by Judith Curry
A paper in press in the Journal of Climate provides some insight into the interaction of cloud feedback with ocean heat transport.
Climate sensitivity to changes in ocean heat transport
Marcelo Barreiro and Simona Masina
Abstract: Using an atmospheric general circulation model coupled to a slab ocean we study the effect of ocean heat transport (OHT) on climate prescribing OHT from zero to two times the present-day values. In agreement with previous studies an increase in OHT from zero to present-day conditions warms the climate by decreasing the albedo due to reduced sea-ice extent and marine stratus cloud cover and by increasing the greenhouse effect through a moistening of the atmosphere. However, when the OHT is further increased the solution becomes highly dependent on a positive radiative feedback between tropical low clouds and sea surface temperature. We found that the strength of the low clouds-SST feedback combined with the model design may produce solutions that are globally colder than Control mainly due to an unrealistically strong equatorial cooling. Excluding those cases, results indicate that the climate warms only if the OHT increase does not exceed more than 10% of the present-day value in the case of a strong cloud-SST feedback and more than 25% when this feedback is weak. Larger OHT increases lead to a cold state where low clouds cover most of the deep tropics increasing the tropical albedo and drying the atmosphere. This suggests that the present-day climate is close to a state where the OHT maximizes its warming effect on climate and pose doubts about the possibility that greater OHT in the past may have induced significantly warmer climates than that of today.
The paper is in press in the Journal of Climate. A complete online version of the paper is found [here].
Introduction
The overall all problem is laid out in the Introduction, which provides an excellent literature review on the topic. A few excerpts:
The oceans absorb heat mainly in the tropical regions where cold water upwells to the surface and lose it in high latitudes where cold and dry winds blow over warm currents during winter time. This implies a net heat transport by the oceanic circulation from the equator to the polar regions that contributes to remove the surplus of heat received in the tropics. Averaged over long times the ocean must gain and lose equal amounts of heat in order to maintain a steady state. The oceanic heat transport is largest in the tropical region and becomes very small poleward of 45° (Trenberth and Caron 2001). At higher latitudes the heat transported by the atmosphere, due mainly to the presence of energetic eddies, is the main contributor to total poleward heat transport.
The circulation of the oceans likely changed over the course of Earth’s history, due to changes in external forcings, e.g, insolation and greenhouse gases, and changes in continental configuration. Thus, a change in ocean heat transport is a common explanation in studies of past climates. For example, Rind and Chandler (1991) propose that 46% greater ocean heat transport during the Jurassic period (200144 million years ago, Ma) would have warmed the climate by 6 K. They also suggest that 68% greater ocean heat transport during the Cretaceous (14465 Ma) would have warmed the climate by 6.5 K. Barron et al. (1993) studied the impact of oceanic heat transport in the Cretaceous using an atmospheric model coupled to a slab ocean. Imposing presentday zonally averaged heat transport but distributed differently among oceans due to a different continentaconfiguration they found that increased ocean heat transport warms the climate. Moreover, they found that the warming is not linearly related to the value of oceanic heat transport: increasing from 0 to present day heat transport increases the surface temperature by 2.6 K, but only 0.6 K from present day to two times present day values. Closer to the present and already with the same continental configuration, Dowsett et al. (1996, 2009) argue that the warmer high latitude ocean temperatures during the midPliocene (~3 Ma) can be explained by a more vigorous North Atlantic Deep Water formation and thermohaline circulation. Finally, Romanova et al. (2006) found using an atmospheric general circulation model that reduced ocean heat transport contributed to global cooling during the Last Glacial Maximum. In general, patterns of decreased equatortopole temperature gradients due to a large extratropical warming, as in the case of the Eocene, are explained as due to enhanced ocean heat transport: larger ocean heat transport decreases sea ice in high latitudes leading to an icealbedo feedback that warms these regions. The tropics may cool or stay close to present values, so that there is overall global warming. In recent years, other studies have suggested that increased ocean heat transport cannot fully explain the decrease in the meridional temperature gradient during the Eocene. Alternative explanations involving high latitude convection feedbacks have been proposed to explain the high latitude warming of past climates.
The undergoing changes in climate caused by human activities will probably affect the oceanic circulation and its heat transport, which then may feed back onto theatmosphere and climate. Nevertheless, the connection between atmospheric and oceanic heat transports is not yet well understood. For example, is it possible to change one component without changing the other one? Everything else being equal (e.g. constant greenhouse concentration), this would result in changes in the albedo of the planet because the total heat transport by the oceanatmosphere system will be different, and thus the system has to gain heat differently at each latitude.
The representation of clouds is one of the main weaknesses of current climate models . In particular, the parameterization of boundary layer stratus clouds has proved to be very difficult and has been a major area of research in the last decade. These clouds have a very weak greenhouse effect, but strongly reflect incoming shortwave radiation, thus modulating the albedo of the Earth. Bony and Dufresne (2005) have shown that the simulation of marine low level clouds is a large source of uncertainty in tropical cloud feedbacks and of climate sensitivity, suggesting that the simulation of tropical responses to different forcings will strongly depend on the parameterization of these clouds, and that results need to be tested using different cloud schemes.
Conclusion
For decreased ocean heat transport we have found very similar results as previously reported by W03 and H05: a decrease in the heat transported by the ocean cools the climate by increasing the sea ice extent and the low oceanic cloud cover, thus increasing the albedo. Moreover, the tropical regions become narrower thus decreasing the moistening of the subtropical atmosphere and thus the greenhouse trapping. These atmospheric changes are such that the atmospheric heat transport tends to compensate for the decreased OHT: there is almost complete compensation in the deep tropics while in the extratropics the total poleward transport of heat is smaller when the ocean circulation is absent. We propose that these changes are robust across models mainly because decreasing the ocean heat transport does not fundamentally alter the circulation of the presentday atmosphere, it essentially represents a small deviation from today’s conditions.
The climatic response for larger than presentday values of ocean heat transport is very different from previous studies and it is highly dependent on the parameterization of low clouds. Taking equatorial regions warmer than the subtropics as a plausibility criterion for the solution, the results are that an increase in OHT tends to warm the climate and that this warming is largest when the tropical region is widest. However, the cloud scheme dictates how much can the OHT increase before the solution becomes unphysical. A highly sensitive scheme suggests that our current climate is very close to the maximum positive effect of the ocean heat transport on climate (less than a 15% increase away); another cloud scheme suggests that the climate can further warm 0.6 K for a 25% increase in OHT. For OHT increases larger than 25% of presentday values, a strong positive radiative feedback between tropical low level clouds and sea surface temperature works, always leading to an unphysical cold climate. In this state, low level clouds tend to cover the tropics which increases the albedo enormously. At the same time, the Hadley circulation reverses, inducing subsidence over the tropics which inhibits convection and dries the atmosphere, thus cooling it further due to decreased greenhouse trapping. As a consequence the tropical atmosphere transports heat equatorward resulting in decreased total ocean+atmosphere heat transport when the OHT increases.
Thus, as long as the cloud cover parameterizations are correct, the results presented here do not support the hypothesis that larger OHT may have led in the past to warmer than presentday climates without changing the total poleward heat transport, as has been suggested in the literature. We argue that the results of Barron et al. (1993) are due to the use of an atmospheric model with simpler physical parameterizations. To test this we repeated the experiment of increasing the OHT using the International Centre for Theoretical Physics (ICTP) AGCM, an atmospheric model with an horizontal resolution of T30 and 8 vertical levels and simpler parameterizations of the physical processes. In this model cloud cover is defined diagnostically from the values of relative humidity in the air column (excluding the boundary layer) and the total precipitation, and cloud albedo is proportional to the total cloud cover. We found that this model warms 0.8 K when the OHT is increased from 0 to present day values and 0.4 K from presentday to two times presentday heat transport. The sensitivity is much smaller than that of ECHAM5, and even compared to that of the model of Barron et al. (1993). However, as in the latter case, an increase in ocean heat transport always warms the climate, and in a nonlinear way. Taken together, the results of this work suggest that the simpler the cloud cover scheme and the cloudalbedo relationship the less sensitive is the model to changes in ocean heat transport. This is mainly due to differences in the parameterization of low level clouds, and their interaction with radiative fluxes.
A caveat of our results is the lack of ocean dynamical adjustment which may act as a negative feedback opposing cloudSST feedback that leads to the large simulated tropical cooling, in a similar way as found by Hazeleger et al (2005). Note that this caveat applies not only for increased values of the OHT, but also for decreased values because all solutions involve changes in the surface winds. Other possibilities include that the schemes used in today’s models are missing important physics to represent correctly the behavior of low clouds, as has been suggested previously , and so past climates could be used as test for models.
To date our understanding of the climatic response to changed OHT comes mainly from atmospheric models coupled to fixed oceans. Our results point that not only is the lack of dynamical adjustment an important issue when using these models, but also the parametrization of low clouds that result in cloudSST radiative feedbacks of different strengths. In the end, only through the use of coupled models that allow the interaction between these processes will be possible to address this question fully. Nonetheless, we believe the results presented here can serve as a guide for future explorations of the role of the oceans in climate.
JC comment: This paper addresses the very interesting problem of cloud feedback and climate sensitivity in response to ocean heat transport, which relates to some of the ideas that Roy Spencer has been developing. The paper uses climate models in the manner in which they are probably most useful: to conduct experiments using different forcing data and model parameterizations to increase understanding of both how the climate system works and the limitations of climate models.
Rulers at home
scientists befuddled
the tracks no longer look straight
how was that Kim?
Government-funded space and climate scientists have been befuddled since agreeing to go along with the “Bilderberg solar model” [1] of Earth’s heat source as “homogeneous, and in hydrostatic equilibrium.” It is not!
Ignoring ocean heat transport is a serious problem, but probably less serious than ignoring space-age evidence that Earth’s heat source is the violently unstable remains of a supernova that gave birth to the Solar System five billion years (5 Gyr) ago [2,3].
Despite support from government nuclear and particle science in claiming that solar neutrinos magically oscillate away before reaching our neutrino detectors, the entire Climategate (government science) fiasco will not be resolved until world leaders and leaders of the scientific community make veracity (truth) and basic scientific principles their #1 priority.
1. Solar Physics 3, 5-25 (1968):
http://adsabs.harvard.edu/full/1968SoPh….3….5G
2. Science 195, 208-209 (1977):
http://www.omatumr.com/archive/StrangeXenon.pdf
3. Nature 277, 615-620 (1979):
http://www.nature.com/nature/journal/v277/n5698/abs/277615a0.html
With kind regards,
Oliver K. Manuel
I’m sure I’ve got some big numbered spikes around here which could knock you right into port.
There’s water ahead;
Pachauri Jones, watch your speed.
Circular models.
=========
Thanks, now I’m stuck humming grateful dead song for the day.
lol, I thought the same thing.
” The paper uses climate models in the manner in which they are probably most useful: to conduct experiments using different forcing data and model parameterizations to increase understanding of both how the climate system works and the limitations of climate models.”
Though doesn’t this assume that the models are correct/complete?
Just a thought on first pass.
Though doesn’t this assume that the models are correct/complete?
Not necessarily (note the “and the limitations of climate models”). Think of the model as a programmatic realization of a theory and the model runs as “what if” scenarios. The output of the runs gives you a picture of what the theory yields under different conditions. Scenarios matching historical conditions could theoretically be used for validation. Even hypotheticals be used for invalidation should they yield something that was clearly outside the realm of possibility.
Agreed, however without understanding the validity of the base assumptions the work- as interesting as it may be, will be misleading at best.
That’s the issue, you miss one key component and while you think you’re refining your understanding of the system, you are infact only refining your understanding of that particular model system, not the climate.
I think it’s dangerous.
The models, I think, should be treated like orange test vehicles, and tested to destruction under various regimes. I.e., where do they begin to get obviously unphysical? There’s some of that in the paper, and it tends to reveal which parameterizations are the Hokey-est.
No. What you are suggesting is that until you know the answer, you can’t ask the question. Science rarely moves forward in great leaps. More often, knowledge moves forward with closer and closer approximations. In this case, testing a model can determine whether the model is sufficient to explain the phenomenon being examined.
Models are best understood to be the best you can do at the time, not a template from God. When you’re lost in the woods and you’re not sure which way to go, you really do need to take the next step. If you don’t, you’re sure to starve to death.
Actually, yes.
I’m not suggesting that at all. I’m suggesting you cannot use a model on an unknown system to qualify said system. Ask any engineer.
Used as a theoretical tool to point us in interesting locations- sure. For anything else no- and to claim that ANYTHING done on a climate model is an actual experiment is pure fantasy.
Stamp-collecting moves forwards in small steps.
Physics has big jumps – Galileo – Newton – Rutherford – Einstein – with zero progress in between.
Biggest source of meridional flows in the atmosphere are predictable and should be included in these calculations, this is why the models don’t do well with ENSO,PDO replication.
http://research.aerology.com/supporting-research/leroux-marcel-lunar-declinational-tides/
“The undergoing changes in climate caused by human activities will PROBABLY affect the oceanic circulation and its heat transport, which then MAY feed back onto theatmosphere and climate. Nevertheless, the connection between atmospheric and oceanic heat transports is NOT YET WELL UNDERSTOOD.” my emphasis.
So what prey tell are they modelling?
Anyhoo, reading the post more thoroughly, it seems that it’s reasonable work. There are FAR too many assumptions for my liking (all being directly fed into the models, which themselves are largely assumptive), but they seem to be approaching the matter in a logical manner.
The introduction as well, i found very informative; for example i’d never put two and two together regarding the continental shift and change in oceanic heat transfer. Fascinating stuff.
I’ll grab the paper and give it a thorough digest when i can.
Yes, sounds incredible:
Nonetheless, we believe the results presented here can serve as a guide for future explorations of the role of the oceans in climate
So, we are beginning to explore the role of oceans, presumably one of the main factors of climate, with the clouds. Which are a great unknown by themselves. Good to know; I am beginning to feel confident.
Judith you wrote :
…. to conduct experiments using different forcing data and model parameterizations to increase understanding of both how the climate system works and the limitations of climate models.
I would reformulate by saying :
… to realize computer runs using different forcing data and model parameterizations to increase understanding of both how the models work and what their variability is.
It is because of that reformulation that I never waste my time to read such computer run reports. I am always amazed how some people (like the authors of such reports about computer runs) don’t even IMAGINE that their computer runs might have no significant relationship with the reality whatsoever.
It is especially amazing when one reads phrases like : “The representation of clouds is one of the main weaknesses of current climate models what is an euphemism for “we have not a clue how they are formed and transported” and yet they seem to be unable to go one logical step farther and ask themselves whether the whole serie of computer runs
is relevant at all .
I have also an information for the authors who are wasting our time even worse – running computer models on Jurassic and Cretaceous.
The Lyapounov time of the Earth is around 10 MY or less what means that the orbital parameters (obliquity , excentricity etc) cannot be computed beyond a few MY .
I would bet that these authors ignore what Poincaré found out already 100 years ago, namely that the Solar system was chaotic with a consequence that orbital parameters can’t be computed beyond a finite time depending on the dynamics of the system.
A technical remark for those who may be curious : orbit stability and orbit computability are 2 different things. E.g a body may stay on semi-stable orbit(s) with widely varying unpredictible parameters.
This means that nobody knows and nobody can know what were the orbital parameters and therefore energy distribution during Jurassic or Cretaceous what makes musings about oceanic currents in such a far past rather irrelevant.
For my general education, I’d have 1 personnal question Judith.
Do you really waste your time with reading “papers” that are only containing reports about computer runs?
Or do you skim them because you consider that it is important for your work to have a rough idea what this or that computer might say?
Or do you do like me, just dismiss them in general and read only if the issue is to compare computer predictions to observed reality?
Tomas, actually I learn more from a paper like this than a paper that compares model simulations/predictions to observed reality. What do we actually learn from a disagreement between model prediction and observations? We already know that models are imperfect, and a key issue to understand is on what space/time scales it makes sense to compare climate model predictions with reality. So I don’t learn all that much from a disagreement between climate models and observations, since we don’t know why they disagree; this is the “confirmation holism” issue that I have discussed previously. Using models to test ideas about how feedbacks work is very useful, IMO.
I spoke at length on the broader issues that you raise on my previous post “What can we learn from climate models?”
http://judithcurry.com/2010/10/03/what-can-we-learn-from-climate-models/
Perhaps that earlier thread is worth revisiting.
From this particular paper, the item of particular interest to me is the link between cloud feedback and ocean heat transport. Too often reasoning about cloud feedback is done in the context of simple 1-D thermodynamics and radiative convective models. Then when people look at cloud feedback in climate models, they throw up their hands since the the different models are all over the place. This paper suggests a new hypothesis to be tested with observations, among other things.
Unfortunatley Dr Curry you STILL have the issue of not knowing what you’re testing is related to the climate.
You can for example examine the role of feedbacks in the climate models, but this does not mean that it advances our understanding on the feedbacks in the climate system as they are two seperate entities.
I understand the ‘experimental’ approach to using models, but i still think, quite strongly that it’s putting the cart before the horse; you cannot use a model on something you don’t understand to increase your understanding of that same subject as you’re always starting from a ‘moving’ start point.
Or to put it another way, we don’t uinderstand the climate enough to fully model it, so using those same models to improve our understanding of the climate to allow us to model it better is utter madness.
In a system with one or two variables, you could probably get there via extensive trial and error followed by real-system validation. With the climate? i think you’re going to struggle to confirm anything but the parameters you first put into the model.
If you view a model as a hypothesis and a hypothesis as a question, you could see a use in model experiments as helping to define which questions need to be answered with observations.
Perhaps, but you are still ham-strung by the issue that what you’re modelling may bear no realtion to the ‘real’ system. This couldeasily lead to erroneous conclusions.
On the other hand, the more questions you ask the more likely you are to find flaws in your assumptions. But I understand your point. If you use models to derive conclusions you are asking a question and using the question as an answser.
Correct. Now, i’m not saying that models are not useful in the right context (look at the engineering world for a prime example), i just question their validity when modelling the climate.
There are just far too many assumptions.
In any field of science on very important step is learning to understand the theories proposed, models and the consequences of the hypotheses made. Only, when the theories and models are understood, it’s possible to test them properly through comparison with empirical data.
Of course there is also the purely empirical path, where data is collected and compared only at elementary level with theoretical ideas, but that’s in most cases not an effective approach, most certainly not, when the system being studied is as complex as the Earth system.
Studying the models and experimenting with them is an essential part of the scientific work.
“Studying the models and experimenting with them is an essential part of the scientific work”
I’m going to have to disagree here. It’s an interesting part of the work yes, but it is not essential- purely because of it’s inherant uncertainty.
To put it another way. If we had any other method for evaluating the climate we wouldn’t touch these models with a barge pole. They’re ‘good’ because they’re all we got. Don’t let that blind you to just how subjective and open to bias they are.
If we would have. Yes, then the answer might be different, but we don’t have, and that situation is likely to persist.
So then the question is this- are we extolling the virtue of climatic modelling because they’re ACTUALLY useful, reproducable, validated and accurate, or because ‘we’re’ desperate and have nothing else?
Just because they’re all people can think of to answer a question, doesn’t make them any better.
This paper on software development at Hadley by Easterbrook and Johns
http://www.cs.toronto.edu/~sme/papers/2008/Easterbrook-Johns-2008.pdf
was linked here recently for its Figure 1.
While the subject of the paper is software development (or perhaps because of that), it’s quite informative on many related issues as well. You might find it interesting. If you have already read it, think again, what it tells about doing climate science.
Pekka, in actuality, that’s just confirmed my fears.
They do NOT perform validation. They are constantly changing the code (even past ‘freeze’ dates for large runs) and they use previous runs as controls. I’ve skim-read and will do a more thorough read later (so i’ts possible i’ve missed something), but i can also find no mention (in the V+V section) on validation against real world results.
Further, there is the constant insistance that these are experiments- which they’re not.
Are there any engineers who can comment on Pekka’s link?
i should have said in place of ‘they do not perform validation’ that what they are calling validation is NOT validation- more, tuning runs.
Labmunkey,
I understand that you may reach that conclusion. That’s natural, when the expectations set for science are those that very many have. I think, however, that such very common thinking is mistaken. Science can be done in many ways and lead to valid results through different routes.
Another link related to that is this short interview of Eduardo Zorita. Here you can find again also confirmation for your doubts in one or two sentences, but more important in my view are the comments of Zorita on the many different types of scientists.
Science is really a very open activity. It’s not limited by any fixed rules, but judged only by the outcome. Judging the outcome takes often time, sometimes very much time. Thus some scientists are noted only posthumously. It’s, however, very important to give science enough leeway and allow a large variety of approaches. For practical reasons everything cannot be supported or noticed. Obvious crap must be discarded and it should not be as cumbersome as it’s right now on this site, but freedom must be given to all approaches of some potential.
Pekka,
i don’t think i can argue with the thrust of your post and i agree that science must be open to many approaches and that what passes as common practice in one field, is not common practice in another (for right or wrong).
However my objection to climatic modelling is from a more practical/pragmatic viewpoint and can basically be summed up by this sentence:
We do not know that what we are modelling (wrt climate) has bears any relation to the actual climate.
‘We’ have some aspects nailed, some quite close, but there are far too many unknowns, far too many assumptions for us to gain any meaningful results from them. I understand the experimental approach, but you cannot then infer conclusions about the ‘real’ climate from the modelled as we do not know that they are the same (or even close; in fact the modelling success rate would suggest that they are entirely different).
We may think that we’re advancing our understanding of the climate this way, but it is just as likely, if indeed MORE likely that we’re only advancing our understanding of the modelled climate and infact could end up making completely wrong conclusions based off these models.
The models are not validated (in the typical sense) so there is literally NO justification for the basis of conclusions off them.
Don’t think i’m standing ground on this particular issue because i’m a skeptic (in fact i’m more of a luke warmer now), or because i’m anti-climate science or anything, i’m just genuinley worried that these models are sending us down the wrong avenues of research or even worse, giving us completely wrong ‘data’ on the climate, and we currently have NO way of knowing.
truncated post- continued below:
… if this is the case.
Also- i’d give my right pen… for an edit function.
I should say that I have not made in this discussion any claims on the actual scientific value of the present models. I think that it’s obvious enough that they have some value, but how large that value is, that remains debatable.
For me the reasons to think that they may have value has enough support. Improving the understanding of the Earth system is an ongoing process and the models have their role in that. Their task is not so hopeless as one might think based on some claims on the dominance of unpredictable chaoticity. Learning more about the models themselves and comparing the models with empirical data taking that learning into account is telling in future more on what the models can really provide. That’s a line of research that’s worth continuing.
I see where you’re coming from, but i just think at present and especially given the large number of unknown variables that at best we can only use models for a sort of ‘fundamentals’ check.
It’s when you try to integrate these fundamentals into the chaotic, non0linear system that i think we’re going to run into problems.
Again, the assumption that the models are useful (past the fundamental examination) relies on another assumption- that they are akin, or close to the real world system. I’ve not seen enough evidence to suggest that they are. So we come back to this;
It’s all we’ve got, so that’s why we use them. I think that’s dangerous.
One more comment.
All the above is about, how to do science, not about, how to use it’s presently available results. That’s a totally different issue and the criteria are totally different.
Good point.
Oh and thanks for taking the time to debate this. One day somethig will get through my thick skull :-)
Thanks Judith, I think I understand (partly).
I must admit that I didn’t pay much attention to the thread you linked precisely for the reasons I mentionned above.
But I will try to revisit it.
Btw I fully agree with your last paragraph and as you know what my idea of a climate science paradigme is , you will not be surprised that I am not surprised that the models are all over the place.
While I fully agree that the ocean-clouds coupling is the SINGLE major issue about the Earth dynamics, we probably disagree about what the computer models can tell about this coupling.
For me the runs are to 90% useless because I estimate that the dynamical states computed are not realistic for 90 % of them.
And as there is no procedure which would allow to distinguish the 10 % possible from the 90 % non realistic, spending time with these runs is at least 90 % wasted time converging to 100 % when one begins to extend the testing from a single model to N models with different properties.
Just to illustrate by an example .
They say “We found that this model warms 0.8 K when the OHT is increased from 0 to present day values and 0.4 K from presentday to two times presentday heat transport.
What is to tell me that our real Earth has anything to do with this model and that these numbers are not just some random numbers between 0 and 1 ?
Especially if they imply that just by changing the parametrization (or model), about any number can indeed be produced.
There is absolutely no theoretical background for the dynamics. Just some qualitative beliefs.
And when they mention far past climate as possible validation of what they are doing, they clearly show their ignorance about the Lyapounov time I reminded above.
Isn’t all this per definition time waste?
Through time and space, the models are strangely attracted to 100% irrelevance.
=============
Judith-It would be interesting to have another thread on the development of “climate models” and how they are validated.
It seems to be a different practice in the development of “climate models” vs. the development of “weather models”. It appears that “weather models” follow the same or a very similar practice to what is used in the development of models used in the aerospace industry, but developers of “climate models” do not.
The way “climate models” are being developed it seems very difficult to validate the model. In the aerospace industry we would not even discuss the output of a model unless it had met the evaluation criteria set when it was decided to develop the model. That would be the only reason to discuss its results.
In “climate science” there seems to be endless discussion of models of dubious quality to simulate the real environment. Why is that– other than those studying the climate often trying to claim they understand the situation more than they really do??
Rob, i’ve done a few threads previously that include this topic, but I agree it deserves more discussion, I’ll add it to the list.
Agreed. As far as i am aware, the climate model validation is crude to non-existant.
A post on this is coming soon
I would be very interested in the model discussion to know whether the climate models “predict” phenomenon like the Gulf Stream, or such realities are just hardwired in.
That would be brilliant.
Dr Curry
But nobody from the climate establishment ever tests the hypothesis with observations. Nor do they, in this case.
Application of UNRELENTING VIGILANCE to keep knowledge gained from observation CLEARLY SEPARATED from alleged “knowledge” gained from abstractions is the ONLY sensible course.
Earth Orientation Parameters (EOP) are the arbiters of climate disputes. OHT & clouds do not have adjustment knobs that are independent from surface winds & circulation more generally. Whether EOP are integrating or aliasing, they are informing CLEARLY about both stationary & changing GLOBAL frequencies. If this info isn’t built into models, I would sternly suggest that the modelers find some time to reflect soberly and escape their blinding paradigmatic immersion.
I will have more to say on this soon…
Best Regards,
Paul L. Vaughan, M.Sc.
At least a discussion about the science without injecting AGW “beliefs.”
His characterization of the tropical weather regime as “low clouds” is quite laughable. As Willis Eschenbach is discussing over at WUWT (
http://wattsupwiththat.com/2011/06/07/further-evidence-for-my-thunderstorm-thermostat-hypothesis/), and as anyone who lives in the tropics knows from daily experience, the weather regime is a a complex and non-linear cyclic phenomenon dominated by massive thunderstorms that develop and move in response to local warmth and moisture conditions, which are a result of solar insolation and interaction with the sea surface.
These heat engines are clearly a fundamental factor in heat transport but they get nary a mention in the models. Sort of like studying the performance of Formula One cars without mentioning their engines.
Am I missing something here, or are the climate models really as primitive as they seem to be?
Yes, you are correct about people living in the tropics, even the subtropics. The mechanism that Willis described is visible to anyone who is fascinated by nature. You state, correctly, that the models simply do not address what Willis described. Being a tad more explicit, I would say that the models do not address either the mechanism described or the observational data that is specified implicitly through the description of the mechanism.
I asked Willis whether there is overlap between his account and that offered by climate models. Of course, that is a huge undertaking and I am not surprised that he did not bite. However, it seems to me that the key to understanding the failure of models is understanding whether they have systemic shortcomings that prevent them from addressing this climate pattern that is easily observable throughout the tropics, subtropics, and maybe elsewhere.
As a first guess in such an investigation, it seems to me that models are designed in such a way that they describe only mechanisms of heat transport that are caused by radiation and that they specify only measurements of such “observable” facts of radiation. By contrast, Willis’ description of the mechanism makes essential reference to sunlight and the effects of clouds on it. Is it possible for models as formulated today to describe sunlight and clouds without treating them as non-entities that must be analyzed into claims about heat transport and radiation? In summary, is it possible for models today to be used to address the humanly observable phenomena that Willis described?
ICTZ only takes couple of degrees of latitude wide band along equator, but tropics take, by definition, 46-degree wide band. ICTZ is broad over continents, as well as wery warm pools of water around sunda shelf and in western pacific, but elsewhere it only occupies narrow band along equator, low stratus clouds occupy much of the rest.
Yesterday Roger Pielke Sr. commented on a recent EOS article concerning models.
http://pielkeclimatesci.wordpress.com/2011/08/15/a-new-agu-eos-article-titled-guidelines-for-constructing-climate-scenarios-by-mote-et-al-2011-which-inadvertantly-highlights-this-flawed-climate-science-approach/
Bob thanks for spotting this, trying to find a copy now.
I think I’ve never heard so loud
The quiet message in a cloud.
Memo to Arthur Smith, documenteur extraordinaire for the APS: Track how long I’ve been saying that please; it’s important.
=====================
In 1924, in his paper ‘World Weather’, Sir Gilbert T. Walker, British mathematician , introduced the terms Southern Oscillation, North Atlantic Oscillation, and North Pacific Oscillation.
Nearly 90 years later, today’s climate scientists’ degree of understanding:
Until now, it is not clear which are the mechanisms driving North Atlantic Oscillation.
The causes of the Pacific Decadal Oscillation are still unknown
Southern Oscillation .. The actual cause of this is unknown. …
http://www.vukcevic.talktalk.net/AP.htm
Any takers?
Here are my thoughts on the “unknown” drivers of the global circulation.
http://research.aerology.com/aerology-analog-weather-forecasting-method/
The study could be important if the changes over time that are related to oceans and cloud cover–which are both nominally dependent on the Sun–help give us a glimpse of what truly is the key independent variable that explains both global warming and global cooling and why, according to Dr. Don J. Easterbrook, 9,099 of the last 10,500 years were warmer than 2010.
Here is the link:
http://www.rmets.org/pdf/ww5.pdf
I think it be a pity when climate models are seen only through the lens of climate prediction. That aspect exists, and much time, labour and electricty goes into it. However it does not constitute a particulary well constructed experimental design.
I have done what I can to look into this and it was well summed up by the hilarious phrase “ensemble of missed opportunity”.
It seems the fashion to suggest that modelling groups do little else but produce gloom and doom scenario runs and that this is the extent of their effort.
To suggest that they do not know the issues with the models and how that makes any failure to interpret the predictions without recourse to meta-knowledge concerning the models, likely to give false impressions. This is a failure that we all have a part in, particularly the media and the pundits.
Even still. it seems to be the case that we have not spent sufficient time and effort on the task of interpreting the modelled behaviour. Letting the models speak to us. I would characterise this as a task at least as daunting as building the models and running them.
The CMIP5 archive, when complete, will be a rich resource but it will be huge and likely opaque. To give an idea of it, I have heard it said that if one wished to download the data using a 8Mb/s broadband link it would not finish loading before 2050.
But there is much more than that, some of the most interesting runs are the most artificial, say a stabilisation run, a run when it is always January, or a run for an ocean world. These are useful in an analytical way, in that amongst other things they allow the statistical nature of the models to be explored and fedback to the theoreticians.
I believe that there is a view amongst some of the modelling groups that it might be better use of the large simulators if they were permitted to explore the parameter space of the models. To conduct well designed experiments on them as well as with them. The data produced having many uses not the least of which is the development of sophisticated emulators for which many users are crying out.
The scenario projections which are in effect predictions are more interesting than useful in any science sense. They are an all singing and dancing Rolls Royce approach to a largely non-science issue. It might be arguable that the same, and vastly more alternative, scenarios could have been produced by emulators, if we had invested in them more and had better explored the parameter space of the simulators to characterise them.
That would not have done. For it would have been unacceptable not to have used the most advanced simulators, for largely political reasons.
It seems also the fashion to suggest that the modellers are somehow in awe of their achievements, or that they believe the models as some sort of font of gospel tracts to be revered. Such may exist, but the internet has recorded modellers talking amongst themselves being rather rude about the models.
I think that Judith, knows some of these people as well as others that have an optimistic but sane overview of the bigger picture. Hopefully she will read this and comment as to where I am at odds with her view. I know I must agree on her points regarding the use of models that she gave above.
Alex
It is difficult to get scientists to talk about models because scientists are by nature realists and insist on talking about “the facts.” As far as I know, no one is doing what I would consider serious evaluation of the models as models, without regard to matters such as prediction. The first I want to know about a model is its list of primitive predicates. To use a familiar analogy, if I were investigating a model used for subatomic physics, I would want to know if it contains a primitive predicate such as “_____has spin.” Having found this predicate, I would know that the model can yield a result about a subatomic particle that spins. Then I would want to know the formulations that the predicate occurs in. My guess for climate models is that they contain only predicates that can be used to describe heat transfer caused by radiation; that is, they contain no or few predicates for describing observable natural processes such as cloud formation in the tropics. Until you have done this kind of analysis, you do not know what your model is talking about (can refer to).
Theo,
Figure 1 on page 3 of this paper gives an overview of the components you might expect to find in a GCM.
Thanks. But you need the list of primitive predicates before you can have a clue what the model can refer to.
Gene,
Excuse me. I should have added that these papers look very good. They are written by programmers. Programmers have a profound understanding of the workings of models that scientists do not have an opportunity to acquire, so long as they do not become programmers rather than scientists.
It rarely occurs to the non-programmer that when you create some huge model you are confronted with the problem of getting it so solve. How is that done? How do you solve your true description of the world? It is done with heuristics that fall outside the realms of science and mathematics. Someday Departments of Programming Heuristics will be as prominent as Departments of Applied Mathematics.
Programmers who support climate modelers are learning a great deal about programming heuristics from their work. Such achievements are very important.
I just wish they would apply for grants to study programming heuristics rather than grants to model climate.
This is somewhat off topic but I’ll justify it by saying the paper is about models and polar climate (which I guess is in some way linked with OHT).
A new paper is under discussion and review at ACPD (http://www.atmos-chem-phys-discuss.net/11/22893/2011/acpd-11-22893-2011.pdf ). It looks at modelling temperature change in the arctic with a new generation Canadian model. The language seems beautifully balanced. Having read one of your earlier comments Judith I’m guessing this is a study that you disapprove of as it looks at the correlation between observed and modeled temp. Although it seems fine to me that this might just be a first step before answering some more interesting questions, Those questions are eluded at in the text but (frustratingly) remain unanswered. I’d also take a look at the three Chylek papers in the reference.
If (amplified) arctic temps are at least in part due to poleward transport of energy, and if in general IPCC 2007 models are getting those temperatures hopelessly wrong for the 20th century whats the basis for thinking these tools are good bases on which to perform these OHT experiments?
Look at the themperature history. Every time it gets warm, it then gets cool. Every time it get cool, it then gets warm. A proper model will have something in it to use warm to cause cool and something in it to use cool to cause warm.
When we are warm, Sea Ice is melted and it snows more.
When we are cool, Arctic Water is frozen and for lack of moisture, it snows less. This is the powerful negative feedback that all of you are missing.
Who, me? Miss?
============
All is wrong. All Always Never should be used carefully.
Most on different sides ignore albedo effects of snow
That is the thermostat of earth
That should have been Herman A Pope and not W
I am on a travel tablet and pushed the wrong thing
The oceans and lakes that comprise the largest part of the Earth’s surface accumulate heat over time. Moreover, the oceans also can transmit heat and cool over time. And, the satellite data has been telling us that the Earth has been cooling for a decade. The Oceans are in a cooling trend. And, in a period when the oceans are cooling, there is no global warming during that period; and, there is no end to the cooling in sight.
Judith, I am so frustrated reading another paper which is “model-centric” and leaves me unconvinced of the conclusion because the methodology could well be meaningless. I believe climate science must try new approaches to better understand this chaotic system. I would like you to start a new thread which challenges your bloggers to propose alternative methods to deal with this problem. A hand full of scientists (Spencer and others ) are attempting such approaches. But we need more of this “out of the box” kind of thinking.
Please do not forget Svensmark. His work submits fully to scientific method. He offers a huge contrast to the climate modelers. Yes, his work might not be the final word on global warming/climate change/climate disruption/climate weirding/climate ?/ but it will live in the annals of science.
Yes, Theo I agree. I felt I learned something important after reading Svensmark’s work. I have been looking for rebuttals. So far it seems quiet.
There is no such thing as climate science. To be a science, it need either to observe the object of the study in several independent or weakly-interacting cases, or be able to construct independent experiments. Computer simulations are not experiments! They never will be experiment! Everyone who call them ‘numerical experiments’ is just an ignoramus. Climate science is not science, and never will be. it may be observational science like paleontology or history, but nothing more. And until we travel to other planetary systems and observe hundreds of other planetary systems, it never will be. Everyone who calls themselves ‘climate scientist’ is just crying out to the world about their own ignorance.
The climate science may be called climate mathematics, but nothing more.
The whole scientific method process went off of the rails in the 50’s when They started “peer review” and subverted all of the funding into numeric modeling and there by shut out any kind of cyclic pattern research into weather and climate drivers. Now that the numeric models have run to the ends of their effective forecast range of 7 to 10 days when the cyclic process over come the artificial “physics” that don’t match reality, the study of the cyclic patterns is the only way expand forecast ability into months and years into the future.
It would have been much more productive if both areas of investigation had been synchronized for a more productive combination of the real drivers smoothly understood than now playing catch up in a hostile environment of CO2 hysterical CAGW collapse burning up all available funding, and endless bickering of the useless talking points as they die.
I would much rather keep searching for the blend of short and long term cycles, coupled with weather models to develop the whole picture so we could proceed with good long range forecasts that will prevent major surprises like the blocking highs, droughts, intense monsoonal flows, that are predictable by the combined cyclic patterns being incorporated into the numeric models, they currently swamp past 5 days.
For the relation between OHT and temperature (and atmospheric heat flux) there seem to be good descriptions using Maximum Entropy Production, a principle that applies to systems far from equilibrium.
A nice overview with MEP explained, and with discussion on various climate variabeles here.
We all know nothing escapes from the Second Law.
The second reference is to the paper by Kleidon. I must say that this paper is not convincing at all. It’s to a large extent based on a very bad example (there are numerous questions on, why the example is set up as it is, and changing the setup invalidates the results).
As a whole the paper makes me very doubtful of the whole concept. The paper tells explicitly that no formal justification exists for the principle of maximum entropy production, and the plausibility arguments appear to me rather implausibility arguments than plausibility arguments.
The concept is too flexible to be a real principle. It’s quite possible that there are situations, where it can be justified on case specific arguments, but that seems to be the best that I can conclude about it.
I have looked at some other material in addition to the Kleidon review and that’s has only strengthened by doubts. The practical attempts to apply the principle appear to have led to erroneous conclusions.
From the theoretical point of view we know that the second law can be derived from statistical physics. It tells that the macroscopically observed results are due to simple laws of statistics and in particular to a large extent determined by the size of the phase space that corresponds to each particular macroscopic state.
If the principle of maximum entropy production would be something similar, as the whole idea behind it is as far as I understand, then there must also be a similar way of deriving it from laws of statistics. Looking at the concept from this point of view tells also, how it should be defined in detail. This approach to the concept has led to my skepticism on its real potential.
I have read the Barreiro/Masina paper, but admittedly in a somewhat cursory fashion, so that I might have ended with a different interpretation if I had scrutinized it more carefully. Nevertheless, it left me puzzled. A major conclusion was that with a more than modest increase in poleward ocean heat transport (OHT) from equatorial regions, we would begin to see a net cooling rather than warming, because increased low stratus cloud formation in the tropics would increase planetary albedo more than warming in polar regions would reduce it through cloud reduction and ice melting.
What puzzled me was the mechanism involved, because the increase in tropical low cloud cover was attributed to subtraction of heat from the tropics (surface cooling) as part of its more rapid transport poleward. This seems to me to be a rather artificial scenario, however. Ordinarily, an increase in OHT is expected as a consequence of increases rather than reductions in tropical SSTs, leading to a meridional gradient that drives heat toward the poles (perhaps more toward the Arctic than the Antarctic but that might be a misconception on my part). Under these circumstances, the increase in OHT would not be accompanied by equatorial cooling, and so the cloud changes that were modeled would appear not to apply.
I assume that ocean heat does not decide on its own to move faster. What would cause OHT to increase while tropical SST declined? Under such a circumstance, the model simulation would be relevant to expected temperature changes, but I didn’t notice in the paper an indication of how that circumstance might have arisen in the past or might again in the future.
Good point, Fred. While many other commenters saying why models are inadequate in general, you explained why it should be easy to ignore this paper. After all, there are only three ways for energy to leave the tropics: a) poleward (which the authors have artificially increased), b) downward (which can’t change when a slab ocean is used), and c) upward. Do we really need models to show that if you increase a) and hold b) constant is it any wonder that the tropical atmosphere is colder and cloudier, depending on how poorly your model handles low clouds? Does the amount of additional energy one can artificially move poleward without causing surrealistic changes really mean anything?
Model and experiments
As in the study of H05 we use an atmospheric model coupled to a slab ocean. This configuration has the advantage of allowing the prescription of ocean heat transport, thus facilitating the study of its role in climate. The slab ocean allows air-sea thermodynamic interactions, but does not allow the ocean to adjust dynamically to changes in the wind stress. Since changes in the surface stress may provide a (positive/negative) feedback that is not realized in the model, the solutions presented in this study have to be further tested in a coupled model configuration.
http://www.fisica.edu.uy/~barreiro/papers/BarreiroCherchiMasina2011.pdf
Vukcevic – you have linked to Walker’s 1932 paper (World Weather V). His 1924 paper (published in 1926) titled Correlation in Seasonal Variations of Weather, IX can be found at http://www.rmets.org/pdf/classicindia2.pdf . In that paper he clearly discusses the NAO and SO.
Thanks, there are number of his papers I went through recently, including series http://www.rmets.org/pdf/ww3.pdf to ww6.pdf , all very interesting, worth revisiting by the contemporary climate science.
Judith,
Have they realized yet that cloud cover NEVER crosses the equator?
Hmmmm….Circular motion, where have I heard that before?…..hmmm.
This paper is a good start and reflects my previously published contention that ocean warming of the atmosphere from below affects the width of the tropics thereby pushing ALL the surface air pressure systems poleward in both hemispheres with implications for global cloudiness, albedo and solar shortwave input to the oceans.
However in my opinion it is only half the scenario.
Additionally the level of solar variability has a similar effect from the top down.
I think it will soon be recognised that all climate change is a consequence of the interplay bewtween the two forces which sometimes supplement each other and sometimes offset each other.
Stephen,
The circumference sizes and different speeds have absolutely nothing to do with circulation?
Still take 24 hours to complete a single rotation in both areas even though the circumference difference at the equator and poles.
So, one has to be revolving slower than the other in a complete rotation.
Of course they do.Where did I suggest otherwise ?
The rotation affects the flows between hemispheres and distorts the patterns but in the end it is the average latitudinal positions of the air circulation that changes to reflect alterations in the global energy budget.
Stephen,
There is not a single piece of literature on the use of planetary size difference in circulation.
This also includes the atmosphere size difference as well which also have different speeds.
Suns distance to the poles and to the equators also means a difference in dissipated heat from the sun. The greater the distance, the less compacted the radiation is. Centrifugal force from the rotating planet allows pressure by exerting pressure to the atmosphere.
This still has not touched the interaction of magnetics of the magnetic field, nor the importance of ocean salt.
All kinds of literature on ocean currents, air currents etc.
Lindzen-Choi have a revised paper on sensitivity which Watts has posted.
http://www-eaps.mit.edu/faculty/lindzen/236-Lindzen-Choi-2011.pdf
I cannot resist to tell an anecdote that I think highly relevant to the “computer model” issue and which also explains where and how my opinion about computer models come from.
Just after having finished my studies I have been drawn in an ambitious project. I was young , about the age of our own Colose or slightly more and in retrospect I realize that I thought that I knew much while knowing (almost) nothing.
The project was to determine certain properties of a multiphasic flow in transient dynamics . So no “equilibriums” or lazy “steady states” – full blown dynamics.
It didn’t take much to realize that the problem could not be treated by simplification and all analytic attempts fast got bogged down in untractable mathematical complexities. But as my generation (compared to older scientists) has been trained with computers, I decided that we’d realise a numerical model.
And so we did. The model was packed full of settled science and peer reviewed papers. It was so darn perfect that it was more real than the real flow itself.
In the first series of runs we got interesting results. It didn’t look stupid and one could always hang behind a chart a qualitative explanation that fitted the chart. “Ah sure, this variable shows this strange oscillation because the gradient must have done this.” The thing seemed to work.
The result were not stupid (I say “not stupid” because I can’t say “correct” – nobody knew what was correct) and we had always a non stupid narrative that fitted the data too.
Then after a few weeks I wanted to see what happens when we go to really large time scales. But as the model was extremely computationaly intensive and it was hard to say what EXACTLY made the model behave as it did , we never got long periods of time available for a single run.
So one day I launched a run on Friday afternoon and let it run with the same set up through the week-end.
Monday morning of course I ran to the computer to see the dynamical state.
Well…the temperature of my flow was enough to vaporize the lab and transform everything in plasma :)
Logically, using normal flow equations for plasma conditions yielded nonsense and the model was just continuing to compute nonsense.
Obviously to allow such dynamics, the model was somewhere not EXACTLY conserving energy or momentum or both.
To make a long story short. There was indeed a problem which was 80% code (so just software) and 20% physics. It took a rather long time to find the software part.
This error was too small to show in the shorter runs we did. It also had different propagation features depending on the conditions of the run.
Once the model corrected, it still showed non stupid results but quite different from the first.
In hindsight we were able to identify anomalies that actually already showed in the first runs but nobody could have found them if he didn’t know what to look for.
So obviously the narratives that fitted the first results were wrong too but as everything seemed plausible, everybody thought that we had the truth of it or very near.
I didn’t tell this anecdote to make an analogy between the dynamics of multiphasic flows in spatio-temporal chaos conditions and the climate.
But I told it to show that as soon as ANY transient physical process needs a large sowtware (hundreds of thousands of lines) then there is absolutely no possibility to be sure whether the results beyond looking “not stupid” are actually correct or approximately correct.
Human brain is simply unable to assess intuitively what those millions of operations really do and what complex error propagations take place due to algorithm or to rounding up.
Anyway as the climate is infinitely more complex than my chaotic multiphasic flow that I studied years ago, I can only draw the conclusion that a non validated climate software has about 0 chance to produce anything correct.
Yes it may look “not stupid” but still be completely off the real dynamics.
Just look at a simple example: the average temperature.
For a given average temperature, f.ex 14°C, there is an infinity of temperature fields who share the same average.
Only one field among the infinity is correct.
10% of the fields are wrong but dynamically near to the correct one.
60% are wrong , farther from the correct one but they look plausible.
40% look stupid.
If your computer shows states (through multiple runs with varying parameters or conditions f.ex) which are mostly among the 60%, and you are happy because they look plausible and you like the average, then you can be horribly wrong because those 60% wrong states will follow a very different dynamics than the one which is correct.
And you would never know that you were wrong unless you finish one day to compare your model with real data over long periods of time.
Tomas,
I agree fully that the issue you describe are real. There are strong versions of confirmation bias in the development of large models. The models are improved and honed until the results are reasonable, but that may be for wrong reasons. I’m sure that all serious Earth system modelers are fully aware of that problem. Unfortunately being aware of the problem is not the same as solving the problem.
The paper of Easterbrook and Johns I refer to above tells, how the people at Hadley work to reduce the problems (preferring sometimes “better physics” to better skil, doing model intercomprisons, etc.). The approach alone doesn’t tell the level of success, the working scientists see much more, but they can be blind to some issues. There are also generic issues of the type that all dynamics cannot be modeled at all with the existing approaches, while still possible for the real system.
All this means that it’s right to be skeptical of models, but it doesn’t tell, how far we should believe anyway in, what the models tell. This is an issue, where one may err seriously in either direction. Without working knowledge on the field of research the changes of proper judgment are close to zero. With working knowledge the best scientists have more basis for their judgment, but how could we decide, how much they really know, or how difficult the most important obstacles for success really are.
FYI -“New paper from Lindzen and Choi implies that the models are exaggerating climate sensitivity.” Posted on August 16, 2011 by Anthony Watts
http://wattsupwiththat.com/2011/08/16/new-paper-from-lindzen-and-choi-implies-that-the-models-are-exaggerating-climate-sensitivity/
Dr. Lindzen has the full paper on his personal website here:
http://www-eaps.mit.edu/faculty/lindzen/236-Lindzen-Choi-2011.pdf
“… Dr. Ernest Njau and Dr. Boris Komitov … believe that we are about to enter our next climate era, one with a long lasting deep cold period. They are highly respected scholars who have helped lead the way in their own countries in conducting in-depth research that has identified solar activity behavior as integral to the prediction of the Earth’s climate variations.” (SSRC, 2008)
I agree with the poster who wrote “We may think that we’re advancing our understanding of the climate this way, but it is just as likely, if indeed MORE likely that we’re only advancing our understanding of the modelled climate and infact could end up making completely wrong conclusions based off these models.”
We already know that AGW model-makers’ GCMs fail grandly. All you have to do is compare GCM predictions to reality to see that the these models are mere toys and are not to be taken seriously.
GCM model-makers simply apply perimeters to make models fit historical data. and in doing so indulge the fiction that their GCMs will be reliable in the long range. They can’t show verifiable results but still they want you to believe that in a time far, far away into the future–where validity is impossible but faith makes everything possible–the schoolteachers are right and we’re all doomed if we do not listen to them and change our evil ways.
Needless to say, thinking that schoolteachers will save the world is the sort of leap of faith that separates global warming anti-science from any real science that is to be taken seriously. A belief that has not been validated and can never be validated is what we call superstition, not science!
Besides, we know what causes global warming. Nominally: it’s the sun.
Whether the models used in the paper represent reality or not, they do represent the reality the AGWers live in. And so from the point of view of AGWers’ understanding of the climate, the results have meaning.
…and from what I could see, that meaning questions their assumptions on sensitivity as defined by past climate because increased OHT values dont appear to have been able to have been able to create the earth they believed existed in their AGW modelled world.
Pekka
I more or les agree with what you wrote.
But if I was to draw an analogy with my experience and the climate, it would be to make a GCM run over thousands of years just to see when the models predict the next ice age and the transit to the next interglacial and how GHG impact the arrival and end of this next ice age.
I know that it involves making variable (very few) parameters that are held constant in runs over shorter time scales and that the computation time would be prohibitively long.
But these parameters are not hard to add and time can be found.
If such a test shows really stupid dynamics, then we’d know that the model is necessarily wrong.
Of course if it shows something not stupid it wouldn’t be a proof that the results are correct but it would qualitatively increase the confidence that in some not well understood way, the model captures some of the defining dynamical features of the real system.
Hi Vukcevic, Thanks for the link to World Weather III. If you find links to WWI or WWII (the papers not the wars) I’d appreciate a reference.
Thanks.
Odd that on a thread with the subject of ocean heat transport, the discussion is mostly atmospheric stuff.
There is a 600 pound gorilla (sorry for cliche) here – or more precisely a several million cubic km “gorilla” – namely the heat content of the oceans themselves.
In addressing why increased ocean circulation (or “OHT”) could cause cooling, one can simplify the question by ignoring the atmosphere and simply addressing the heat and temperature distribution in the ocean – in particular the vertical profile. This is characterised by strong vertical stratification and a gradient from warm surface temperatures, through the thermocline to a uniformly cold deep water column. There is one very simple corollary from this: ANY significant increase in deep vertical mixing, either locally or globally, results in downward movement of heat – movement of deep cold water to the surface and vice versa.
On the basis of this simple consideration, it is inevitable that when OHT involves deep vertical mixing, then SST and thence atmospheric cooling will result.