. Illustrating that there have been some poor statistical practices does not imply that it has rarely been the case. There are plenty of good statisticians working in climate science.

    I know you don’t want to talk about Mann, but that is another example where statistician involvement at the onset could certainly have saved a lot of wasted energy.

    I doubt it would have made any difference. I don’t believe that the attack on MBH98 was motivated by a desire to do sound statistics. You can live in your fantasy world where you believe that to be the case. I’ll remain in the real world where people appear to attack anything that present results that they perceive to be inconvenient. Finding a problem with a complex analysis that has not been done before is probably quite easy.

    Nic is I think well versed in the math and arguments for these simple models and what he is discussing goes very far beyond statistics.

    Yes, he argues strongly for them and rarely acknowledges the caveats.

    I really doubt Otto and Allan wouldn’t have included him in their paper if he was not competent.

    I don’t doubt his competence, but that he was included on a multi-author paper isn’t really strong evidence of its existence.

    What I really think is needed here is for people like Nic and James Annan to get together and try to reconcile paleoclimate with recent observationally based estimates.

    What I think is needed is for people to simply do science, be they Nic Lewis, James Annan, or anyone else. However, if Nic Lewis continues to writes post on Climate Audit where he insults other climate scientists (such as Marotzke & Forster), he may well find (quite justifiably) few who would actually be willing to work with him.

  165. Brandon Gates says:

    David Young,

    This is not hard and ignorance does not justify sarcasm and expressions of lack of personal emotional control.

    Hmm ….

  166. > I don’t doubt his competence, but that he was included on a multi-author paper isn’t really strong evidence of its existence.

    Well, it does argue in favor of the existence of Nic Lewis. Or does it?

    G. W. Peck is a pseudonymous attribution used as the author or co-author of a number of published mathematics academic papers. Peck is sometimes humorously identified with George Wilbur Peck, a former governor of the US state of Wisconsin.

    Peck first appeared as the official author of a 1979 paper entitled “Maximum antichains of rectangular arrays”.[2] The name “G. W. Peck” is derived from the initials of the actual writers of this paper: Ronald Graham, Douglas West, George B. Purdy, Paul Erdős, Fan Chung, and Daniel Kleitman. The paper initially listed Peck’s affiliation as Xanadu, but the editor of the journal objected, so Ron Graham gave him a job at Bell Labs. Since then, Peck’s name has appeared on some sixteen publications, primarily as a pseudonym of Daniel Kleitman.

    http://en.wikipedia.org/wiki/G._W._Peck

    I think I recognize him around 2:45 in this report for the Proceedings of the Xanadu Society:

  167. BBD says:

    Turbulent Lucifer

    And the CO2 emissions are declining for most of the developed world, now even China.

    But not so much that you’d notice.

  168. BBD says:

    DY

    I personally find it a stretch to think we can know albedo from the deep past for example.

    We’ve been through this already.

  169. BBD,
    Yes, DY doesn’t appear to be familiar with the “argument from incredulity” fallacy.

  170. > we don’t need teh modulz (I so want to steal that, but won’t)

    Please do. Don’t forget the longer version: “teh modulz are stoopid.”

    ***

    > It’s exactly the sort of thing I need to read more of because it’s a common argument, and one which I think is most accessible for me to argue as a lay person.

    If you like falsificationnism, I’ve got a collection entitled PopperForBloggers:

    http://neverendingaudit.tumblr.com/tagged/PopperForBloggers

    Searching to eliminate stuff from it (it’s hard to keep at 3,333 notes), I stumbled upon this bit from Vaughan:

    But Gerard, there is an easy way to falsify this theory. If increasing CO2 is supposed to be making the weather fluctuate more violently, and it returns to the relatively calm state it was in before 1960, as we would all love it to do, then you have your falsification right there!

    If on the other hand it continues to get more and more violent, then those of you continuing to say there’s nothing out of the ordinary with the weather will look more and more stupid with every passing year, because what the so-called alarmists were warning about will be coming to pass and you will be refusing to acknowledge that it is happening. Your head will be in the sand.

    In the parable of the boy who cried wolf, the wolf eventually put in an appearance.

    http://neverendingaudit.tumblr.com/post/31982609327

    Don’t forget the quote where Popper himself admits that no conclusive disproof of a theory can ever be produced.

    ***

    I’ll try to return to your other remarks, Brandon Gates, but I need to get other things done first.

  171. > Modest warming is irrelevant.

    Here would be one way to substantiate that argument by assertion which transforms Koonin’s “but it’s small” with “but it’s moderate”:

    Climate hysterics seem to be in denial that:

    1. observed warming is lower than the low end predicted rates
    2. The models have failed in meaningful ways ( Hot Spot, albedo, rate, … ).
    2. Climate adversity has always occurred, for reasons that have nothing to do with global energy balance.
    4. The imagined effects of warming have not transpired ( largely because global average temperature is just not an important term to climate, certainly not at the rates we observe ).
    5. Humanity certainly did OK during the Holocene Climatic Optimum, which had hotter northern summers for millenia. Again, temperature wasn’t all that important.
    6. Carbon dioxide enhances plant growth, crop yield, and drought tolerance.
    7. More people die during the cold season than the warm season.
    8. Rates of forcing have declined from their peak and are likely to decline further.
    etc. etc.

    Hysterics are hysterical precisely because they are deniers.

    http://judithcurry.com/2015/04/12/week-in-review-politics-and-policy-edition/#comment-692979

    Comments like these are a real time saver for the Contrarian Matrix:

    https://contrarianmatrix.wordpress.com/

    Many thanks!

  172. Willard, well, the d word does get inflammatory.

    But these are all settled science, so I’m sure denying them doesn’t apply to you.

    1. observed warming is lower than the low end predicted rates
    2. The models have failed in meaningful ways ( Hot Spot, albedo, rate, … ).
    3. Climate adversity has always occurred, for reasons that have nothing to do with global energy balance.
    4. The imagined effects of warming have not transpired ( largely because global average temperature is just not an important term to climate, certainly not at the rates we observe ).
    5. Humanity certainly did OK during the Holocene Climatic Optimum, which had hotter northern summers for millenia. Again, temperature wasn’t all that important.
    6. Carbon dioxide enhances plant growth, crop yield, and drought tolerance.
    7. More people die during the cold season than the warm season.
    8. Rates of forcing have declined from their peak and are likely to decline further.
    etc. etc.

  173. Eddie,
    How many of your points are actually correct?
    1. Seems wrong to me.
    2. Albedo?
    3. Okay, so what?
    4. What imagined effects? I don’t think anything was predicted to have transpired now.
    5. Again, so what? We’re projected to get much warmer under a high emission scenario.
    6. Hmm, not my area but not convinced that it’s quite that simple.
    7. Except we have a pretty definite upper limit.
    8. Seems you’re willing to make confident projections of your own, while dismissing those of others.

  174. David Young says:

    ATTP, I know this is a sensitive subject for you but it seems to me very true that in many fields of science, including CFD and climate science, professional (independent) statisticians are not involved in research where statistics is in some cases central to the findings.

    In CFD, its just that the quality of the data is only mediocre and the numerical experiments so poorly thought out or with very small sample sizes that statistical analysis is often simply not justified. This is a strong condemnation of the literature in this field for me anyway. However, one could argue that one could compare methods without statistical analysis if you took care to eliminate or minimize sources of error, such as coarse grids, poor residual convergence, poor boundary resolution, etc.

    In medicine this was a problem for a while and now it is difficult to get a study published without independent statistician involvement and that’s a good thing. Can we at least agree on that?

  175. DY,
    It’s not a particularly a sensitive subject for me. It annoys me to have people outside a field pontificate about what would make it better, especially when they seem to have an undisclosed bias of their own. You seem to ignore that climate science is made up of a diverse group of people with a wide range of skills, including statistics. Also, from what I’ve seen, many of the supposed statistical criticisms have been flawed themselves. The criticism of Marotzke & Forster being the most recent example. I’m not aware of Nic Lewis withdrawing his claim of a trivial and obvious statistical mistake. Feel free to prove me wrong.

    In medicine this was a problem for a while and now it is difficult to get a study published without independent statistician involvement and that’s a good thing. Can we at least agree on that?

    Yes, but at what stage does an independent statistician become part of the field? This is the bit that you seem to overlook. A good number of the people in my field are extremely skilled at statistics and yet they would not call themselves statisticians. You can’t keep adding new independent statisticians. What you do is improve the skills of those involved in the field already.

  176. David Young says:

    ATTP, I was surprised Nic didn’t add a correction to his post saying that he may have been wrong about the circularity and that more study was required of the result. But this behavior, while not what I prefer, seems to be the norm in climate science. I’ve never seen any corrections of the use of uniform priors, but maybe I missed them. I think people simply stopped doing it.

  177. > well, the d word does get inflammatory.

    “Hysterical” is so much more moderate it becomes irrelevant, I guess. I might as well repost my response at Judy’s:

    > Hysterics are hysterical precisely because they are deniers.

    The history of the concept of hysteria’s a bit more complex than that:

    Hysteria is undoubtedly the first mental disorder attributable to women, accurately described in the second millennium BC, and until Freud considered an exclusively female disease. Over 4000 years of history, this disease was considered from two perspectives: scientific and demonological. It was cured with herbs, sex or sexual abstinence, punished and purified with fire for its association with sorcery and finally, clinically studied as a disease and treated with innovative therapies. However, even at the end of 19th century, scientific innovation had still not reached some places, where the only known therapies were those proposed by Galen. During the 20th century several studies postulated the decline of hysteria amongst occidental patients (both women and men) and the escalating of this disorder in non-Western countries. The concept of hysterical neurosis is deleted with the 1980 DSM-III. The evolution of these diseases seems to be a factor linked with social “westernization”, and examining under what conditions the symptoms first became common in different societies became a priority for recent studies over risk factor.

    http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3480686/

    It’s always nice to see Denizens rediscover the old classics.

    Please, do continue.

    http://judithcurry.com/2015/04/12/week-in-review-politics-and-policy-edition/#comment-693077

  178. DY,

    But this behavior, while not what I prefer, seems to be the norm in climate science.

    A poor excuse, IMO.

    I’ve never seen any corrections of the use of uniform priors, but maybe I missed them. I think people simply stopped doing it.

    I’m unaware of this uniform prior situation (well, other than I know they were used), but there’s a vast difference between a technique turning out to be poor and people changing what they do, and someone making a strong claim about a trivial, embarassing error in another paper. The first is normal in science; you don’t need to apologise for discovering that there is a better technique than the one you’ve used in the past. On the other hand, it’s decent to apologise if you’ve pubicly accused others of making a stupid, embarassing mistake when they, in fact, did not. In fact, I would argue that the ideal would be that you don’t make the claim in the first place.

  179. David Young says:

    Of course, its annoying. That’s the whole idea. The idea that experts are immune to normal human frailty is an idea that has no place in modern science. The involvement of independent statisticians will continue to increase the credibility of papers and that’s a good thing. Quibbling about when a statistician ceases to be a statistician is not helpful. Statistics is a very big field and you will never have trouble finding a “real” statistician to advise your team.

    All this involves more work, delays publication, and makes people upset. Life is like that sometimes. The real problem is that many papers in some fields don’t have adequate data quality or good enough experimental design to withstand such scrutiny.

  180. BBD says:

    Hi David

    Thanks for ignoring my comment!

  181. David Young says:

    I need to exit this discussion as I need to work on our paper about statistical analysis of CFD. 🙂

  182. Vaughan wrote:


    Yes indeed. My current model does something similar but more physically motivated than just a delay, more details in the write-up.

    Nice write-up.

    FYI, I use a damped exponential lag of 4 years. In other words, essentially a convolution of the LOD with the impulse response of a damped exponential with a time constant of 4 years. This becomes a temporally smeared response in temperature to LOD changes.

  183. > Statistics is a very big field and you will never have trouble finding a “real” statistician to advise your team.

    Well, sometimes, somebody somewhere does:

    Once upon a time, I was doing some research and a statistics issue came up. I sought to engage someone in our statistics department. I thought that the issue might be of sufficient theoretical interest to make it worth a statistician’s time. So I contacted someone I knew on the statistics faculty and asked if he could refer me to someone.

    The conversation went something like this: Statistics person: “Sure thing. How much money do you have?” Me: “What do you mean?” Statistics person: “I doubt you could find anyone to work with you for less than $20,000.”

    This was not a bribe, per se, but would go to cover the salary of the statistician for the weeks that he or she would spend working on my problem.

    I didn’t have the $20,000 to spare from my research grant. Hence, I did not engage a statistician.

    Moral: Engaging a statistician is easier said than done.

    At the moment, I have a great problem in temporal-spatial statistics that I’m working on, but no grant money for the problem. If anybody out there knows a statistician familiar with that area that will work for free, please put them in touch.

    http://neverendingaudit.tumblr.com/post/9005026590

    ***

    The discussion has turned normative again, out of a sudden. Let’s wonder why.

  184. BBD says:

    The idea that experts are immune to normal human frailty is an idea that has no place in modern science.

    High frequency whistle on my audio gear. What can it be? [Opens signal analyser and studies screen closely]. Ah! Groupthink. And a… torquing of the science?

  185. DY,

    Of course, its annoying.

    I don’t think you understand my use of the term. It’s annoying that someone who I think does not understand how science actually works has the gall to pontificate about what would make it better.

    The idea that experts are immune to normal human frailty is an idea that has no place in modern science.

    Obviously! The same applies to independent statisticians.

    The involvement of independent statisticians will continue to increase the credibility of papers and that’s a good thing. Quibbling about when a statistician ceases to be a statistician is not helpful

    They’re already there is my point! And how many climate science papers does an independent statistician help with before they’re just a climate scientist and you have to then go and find another one because of some silly idea that you need them to help the statistically illiterate climate scientists.

    Statistics is a very big field and you will never have trouble finding a “real” statistician to advise your team.

    Again, they’re already there. If I went to find some supposedly independent statistician, I’d probably have to spend all my time explaining the physics to them and explain why their supposed superior statistical method wouldn’t work.

    My own view is that this “independent statisticians would make everything so much better” meme is nonsense. If some statisticians want to start doing climate science, good. Do it. There are many who already do. Insisting that some “independent” group should oversee in some way to help is just silly.

  186. I need to exit this discussion as I need to work on our paper about statistical analysis of CFD. 🙂

    Who’s your independent statistician?

  187. BBD says:

    Turbulent Lucifer

    Can you point me to the reduction in the atmospheric concentration of CO2 in this observational record? I seem to be missing something.

  188. BBD says:

    Willard

    Moral: Engaging a statistician is easier said than done.

    Too true. Even Tamino charges for his services, according to the sidebar on his blog.

  189. BTW, if you’re worried that you’re a hysteric, here’s a good test.

    If you think this guy makes sense and is not an idiot with more money than brains,
    you may be a hysteric:

  190. Brandon Gates says:

    Willard,

    Please do. Don’t forget the longer version: “teh modulz are stoopid.”

    If I use it, you will be cited. And I think I’ll probably use it.

    Don’t forget the quote where Popper himself admits that no conclusive disproof of a theory can ever be produced.

    Try telling that to this guy: You appear to agree that the equilibrium temperature cannot be observed but feel that one can assign a value to the equilibrium temperature by assigning a value to ECS.. The hypothesis of the constancy of ECS is, however, non-falsifiable in view of the lack of observability of the equilibrium temperature. Thus this hypothesis is unscientific..

    I’ll try to return to your other remarks, Brandon Gates, but I need to get other things done first.

    But of course, do as you will. You’ve been more than helpful with the info, for swapping war stories and as a sounding board. Carry on.

    In the off chance someone here doesn’t already have it, my return contribution is: http://strangebeautiful.com/other-texts/popper-logic-scientific-discovery.pdf

  191. > Even Tamino charges for his services, according to the sidebar on his blog.

    You’re right, BBD. Even Tamino, who’s not even a statistician, but a blogger with an attitude:

    http://neverendingaudit.tumblr.com/post/12239577107

    Perhaps commenters should charge too.

    There could be clients willing to pay to have comments. There could be clients willing to pay not to have comments. Imagine how much money one could make by honestly brokering the two types of clientèle!

  192. Brandon Gates says:

    ATTP,

    Who’s your independent statistician?

    Bet you a fiver his initials are WMB.

  193. oneuniverse says:

    ATTP:

    There is no physically plausible way in which you can bring energy from the deep ocean in a way that would increase the temperature of the surface.

    A reduction in the net flow of energy from surface to ocean is all that’s needed to raise the surface temperature, though. That’s not only plausible, but has been observed over a multi-decadal timescale (as discussed in the Chen and Tung I referred to earlier).

    David Young:

    “I personally find it a stretch to think we can know albedo from the deep past for example.”

    BBD:

    “We’ve been through this already.” [referring to Hansen and Sato 2012]

    H&S2012 acknowledge that “[..] interpretation of an empirical climate sensitivity that treats natural aerosol changes as a forcing is complex, and the error bar on the derived sensitivity is necessarily large.”. They then go on to make an assumption that allows an estimate with much smaller error bars to be derived, namely, that all aerosol and other non-ice albedo-related radiative changes can be treated as fast-feedback responses to temperature. They justify the assumption as follows :

    Of course, there are aerosol changes on long time scales; for example, some periods are dustier than others. But these aerosol changes are analogous to the cloud changes that occur between climates with or without an ice sheet. Changed surface conditions (e.g., ice sheet area, vegetation cover, land area, and continental shelf exposure) cause clouds and aerosols to exhibit changes over long time scales, but the adjustment time of clouds and aerosols to surface conditions is fast.

    A fast cloud/aerosol adjustment time to conditions, howerver, doesn’t mean that radiative forcing changes caused by the changing conditions are a fast feedback. To actually justify the assumption, the changing conditions themselves need to be fast responses to temperature. This view of aerosols/vegetation/land change as a fast-feedback to temperature isn’t supported in the IPPC AR5 WG1 Chapter 5 paleoclimate archive review.

  194. oneuniverse,

    A reduction in the net flow of energy from surface to ocean is all that’s needed to raise the surface temperature, though.

    Yes, I realise. I was simply referring to the reason why you can’t take energy from the deep ocean, and warm the surface. You’re correct, though, that changing the flow of energy within the ocean can result in surface warming.

  195. BBD says:

    oneuniverse

    H&S12 is generous with its assessment of uncertainty and ends up with a lower bound of ~2C for ECS. Surely even this provides little wriggle room for policy?

  196. BBD says:

    Willard

    You’re right, BBD. Even Tamino, who’s not even a statistician, but a blogger with an attitude

    I certainly agree that he is a blogger with an attitude and I do not condone that. But I didn’t realise that he wasn’t a statistician. Do you know what his qualifications are?

  197. > Do you know what his qualifications are?

    I believe I do, BBD, but I don’t think formal qualifications matter much in general. It’s not what you did when you were young in school that matters, but what you can show for yourself.

    This underlines one problem with DY’s suggestion: requiring statisticians on board of every project would lead to more corporatist crap surrounding grants. We have enough of that already.

    I think this kind of INTEGRITY ™ concerns don’t fit so well with Grrrowth solutions. Grrrowth should be are main concern. Grrrowth is ultimately the only concern.

    Thank you.

  198. BBD says:

    And bless you too, Willard 😉

  199. Eli Rabett says:

    Myles Allen was using a uniform prior. The problem was not that the prior was uniform, but that it extended to a change of 20 C. James Annan pointed this out. Eventually the uniform prior to 20 C vanished. James OTOH does not like uniform priors at all because they weight the wings too heavily. Eli would make the same point about Lewis’ priors that they unweight the wings to lightly.

  200. I would defer to Ms. Rabett on matters of priors:

    [Eli] Eli’s take on all this is that starting with priors (from models/theories/other data sets) which are close to the data set under analysis will result in improved statistical estimates. The (very old language here) surprisal, the difference between the prior and the posterior, will be small and one may be able to used it to extract meaningful dynamics from under the statistical noise.

    [Ms. Rabett, looking into Eli’s eyes] Very meaningful dynamics indeed.

    [Eli, keeping his cool] However, if the prior is awful, the result may actually diverge from the underlying statistical information in the data set, so with Bayes, you have to know the answer, or a good approximation to it to make progress, or, as Gelman points out

    [Eli, using Andrew’s voice] If the prior is derived from previous work, the data set may be crap, in which case the use of the Bayesian statistics is to identify crap data.

    [Eli] So how good is Eli’s prior?

    [Ms. Rabett] And posterior, which I admire on occasion.

    http://rabett.blogspot.com/2013/02/on-priors-bayesians-and-frequentists.html

    On matters of pryors, I take Richard because he’s a ninja:

  201. David Young says:

    @TP, People can decide for themselves who understands how “science works.”

    You are good at just asserting that statistician involvement is already satisfactory. I don’t know why you would say that or what your evidence is. I pointed out a couple of examples where it was not satisfactory. You said you were unfamiliar with them. Sometimes those whose job is to justify science are a little slow to look at things from the outside.

    I have two independent statisticians, an experienced one and a young and very brilliant one who has little experience in the field we are analyzing. You will find out when the report comes out. You might want to contact them the next time you have an issue.

  202. David Young said:


    This is not hard …

    I know it’s not hard, just solving a differential equation.

    How is that coming along Young ?

    Here is a helping hand:
    http://forum.azimuthproject.org/discussion/comment/14488/#Comment_14488

    🙂

  203. DY,

    People can decide for themselves who understands how “science works.”

    Of course. People can also express an opinion about it if they wish.

    You are good at just asserting that statistician involvement is already satisfactory. I don’t know why you would say that or what your evidence is.

    I didn’t assert any such thing. I pointed out that there are already skilled statistician doing climate science. I’m not suggesting that there shouldn’t be more, or that people doing climate science wouldn’t benefit if their statistical skills were to improve. My issue is with idea of independence. What does it mean? In what way would some statistician be independent. Unbiased? Really, in what way? Never done climate science before? How would that be a good thing? And – as Willard points out – would lead to more corporatist crap surrounding grants. I think the idea that climate science would improve if some group of independent statistician were to oversee their work is naive, simplistic and simply foolish. That it could improve if more statisticians became involved in climate science, on the other hand, may well be true.

    I have two independent statisticians, an experienced one and a young and very brilliant one who has little experience in the field we are analyzing.

    In what way are they independent? It sounds like they’re part of your team. Are you trying to make my point for me?

  204. Brandon said:


    The physics view of boundary conditions is that they’re set by applying known parameters to a differential equation. The more stats-minded (like me) see “boundary” and think “observed minima/maxima”.

    You are on the right track in terms of the math. For a DiffEq, the boundary conditions are largely set by the forcing function, which tend to guide the solution. The initial conditions of the DiffEq are set e.g. by the value and first derivative for a 2nd-order equation.

    So by adjusting the parameters of the DiffEq, the forcing parameters, and tweaking the initial conditions (if necessary) one can match the model to the data. This includes matching the “observed minima/maxima” as you suggest, which correspond to peaks and valleys in the ENSO behavior.

    I wrote a comment here last night describing some results:
    https://forum.azimuthproject.org/discussion/comment/14488/#Comment_1448


    What I accept absolutely because it makes the most sense to me is that ocean/atmospheric couplings are deterministic, meaning that a stochastic model with some autocorrelation thrown in to look like the real thing explains absolutely butkus. In that sense, I think “random” means “stuff we can’t explain causally, but we can tell you what it looks like”.

    My question to you now is, how does one apply a model such as you’ve shown to an AOGCM, if at all?

    My rationale for doing any of this is that a stochastic model is useless because ENSO is a singular phenomenon. It is a standing wave dipole sitting in the equatorial Pacific. This is not some random physics happening. It is either some chaotic yet deterministic behavior (which is what many believe) or it is simply quasi-periodic and obscured by the forcing factors driving it. The latter is my view.

    Now, to your question as to how to apply this to a GCM, is to use the results of the ENSO to force the GCM climate models. In my view, that would go a long way toward unifying all the disparate climate models, as they would have an additional common forcing factor (besides the solar insolation, GHGs, etc).

  205. Willard says:

    > I pointed out a couple of examples where it was not satisfactory

    “Not satisfactory” sounds normative.

    DY’s just peddling the same stuff over and over again. If it’s not his quest toward solving one of the Millennium Prize problems, it’s Nic.

    Speaking of insatisfaction, here’s Radford Neal:

    The bets based on Jeffreys’ prior will not win as much money on average as those based on the “subjective” uniform prior.

    http://climateaudit.org/2014/04/17/radiocarbon-calibration-and-bayesian-inference/#comment-547444

    Radford’s point seems to “have been accepted even though without any admission of error,” as DY himself says. Nic seems to have “moved on”. Perhaps this indicates that we’d need independent machine learning guys to check out the independent stat guys. At the very least, they are less impressionable by Millennium Prize problem arm waving.

    Speaking of which, AT, have you noted how DY, instead of defending his preference for Nic’s stuff, redirected Vaughan’s attention in the general direction of the auditing head office?

    INTEGRITY ™ – Head to the Head Office

  206. AT said:


    WHT, said something that – IMO – illustrates why trying to interact on CA is generally pointless,

    ‘as far as climate science is concerned, physics + logic beats statistics. ‘

    I was just echoing what Vaughan said:


    What I failed to reckon with is that Tamino is a statistician, not a logician (and seemingly not equipped with much more than freshman physics plus experience with variable star data and climate data).

    Vaughan is a very clear thinker and I also think Tamino does good work. It’s just that if the noise in the measurements gets in the way, you clearly have to do some statistics, but if the signal of interest is clearly exposed, then physics+logic is the way to go. In that case, one uses statistical techniques to evaluate whether you may be over-fitting, for example. But that again relies on knowledge of the physical factors too.

    You can also tell whether a statistician or a physicist is working on a noise model. For a statistician, they will call red noise an autogressive AR model, for example. In contrast, a physicist may refer to red noise as an Ornstein-Uhlenbeck reversion-to-the-mean random walk model. The latter is based on an actual physical model, whereas the AR process is simply a mathematical heuristic to a statistician, devoid of any physical meaning.

    I agree that CA is kind of a navel-gazing forum, as is the Blackboard. A lot of pointless activity IMO, but good for the occasional #OwnGoal.

  207. John Hartz says:

    The following is a very eloquent illustration of many discussions about manmade climate change. It should should be framed.

    (CNN)The world’s glaciers are disappearing.

    Climate change skeptics are not.

    The people who study our oceans and skies worry about both of these phenomena.

    But in wide swaths of 21st century America, the argument is still stuck on what to call the other side.

    “Denier” can also describe those who reject the Holocaust, so some who refuse to believe that human behavior is warming the planet prefer the label “skeptic.”

    But scientists are de facto skeptics, and many resent that word being hijacked by people who sneer at their work and accuse them of fraud.

    They suggest “denialists.”

    The other side counters with “fearmongering warmists.”

    Meanwhile, the glaciers don’t care.

    They just melt.

    Goodbye, glaciers?, Bill Weir, The Wonder List, CNN, Apr 9, 2015

  208. Kevin ONeill says:

    WHUT says: :Now, to your question as to how to apply this to a GCM, is to use the results of the ENSO to force the GCM climate models.

    I don’t think that’s how CSALT can be used. The GCMs are already performing the underlying physics of the CALT terms. And the SOI (‘S’ in CSALT) is an index that is the result of a physical process. I doubt there is anyway to force the GCMs with this. Instead, an examination of why the SOI is not accurately reproduced in GCMs would have to be undertaken and made more accurate. The literature is not filled with papers delving into the details of what factors are most influential on the GCM’s SOI output.

    It might also be instructive to look at those model runs (if any) that do correctly model SOI and build an ensemble from them to see if they also model ENSO correctly and to see if there is any predictive capability.

  209. Frank Chmod says:

    Andy’s response reminded me of Geoffrey Howe’ s resignation speech: polite, well argued and … devastating!

  210. KO, OK, I get what you are saying. My simplified model of ENSO that seems to capture the dynamics more than adequately could be expressed in terms of a GCM. In fact the wave equation that I am using to model ENSO is a direct rip-off of the wave equation that hydrologists use to approximate the sloshing of liquids in a tank (see papers by Frandsen or Faltinsen). They in fact check that their simplified “toy” model matches the results of their fluid dynamics numerical simulations (the equivalent of a GCM) , which is why they place confidence in the simple equation — because it is a useful model of sloshing that doesn’t require the full Navier-Stokes treatment.

    Calling David Young and Nick Stokes, hello, knock, knock, anybody in there?

    Yet the larger question is why does the ENSO signal have such a large impact on the rest of the global temperature signal? I would suggest that it is a forcing function within the coupled climate network. So that the periods that show up in the SOI or NINO3.4 measures also show up partially in the NAO and AMO. That is how it acts as a forcing function, to second-order in the climate network.

    BTW, the CSALT model is separate from the SOI model. I call the latter the SOIM, and it can act as an input to CSALT for when the SOI data is not available (as in the future). In fact that is why I started trying to model ENSO, because people claimed CSALT was useless as a predictive unless the SOI could also be predicted. So, voila, a twofer.

  211. Pingback: Andy Lacis responds to Steve Koonin

  212. robhon says:

    Wow. It’s fascinating to watch Don Montford going off the rails after Andy’s comment on Judith’s site.

  213. robhon says:

    Correction. I had thought that was Montford who wrote the HS Illusion. It’s Don Monfort. Someone completely different.

  214. robhon,
    Yes, it is someone different. That wasn’t Don Monford going off the rails.

  215. BBD says:

    Andrew Montford.

  216. Brandon Gates says:

    I’ve had “Don Don” go ballistic on me before. He can be singularly unpleasant when he has no argument to stand on but a position he wishes to defend.

  217. Brandon Gates says:

    Web,

    I have had similar thoughts, and questions, about how to “synchronize” AOGCM hindcast runs as you suggest, albeit not with such detailed appeals to specific maths and physics — as well as the possibility of using such methods in projections. Yours are the best answers I’ve seen in line with my own thinking, however it’s still an outstanding question in my mind whether it’s a viable solution overall. I don’t have the expertise to prosecute an argument either way, but you’ve given me some help in being better able to follow the argument, which I appreciate. Thanks.

  218. Kevin ONeill says:

    WHUT writes: “Yet the larger question is why does the ENSO signal have such a large impact on the rest of the global temperature signal?

    I’m not sure what you mean. If, as you’ve shown, ENSO can be described by CSALT, then the only things obviously missing that would have an effect on temperature are albedo and clouds. The sun, CO2, aerosols, and LOD would seem to cover most of the temperature signal by themselves. Since ENSO consists of all these effects, it would be surprising if it didn’t have such a large impact on the global temperature signal.

  219. Brandon Gates says:

    JH,

    I am fond of telling our friends across the divide that the planet doesn’t give a whiff about our politics. I am finding that the less I care what they say about my person, the better such politicking is received.

  220. KO, ENSO is not described by CSALT. ENSO is described by something I call the Southern Index Model.

    The top panel below is the Model against the SOI data for the last 130+ years. The middle panel is how well it categorizes El Nino vs La Nina excursions. The bottom two panels are wavelet scalograms which exposes how well the model works to reproduce all the important time scales.

    The reason that ENSO has such a big impact on the global temperature signal is because it is a singular process, acting as a kind of a power plant that can disperse heat to or absorb heat from the rest of the world.

    The ENSO signal along with other factors such as CO2 and volcanoes are inputs to the CSALT model and then CSALT does a multiple linear regression to determine the contributions to the trend and natural variability of the climate signal. CSALT is essentially what Vaughan Pratt is also describing elsewhere in this thread.

  221. dcpetterson says:

    I’m arriving a couple of days late to this discussion. I apologize.

    I want to comment on the topic of plausible deniability, because I have seen this technique used in countless other circumstances, and I wish to challenge it.

    One has to wonder what the intent of Koonin’s piece was, if not what is implied. That is, he claims the effects of human influences are “small” and therefore hard to certify. Some of Koonin’s defenders seem to be saying that reading anything further into Koonin’s statements is unjustified, and further implications exist only in the prurient fantasies of Koonin’s detractors. Are readers unjustified in drawing any conclusions from his statements? is his opinion piece to exist in a vacuum, with no conceivable implications? Indeed, was he not expressing an opinion in his opinion piece, but merely listing a bunch of facts as one might in a game of trivia? There seems to be something wrong with that argument; the WSJ is not a forum know for providing intentionally useless factoids meant to amuse and nothing more.

    Since he wrote for WSJ (and not, for instance, for a technical journal of some sort), his audience is not limited to scientists who might take his statements as a challenge to find new ways of making measurements of human effects more precise and reliable (assuming the truth of his statement that such measurements are difficult and therefore potentially meaningless). Instead, his audience (intended or not) consists largely of political and economic wonks (that is the stereotype of WSJ readers). It is not unreasonable to expect political and economic implications to be taken from an opinion piece published in WSJ. Given that Koonin says the human impact on climate is “small”, and advises us to compare this to a list of various other large and vital economic factors, one can reasonably infer he sees the impact of humans on the climate as being negligible in comparison.

    That inference is not, however, what some of Koonin’s defenders say is actually contained in Koonin’s opinion piece. So, for Koonin’s defenders who claim that he was not intending to minimize and dismiss human impact on climate, I ask you — what actually was the opinion he meant to express in this opinion piece? Or if you don’t want to delve into his motives — what is the conclusion that YOU draw from the piece? Would you say the facts and logic Koonin supplies could more easily be used as a possible source of support for — or against — immediate and massive action to curb CO2 emissions?

  222. John Hartz says:

    FYI — ATTP’s OP has been reposted on the Skeptical Science website. You may want to particpate in the discussion over there as well as here.

  223. oneuniverse says:

    (Sorry for the delayed response)

    ATTP, thank you. Hence my query of Dr. Lacis’ statement that “[..] the ocean cannot cause a decadal warming spurt – the deep ocean is colder than the surface biosphere, so it cannot be a source of heat.”. It seems to be at odds with literature and with what I thought was the general understanding of the ocean as a variable heat sink. Shifts in the ocean currents and winds can alter the amount of heat sequestered in the ocean, and these shifts can be multi-decadal. (I suppose a lot depends on how one defines “warming spurt” – what it’s relative to, and how that’s calculated).

    BBD:

    H&S12 is generous with its assessment of uncertainty and ends up with a lower bound of ~2C for ECS.

    I don’t think the aerosol/cloud albedo assumption underlying their ECS range is plausible or supported by evidence, as I mentioned, and I note that AR5 didn’t include their estimate. (Incidentally, their 3 +/- 1 K estimate is for 1-sigma uncertainty, the 2-sigma includes some values below 2K). WG1 Chapter 5 Table 5.3 lists a number of studies with their 5-95% range for ECS. The lower bounds of the 5-95% CI’s are 1.0, 1.2, 1.2, 1.4, 1.4, 1.6 and 2.0. Similarly, the “Making sense of palaeoclimate sensitivity” article in Nature (PALEOSENS 2012) examined a large number of paleo CS studies (including H&S12), and concluded that the 5-95% ECS range is 1.1 – 7.0 (2.2 – 4.8 K for the 68% range).

    Earlier in the aerosol forcing thread, you wrote to Ron Graf:

    That’s good because you *still* haven’t explained to us how you square 1.5C with palaeoclimate behaviour which is incompatible with an ECS below 2C and strongly suggestive of a value closer to 3C.

    I don’t think that’s in agreement with the consensus view, which seems to be that ECS values below 2 K are thought to be unlikely, but not very unlikely (following the IPCC language). It may be quibblesome, but there’s a large semantic gap between unlikely and incompatible. The latter suggests the <2K results are invalid (or that the paleo estimates are invalid).

  224. Eli Rabett says:

    dc, Eli has always called the stuff like Koonin’s implausible deniability, because to believe that they have left the dog whistle at home is to call them stupid, and that is insulting.

  225. BBD says:

    oneuniverse

    Funnily enough I was going to mention Rohling et al. to you if we continued. Okay, I agree, ‘incompatible’ was too strong. However, as Rohling et al. shows, ECS is unlikely to be below 2C so there is really no policy wriggle room. The central estimates from palaeoclimate data point to ECS being close to 3C and so that is what I regard as being the most likely approximate value. I see no merit at all in picking an unlikely value from the bottom of the range and arguing it to be correct – and that is what lukewarmers tend to do.

  226. Eli,
    Yes, I have the same problem. When someone, who should know better, writes an article like that written by Koonin, it’s hard not to see it as intentionally misleading because the alternative is that they’re actually just stupid, and that seems insulting and also rather implausible.

  227. dcpetterson says:

    Eli and ATTP,

    Precisely. Which is why the comments early in the thread arguing that Koonin didn’t mean what he whistled moved me to comment.

    I need a general technique for handling the purveyors and defenders of implausible deniability, other than being suckered into arguing obscure semantic or tangential points at the sacrifice of the central issue.

    By the way, I’m new to this blog, and I’m enjoying it immensely.

  228. dcpetterson,
    Welcome, and thanks.

  229. Eli Rabett says:

    Which is why implausible deniability works. Or you could always link to thie

  230. oneuniverse says:

    BBD, fair enough, although I’m not sure why you feel that policy is therefore determined. (Which specific policy, btw) . Any consensus on the timescale of >2C ECS manifesting, and particularly on the consequences for humanity et al is much looser than that for CS estimates (IMO).

    As for arguing for lower-range (or upper range) values of CS, how much one should be swayed would ideally depend on the strength of the argument and the quality of the evidence. The studies with estimates of ECS on the low side of the old canonical 3C (eg. Aldrin et al 2012, Lewis 2013, Otto et al 2013, Johansen et al 2015) are, I believe, more than just motivated arguments for a low value – it’s just a natural consequence of the definition of CS and the recent low trend in surface temperatures. I suppose the next 20 years or so will give us a much better idea of whether TCS is on the low or high side.

  231. John Hartz says:

    BBD & oneuniverse: Directly related to your ongoing discussion…

    Economic Collapse Will Limit Climate Change, Predicts Climate Scientist by Till Buckner, Huffington Post, Apr 14, 2015

  232. dcpetterson says:

    Eli, excellent cartoon. Thank you.

  233. BBD says:

    oneuniverse

    BBD, fair enough, although I’m not sure why you feel that policy is therefore determined. (Which specific policy, btw)

    It’s that I don’t feel that there is any case for adaption-only or BAU given the physics. Lukewarmers have a habit of arguing for both. That was the point I was trying to make. I wasn’t being very clear, sorry.

    As for arguing for lower-range (or upper range) values of CS, how much one should be swayed would ideally depend on the strength of the argument and the quality of the evidence.

    Of course. And there are valid arguments about the tendency of ‘observational’ estimates to be on the low side which is why I treat them with caution. Transient variability in the rate of ocean heat uptake wouldn’t even effect TCR (as formally defined) never mind ECS, but it *does* affect these ‘observational’ estimates. And then there’s the guesswork about aerosol forcing over the C20th… EBM limitations and assumptions etc. So rather low confidence in NL, Aldrin etc.

  234. JCH says:

    Nic Lewis is apparently using the AMO to determine CS. What I love about the AMO is there are so many examples of where it’s obviously driving the direction of the SAT/GMT, like this recent one: the kickoff for the stadium wave.

  235. Willard says:

    From that old thread at Eli’s:

    But instead of telling Mom all this complicated stuff, just take two sheets of glass outside on a sunny day and hold one over a white sheet of cardboard and the other over a black one, with the sunlight streaming through the glass to warm up the cardboard, with an air gap so no one can complain the cardboard is heating the glass by conduction. Wait five minutes and then feel the glass sheets, holding one in each hand for a few seconds and then switching hands to feel the effect better. Explain to Mom that the reason the glass over the black sheet feels warmer is that the sunlight did not heat either sheet appreciably because they’re transparent to visible light, but heated the black sheet more than the white one and the radiation from the former then heated the glass because it’s opaque to the infrared light emitted by the warmer black sheet. Also point out that the side facing the black sheet is warmer than the side facing the sun and tell her that “lapse rate” refers to the temperature gradient across the glass.

    The effect can be strengthened by waiting longer, and by insulating the setup in various ways: thicker glass, perspex (aka acrylic, plexiglas. lucite), or a saran-wrap blanket wrapped loosely but completely once around each of the four objects. The saran wrap works by trapping a thin blanket of insulating air that reduces convective cooling of the objects while still allowing both visible and infrared radiation to pass, being so thin that its main optical impact is not absorption but reflection. It has the additional benefit of further reducing any chance of convective warming of the glass by the cardboard, the point being to demonstrate the effect of radiation in the absence of convection.

    Remarkably, many climate scientists deny that the glass warms up for the same reason the atmosphere does, citing physicist R.W. Wood’s little note in the February 1909 issue of Philosophical Magazine purportedly showing that radiation plays a negligible role by comparison with convection. I thumbed through a few subsequent issues of Phil. Mag. expecting someone to raise the obvious objection (since de Saussure had demonstrated strong warming by this effect in 1767, see http://www.solarcooking.org/saussure.htm ), and bingo: there in the July 1909 issue was a much longer paper by Charles Greeley Abbot, director of the Smithsonian Astronomical Observatory, thoroughly refuting Wood. See http://boole.stanford.edu/wood for more details about that as well as a second failed attempt by Wood a few months later to debunk an unrelated optical effect. In that regard Wood was the Will Happer of his day, though he made up for it with much excellent work in other aspects of optics, inventing Wood’s glass, debunking N-rays, etc.

    At 0.39‰ (0.039%) by volume, atmospheric CO2 seems a negligible component of the atmosphere. One way for Mom to better appreciate it is to point out to her that at a level of 1‰ (0.1%), atmospheric CO2 in its solid form, dry ice, would coat the Earth with a layer 1 cm thick, a handy coincidence. The current level of 0.39‰ would therefore be 0.39 cm thick. Since that’s close to the thickness of ordinary glass, and since CO2, glass, and perspex block IR by the same mechanism, namely absorption by triatomic molecules, if not at the exact same wavelengths, the analogy is an excellent one even quantitatively!

    http://rabett.blogspot.com/2010/03/simplest-explanation.html?showComment=1269282501514#c7028382147650688144

    Some things should not stay at Eli’s.

  236. John Hartz says:

    Providing context for the OP and much of the discussion that has already taken place on this thread…

    What a group of physicists think about climate change matters greatly because climate science is, after all, a branch of physics, and most atmospheric scientists are based in physics departments. And the APS’s route to its 2015 statement has been dogged by controversy and its own internal politics, including a coup d’état that led to the resignation of Steve Koonin, the head of the climate statement review committee and a former Obama administration official.

    The nadir of all this was reached when Koonin wrote an editorial in The Wall Street Journal that appeared to question climate science.

    The process of putting the statement together was “painful,” said Robert Rosner, an astrophysicist at the University of Chicago and an APS member involved in the drafting process until December 2014.

    That a group of seemingly staid physicists could come to metaphorical blows over a previously accepted climate statement reflects just how politically charged the issue remains.

    Physicists Battle over Meaning of Incontrovertible in Global Warming Fight by Gayathri Vaidyanathan, ClimateWire/Scientific American, Apr 14, 2015

  237. oneuniverse says:

    BBD

    Thank you for the clarification on policy. FWIW I think there are good arguments for reducing our reliance on fossil fuels.

    re: CS
    Ok, it seems that more heat was sequestered in the deeper ocean in the last decade than expected, an expectation based on earlier observations (Balmaseda ea 2013, England ea 2014, Chen and Tung 2014). The latter two papers estimate that the current increased ocean heat uptake efficiency will continue on a decadal scale, which will probably cause observationally-based estimates of TCS (and ECS) to continue to drop lower. At some point, it is expected that the efficiency will decrease.

    However, what makes you say that the increased efficiency is transient ie. will not re-occur (if I’ve understood what you mean by ‘transient’ correctly) ? None of the papers I’ve read make this assertion.

  238. snarkrates says:

    oneuniverse: “However, what makes you say that the increased efficiency is transient ie. will not re-occur (if I’ve understood what you mean by ‘transient’ correctly) ?”

    I think you are missing the point. As long as the surface (or really the tropopause) is not warming, we aren’t approaching equilibrium, because we aren’t radiating away the heat we’ve been absorbing. Thus, hiding heat in the ocean may delay the warming, but it won’t end it. In any case, a warming deep ocean is NOT a good thing for the health of the oceans.

    Also, a larger heat reservoir, unless you are proposing that it only magically kicks in for the current epoch, ultimately would raise climate sensitivity estimates when all the thermodynamics are properly accounted for.

  239. As long as the surface (or really the tropopause) is not warming, we aren’t approaching equilibrium, because we aren’t radiating away the heat we’ve been absorbing.

    Yes, this is a good point. Energy balance comes through the outgoing flux (on average) balancing the incoming flux. This depends on the surface/troposphere temperature, not on the energy in the oceans. Until the surface temperature rises to the point where we’re back in energy balance, we will remain out of balance.

  240. oneuniverse says:

    Hi John,

    Thanks for the link to the article (1:04 am). The interviewee Christopher Reyer, of the Potsdam Institute, figures we’re headed for an +8C rise by 2100 (on RCP8.5), but won’t get there because economies will collapse before that from stresses caused by climate change. Reyer is the co-author of Turn Down the Heat: Confronting the New Climate Normal, which I hadn’t come across before – thanks again.

  241. BBD says:

    oneuniverse

    Let’s take a middle-road position and say that ECS is ~3C, perhaps a bit lower (eg. GISS Model E’s ~2.7C). TCR pans out at ~1.6C but of course everything is ongoing because of the tedious physics. And there is an assumption that we stabilise at ~560ppm CO2.

    Too alarmist? Step it down to ECS in the range 2C – 2.5C.

    Looking at the Eemian for rough guidance, that could pan out as ~6m SLR over the next millennium, which is a hell of a legacy.

  242. oneuniverse says:

    Snarkrates, ATTP, that’s a good point, and a lower TCS might imply a higher ECS (although I can’t find the reference for this at the moment – Isaac Held I think).

    How the retained energy is partitioned into the atmosphere, land, and ocean layers is very important, though. More heat into the deep ocean means lower TCS, and a lower rate of rise of the surface temperature. Since one of the major concerns about AGW is the high rate of change , that’s significant. (The increased heat uptake efficiency in the deeper layers also suggests that the ocean might have a greater efficiency at sequestering CO2 as well, since the higher efficiency is due to increased mixing rather than increased conduction.).

    BBD:

    Looking at the Eemian for rough guidance, that could pan out as ~6m SLR over the next millennium, which is a hell of a legacy.

    How do you arrive at the 1000 years figure? Most of the sea level rise during the Eemian is thought to have come from Greenland. AR5 WG1 Chapter 13.4.3.3 states: “The complete loss of the [Greenland] ice sheet is not inevitable because it has a long time scale (tens of millennia near the threshold and a millennium or more for temperatures a few degrees above the threshold)”. The GMST-above-preindustrial threshold values it lists are :

    Gregory and Huybrechts (2006) [1.9 to 4.6] °C
    Gregory and Huybrechts (2006), following Church et al. (2001), [1.5 to 3.0] °C, best estimate 2.1 °C
    Rae et al. (2012) [2.1 to 3.4] °C
    Robinson et al (2012) [0.8 – 3.2] °C, best estimate 1.6 °C
    Fettweis et al. (2013) [2.1 to 4.1] °C

    In your described scenario, with a 2.5 C ECS and a CO2 doubling to 560ppm from pre-industrial, and taking the mid-way value of the threshold range as best estimate where none is listed, a GMST of 2.5C above preindustrial is below some of the best estimates of the threshold, and less than 1 C above any of them. So following the AR5 assessment, the best estimate for a Greenland ice loss and 6m sea level rise would be at least several thousand years, possibly closer to ten thousand years or more (or ‘never’, if the threshold turns out to be too high).

  243. BBD says:

    How do you arrive at the 1000 years figure?

    By a generous estimate of how long the gravity-driven drainage of the WAIS will take once fully underway.

    Most of the sea level rise during the Eemian is thought to have come from Greenland.

    No it isn’t. Current thinking is that about 2m came from the GrIS, the rest from the WAIS and perhaps some EAIS (Colville et al. 2011; Dahl-Jensen et al. 2013).

    Eemian GAT was ~2C above Holocene at most. Eemian MSL was ~6m above Holocene.

  244. oneuniverse says:

    re: Melting of the Greenland ice-sheet
    Robinson et al (2012), the study with the lowest threshold range, has a figure 3b which describes the change in relative ice volume over time for different constant regional summer temperature anomalies.

    The GISS data for summer (JJA) temperatures recorded at the Greenland stations with data up to 2015 (2013 in the case of Prins Christ), namely Angmagssalik, Danmarkshavn, Egedesminde, GodthabNuuk, Jan Mayen, Nord Ads and Prins Christi, shows that since 1951 (the date at which stations all have data), summer temperatures have risen by about 1 C at all the stations. Over the same period, GMST has risen by about 0.5C. Assuming the ratio continues, a 2.5C GMST rise will correspond to a 5C Greenland summer temperature rise. According to Robinson’s fig.3b, that corresponds to about 5000 years for a total melt. After 1000 years, the model predicts that over 80% of the ice sheet volume will still be present.

  245. BBD says:

    Try responding to what is written.

    You remind me *very* much of the Imaginary Number Guy (-1).

  246. oneuniverse says:

    Hi BBD,

    Would you mind providing a reference on the gravity-driven drainage of the WAIS? The AR5 estimates (fig.13.13) that the net contribution of the Antarctic to SLR will be negative over the next 600 years for a medium 500-700ppm CO2 scenario, the main contributor to the rise being thermal expansion and melting of Greenland ice. The total SLR for 2500 for that scenario is projected to be between 0.1m and 2.15m. By these estimates, a 6m rise by 1000 years is at the extreme end of projections.

    Levermann ea. 2013 estimate 2.3m/C rise after 2000 years (per C above preindustrial) which would correspond to 5.8m rise after 2000 years under the 2.5C ECS scenario.

  247. oneuniverse says:

    Try responding to what is written.

    I wrote the comment on Greenland summer temperatures and Robinson ea before I read your at 1:57 pm, and thought I’d submit it as I’d done a bit of work checking the temperatures and thought it might be of interest (it was to me).

    By the way, still curious to hear why you think that the increased ocean heat update efficiency is transient – references to the literature if possible, please.

  248. BBD says:

    The AR5 estimates (fig.13.13) that the net contribution of the Antarctic to SLR will be negative over the next 600 years

    ?

  249. BBD says:

    Perhaps you would now acknowledge that this was bollocks?

    Most of the sea level rise during the Eemian is thought to have come from Greenland.

  250. Andrew Dodds says:

    oneuniverse –

    Obvious question.. does this work incorporate the connection of the ocean to the interior of the GIS via the Petermann glacier (and possibly elsewhere). I suspect that there could be some unmodelled physical processes going on here.

    And from what I understand the Paleo evidence is for faster rather than slower melt.

    But we’ll just have to run the experiment and see.

  251. BBD says:

    By the way, still curious to hear why you think that the increased ocean heat update efficiency is transient – references to the literature if possible, please.

    I can’t find any references that say it is not transient. Can you?

  252. BBD says:

    Would you mind providing a reference on the gravity-driven drainage of the WAIS?

    I had this in the back of my mind Joughin et al. (2014):

    Resting atop a deep marine basin, the West Antarctic Ice Sheet has long been considered prone to instability. Using a numerical model, we investigated the sensitivity of Thwaites Glacier to ocean melt and whether its unstable retreat is already under way. Our model reproduces observed losses when forced with ocean melt comparable to estimates. Simulated losses are moderate (1 mm per year of sea-level rise) collapse in the different simulations within the range of 200 to 900 years.

  253. oneuniverse says:

    I wasn’t aware that recent studies show that the Antarctic is thought to have contributed more to SLR during the Eemian. The previous understanding was that Greenland was the main ice-source SLR contributor. (The Wikipedia article on the Eemian also refers to Greenland as the main cause of SLR. This is not an admission that I would look anything up in a slipshod manner on Wikipedia 🙂 ). Thanks for the links to the new papers.

    BBD:

    Current thinking is that about 2m came from the GrIS, the rest from the WAIS and perhaps some EAIS (Colville et al. 2011; Dahl-Jensen et al. 2013).

    Actually, that’s “bollocks” too, since you omitted thermal expansion.

  254. BBD says:

    Actually, that’s “bollocks” too, since you omitted thermal expansion.

    And you you are being silly. But sure, add in some thermal expansion and be happy.

  255. oneuniverse says:

    re: Joughin ea 2014
    Thanks – the study is just for the Thwaites glacier though, which is an area of the WAIS losing ice at one of the the highest rates. Thwaites is less than 20% of the WAIS (possibly much less, that’s just a very rough estimate by eye). The rest of the WAIS isn’t estimated to lose ice at the same pace, although Thwaites’ collapse is thought to contribute to the rest’s instability.

    re: thermal expansion
    It’s one of the significant contributors – I missed it out, as did you. I don’t think it’s silly to correct the mistake, you may feel differently.

    BBD: “I can’t find any references that say it is not transient. Can you?”

    Yes, eg. Kosaka & Xie 2013 : “We conclude that the recent cooling of the tropical Pacific and hence the current hiatus are probably due to natural internal variability rather than a forced response. If so, the hiatus is temporary, and global warming will return when the tropical Pacific swings back to a warm state. Similar hiatus events may occur in the future and are difficult to predict several years in advance owing to the limited predictability of tropical Pacific SST.”

    Chen and Tung 2014 (linked upthread) propose that a salinity mechanism can explain changes in ocean heat uptake. (“A mechanism that can account for the speed with which heat penetrates to such great depths is deep convection caused by vertical density differences”). They also note that “[..] over the longer period reported here, there was very little trend in the salinity, only a vacillating cycle.”. Their figure 6 shows the correlation between salinity and OHC at depths. It’s clear from that figure that a similar (but weaker) high heat uptake event around 1965-1975. If it’s happened twice in the last 50 years, by definition it’s not a one-off (“transient”), and it’s highly plausible that it’s going to happen again.

  256. oneuniverse says:

    Andrew, I don’t know – there’s no mention of the Petermann glacier in the paper. They do use the SICOPOLIS software to model the evolution of the ice, which seems to take the bedrock topography into account.

  257. BBD says:

    Transient is fully compatible with ‘natural variability’. Variability is almost synonymous with ‘transient’. That such events recur occasionally is fully compatible with them being transient episodes of natural variability. Nothing here need have a significant effect on TCR and certainly not ECS. And I can’t find anybody who is arguing that natural variability in the rate of ocean heat uptake is anything other than transient. It’s odd that you are pushing this position.

    * * *

    Thanks – the study is just for the Thwaites glacier though, which is an area of the WAIS losing ice at one of the the highest rates.

    It’s not just Thwaites. See Rignot et al. 2014 Widespread, rapid grounding line retreat of Pine Island, Thwaites, Smith, and Kohler glaciers, West Antarctica, from 1992 to 2011:

    We measure the grounding line retreat of glaciers draining the Amundsen Sea sector of West Antarctica using Earth Remote Sensing (ERS-1/2) satellite radar interferometry from 1992 to 2011. Pine Island Glacier retreated 31 km at its center, with most retreat in 2005–2009 when the glacier ungrounded from its ice plain. Thwaites Glacier retreated 14 km along its fast flow core and 1 to 9 km along the sides. Haynes Glacier retreated 10 km along its flanks. Smith/Kohler glaciers retreated the most, 35 km along its ice plain, and its ice shelf pinning points are vanishing. These rapid retreats proceed along regions of retrograde bed elevation mapped at a high spatial resolution using a mass conservation technique that removes residual ambiguities from prior mappings. Upstream of the 2011 grounding line positions, we find no major bed obstacle that would prevent the glaciers from further retreat and draw down the entire basin.

    Bear in mind the GrIS is melting all the while that the WAIS is destabilising then consider the implications for the rate of millennial sea level rise of Ice plug prevents irreversible discharge from East Antarctica:

    Changes in ice discharge from Antarctica constitute the largest uncertainty in future sea-level projections, mainly because of the unknown response of its marine basins1. Most of West Antarctica’s marine ice sheet lies on an inland-sloping bed2 and is thereby prone to a marine ice sheet instability3, 4, 5. A similar topographic configuration is found in large parts of East Antarctica, which holds marine ice equivalent to 19 m of global sea-level rise6, that is, more than five times that of West Antarctica. Within East Antarctica, the Wilkes Basin holds the largest volume of marine ice that is fully connected by subglacial troughs. This ice body was significantly reduced during the Pliocene epoch7. Strong melting underneath adjacent ice shelves with similar bathymetry8 indicates the ice sheet’s sensitivity to climatic perturbations. The stability of the Wilkes marine ice sheet has not been the subject of any comprehensive assessment of future sea level. Using recently improved topographic data6 in combination with ice-dynamic simulations, we show here that the removal of a specific coastal ice volume equivalent to less than 80 mm of global sea-level rise at the margin of the Wilkes Basin destabilizes the regional ice flow and leads to a self-sustained discharge of the entire basin and a global sea-level rise of 3–4 m. Our results are robust with respect to variation in ice parameters, forcing details and model resolution as well as increased surface mass balance, indicating that East Antarctica may become a large contributor to future sea-level rise on timescales beyond a century.

    Six metres in the next thousand years really isn’t at all implausible. So why all the pushback? It’s actually rather odd.

    You do remind me ever so much of ‘-1’ the Imaginary Numbers Guy.

  258. BBD says:

    Bougamont et al. (2014) Sensitive response of the Greenland Ice Sheet to surface melt drainage over a soft bed.

  259. BBD says:

    Missed link:

    Mengel & Levermann (2014) Ice plug prevents irreversible discharge from East Antarctica

  260. oneuniverse:


    Yes, eg. Kosaka & Xie 2013 : “We conclude that the recent cooling of the tropical Pacific and hence the current hiatus are probably due to natural internal variability rather than a forced response. If so, the hiatus is temporary, and global warming will return when the tropical Pacific swings back to a warm state. Similar hiatus events may occur in the future and are difficult to predict several years in advance owing to the limited predictability of tropical Pacific SST.”

    One of the primary contributions to internal variability is ENSO. We are working this problem at the Azimuth Project, and are getting close to cracking the code.

    What we have here is the Imitation Game being played out. The skeppies are trying to convince everyone to make the wrong decision while the realists are here to aid in making the right move.

    The only problem is that this is not a game.

  261. BBD says:

    re: thermal expansion
    It’s one of the significant contributors – I missed it out, as did you. I don’t think it’s silly to correct the mistake, you may feel differently.

    Looks like another of your mistakes, not mine.

    McKay et al. (2011) The role of ocean thermal expansion in Last Interglacial sea level rise

    A compilation of paleoceanographic data and a coupled atmosphere-ocean climate model were used to examine global ocean surface temperatures of the Last Interglacial (LIG) period, and to produce the first quantitative estimate of the role that ocean thermal expansion likely played in driving sea level rise above present day during the LIG. Our analysis of the paleoclimatic data suggests a peak LIG global sea surface temperature (SST) warming of 0.7 ± 0.6°C compared to the late Holocene. Our LIG climate model simulation suggests a slight cooling of global average SST relative to preindustrial conditions (ΔSST = −0.4°C), with a reduction in atmospheric water vapor in the Southern Hemisphere driven by a northward shift of the Intertropical Convergence Zone, and substantially reduced seasonality in the Southern Hemisphere. Taken together, the model and paleoceanographic data imply a minimal contribution of ocean thermal expansion to LIG sea level rise above present day. Uncertainty remains, but it seems unlikely that thermosteric sea level rise exceeded 0.4 ± 0.3 m during the LIG. This constraint, along with estimates of the sea level contributions from the Greenland Ice Sheet, glaciers and ice caps, implies that 4.1 to 5.8 m of sea level rise during the Last Interglacial period was derived from the Antarctic Ice Sheet. These results reemphasize the concern that both the Antarctic and Greenland Ice Sheets may be more sensitive to temperature than widely thought.

  262. oneuniverse says:

    re: thermal expansion
    McKay ea estimate Eemian peak temperatures as 0.7 +/- 0.6 C compared to late Holocene. This is on the low side compared to other estimates, which are closer to 1-2C I think. Even with this low value, they calculate thermal expansion as contributing 0.4 +/- 0.3 m, which I still consider significant, so I stand by my statement. Maybe we have a different understanding of “significant”.

    BBD: “Six metres in the next thousand years really isn’t at all implausible.”
    For CO2 concentrations of 560ppm and an ECS of 2.0-2.5C, which is the scenario you had proposed and were describing, I think that is considered very unlikely. 560ppm corresponds roughly to RCP4.5. S. Jevrejeva ea 2011 fig. 4a shows RCP4.5 SLR projection for 2500 at under 2m, and having levelled off.

    re: ocean heat uptake
    BBD: “That such events recur occasionally is fully compatible with them being transient episodes of natural variability. Nothing here need have a significant effect on TCR and certainly not ECS.”
    I’m glad we’ve established that you agree that they will likely re-occur (I just wanted to make sure, as I made said in my original question, that by transient you didn’t mean “one-off”). Do you agree that an increased average ocean heat uptake efficiency will mean a lower TCR? If so, since you seem to say (your wording isn’t clear) that TCR is unlikely to be affected, I take it then that you believe that an increase of the average efficiency is unlikely?

  263. BBD says:

    Do you agree that an increased average ocean heat uptake efficiency will mean a lower TCR?

    Where does the ‘increased average’ stuff come from?

    At the outset I made it clear that I was talking about TCR as formally defined – the point when CO2 reaches double the pre-industrial average:

    Transient variability in the rate of ocean heat uptake wouldn’t even effect TCR (as formally defined) never mind ECS, but it *does* affect these ‘observational’ estimates.

    Since this doubling is unlikely to occur until the second half of this century, I think it is reasonable to argue that the effects of transient natural variability in OHU will have little net impact. Of course doubling of CO2 could occur during a decade or two when natural variability suppresses OHU and increases the rate of surface warming. This might create the misleading impression that TCR had been under-estimated.

  264. BBD says:

    For CO2 concentrations of 560ppm and an ECS of 2.0-2.5C, which is the scenario you had proposed and were describing, I think that is considered very unlikely.

    Jolly good.

  265. BBD says:

    Attentive readers will have noticed that the results of Jevrejeva et al. (2011) are rendered conservative and obsolete by more recent work (see above).

    The really smart among them will recognise that every year, confidence in low, slow centennial to millennial scale SLR diminishes as research into the WAIS and EAIS progresses.

    With this in mind, the super-mega smart (and they know who they think they are) will return to the original problem and think again. If Eemian GAT was only ~2C warmer than the Holocene but MSL was ~6m higher, what does this tell us about marine ice sheet stability under even relatively moderate warming?

  266. Andrew Dodds says:

    @BBD

    Not sure that Eemian temperatures were that much higher on a global average basis, it seems to be more of a distribution effect with greater insolation towards the north pole.

    Whereas now we have extra heat coming in everywhere plus Arctic amplification. Oops. I think you are being over optimistic..

    Interestingly, sea level is already going at 3m/1000 years.

  267. BBD says:

    Andrew Dodds

    I’m not sure the Eemian peak GAT was as much as 2C above the Holocene either (~1C would be my preferred estimate) but I’m being generous with uncertainty.

    Interestingly, sea level is already going at 3m/1000 years.

    I wish I’d said that 😉

  268. oneuniverse says:

    BBD:

    Attentive readers will have noticed that the results of Jevrejeva et al. (2011) are rendered conservative and obsolete by more recent work (see above).

    Recent work also finds that the Greenland ice sheet may have been more stable than previously thought during the Eemian interglacial. Sensible readers who are not experts in the field will, I suspect, on the whole, wait and see how the experts revise their multi-centennial and multi-millenial projections on the basis of new work. I referred to AR5 as that’s the most comprehensive recent review that I’m aware of. For multi-millenial SLR, the AR5 states (chapter 13) : “Taken together, these results imply that a sea level rise of 1 to 3 m °C–1 is expected if the warming is sustained for several millennia (low confidence) (Figure 13.14e, 13.14j).”. As I understand it, this puts your 6m SLR in 1000 years for an ECS 2.0-2.5C and 560ppm CO2 at the extreme end of projections, similarly for the more recent post-AR5 SLR estimate of Levermann ea 2013 which I cited earlier. You still haven’t referenced any SLR projections that support your statement.

    If Eemian GAT was only ~2C warmer than the Holocene but MSL was ~6m higher, what does this tell us about marine ice sheet stability under even relatively moderate warming?

    I don’t feel I’m going out on a limb by suggesting that the IPCC AR5 authors and scientists on whose work the report was based were aware of this (and of recent SLR rates) and gave it due consideration in their estimates of future SLR. The text certainly suggests that they did.

  269. BBD says:

    AR5 is ‘optimistic and overconfident’ according to an expert in the field because it does not sufficiently account for ice sheet dynamics. Just as I have been trying to explain to you for rather a long time now.

    The problem here is that you are ill-versed in this topic (witness the nonsense about GrIS driving Eemian SLR; the size of thermal expansion component in Eemian MSL highstand etc) but won’t stop talking, won’t listen and won’t learn.

    So this is a waste of my time.

  270. BBD says:

    Recent work also finds that the Greenland ice sheet may have been more stable than previously thought during the Eemian interglacial.

    Recent work which I referenced for you above. Recent work which demonstrates that the WAIS and sectors of the EAIS are more unstable than hitherto recognised and contributed much *more* than previously thought to the Eemian MSL highstand.

    But in your twisted version of this exchange, this gets to be support for your contrarianism. Either you don’t understand what has been explained or you are playing tedious games.

  271. John Hartz says:

    Breaking news relevant to the ongoing discussion on this thread…

    Although Arctic sea ice set a record this year for its lowest ever winter extent – that was not the case for its volume, new data reveals.

    ‘3D Cryosat’ tracks Arctic winter sea ice by Jonathan Amos, Science & Environment, BBC News, Apr 17, 2015

  272. oneuniverse says:

    BBD: “But in your twisted version of this exchange, this gets to be support for your contrarianism. Either you don’t understand what has been explained or you are playing tedious games.”

    I leave others to read the above exchange and judge whether either of these is the case.

  273. BBD says:

    oneuniverse

    Fair enough.

  274. WHT: You can also tell whether a statistician or a physicist is working on a noise model. For a statistician, they will call red noise an autogressive AR model, for example. In contrast, a physicist may refer to red noise as an Ornstein-Uhlenbeck reversion-to-the-mean random walk model.

    Even simpler test: statisticians model noise discretely (time t = 0,1,2,…) whereas physicists model it continuously (time stepping infinitesimally).

  275. @BBD: AR5 is ‘optimistic and overconfident’ according to an expert in the field because it does not sufficiently account for ice sheet dynamics.

    Quite right. Projections of sea level are frequently based on linear fits to the 20th century, during which the level rose at an average rate of 1.5 mm a year, as can be seen from this plot from Chapter 5 of AR4:

    By now people know not to do this for projections of temperature based in this way, since that would forecast a rise of only 0.63 °C/decade when it was actually 1.5 °C/decade over the last third of the century (1966.7 to 2000), and likely to be even steeper for 2000-2033. (It is wishful thinking to expect the pause to continue past 2010.)

    Odd that the lessons from temperature are transferring to sea level so slowly. Close examination of the above figure shows that it is not rising linearly but following a curve trending more steeply lately, with a rise of 2.5 mm/year over the last third of the century.

    However merely looking at sea level as following a curve does not take into account the more recent phenomenon of rivers developing under glaciers, which tend to yield abrupt sliding of glaciers into the sea when sufficient melting has allowed them to break loose and slide freely. Hence even recent trends need not be a reliable indicator of future rates of rise in sea level, which by such means could easily increase to 10-20 mm/yr.

  276. Pingback: Steve Koonin in the New York Times | …and Then There's Physics

  277. PW Anderson wanted to know more about Koonin, and was very impressed with this! Thanks.

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