Bishop Hill Great Evans above
Sep 26, 2015 Climate: Models Jo Nova carries a rather interesting piece today about some work done by her husband David Evans, who thinks he has uncovered a rather major flaw in the mathematics at the core of the basic model of the climate.
The climate models, it turns out, have 95% certainty but are based on partial derivatives of dependent variables with 0% certitude, and that’s a No No. Let me explain: effectively climate models model a hypothetical world where all things freeze in a constant state while one factor doubles. But in the real world, many variables are changing simultaneously and the rules are different.
Partial differentials of dependent variables is a wildcard — it may produce an OK estimate sometimes, but other times it produces nonsense, and ominously, there is effectively no way to test. If the climate models predicted the climate, we’d know they got away with it. They didn’t, but we can’t say if they failed because of a partial derivative. It could have been something else. We just know it’s bad practice.
This sounds plausible to me. What do readers here think?
Reader Comments (199)
Though Golf Charlie tends to be <3% right 97% of the time, experience suggests that if he had more time at his disposal he'd work his way up into A league climate denial by being 97% wrong 103% of the time.
This is really remarkably simple and clear despite the Climateballer presence here. Even a Pesky Wrabbit is right every once in a while.
1. There is no inconsistency in the use of partial and total derivatives in modern PDE theory or practice and that includes Navier-Stokes, even though in GCM's due to the pile of FORTRAN problem it is remarkably difficult to tell due to the use of outdated and non-conservative methods.
2. Sub grid models are where Evans may have a point, since they are wrong, the only question is if they are disasterously wrong.
3. ATTP (Ken Rice) had a post last week on a paper by Trenberth showing the truly Herculean nature of the task of even getting close to answering this question of how wrong they are.
4. Problem areas that immediately present themselves are tropical convection (Isaac Held has an old post on this showing its ill posed), clouds, and turbulence. These things can make a big difference to the (relatively very small) changes in energy flows in the system despite the insistence of Climateballers that they don't, based on conservation of energy.
My advice to Evans is to go back and think this through again and take down his post until that time. He can only do more harm than good by posting obviously wrong material.
Funding
http://pjmedia.com/tatler/2015/09/22/government-funded-scientist-wants-to-use-rico-laws-to-investigate-climate-change-skeptics
I'm not sure that invoking Navier-Stokes means that GCMs are off the hook in that regard. In particular that while in ideal model worlds (i.e. Newtonian and PDE), to be tractable in GCM world N-S needs to be simplified and its solution produced numerically in time slices, all of which needs assumptions to be made. At the most trivial the range of model outputs shows funny things may be going on GCM world caused by their implementation of N-S alone (and we don't get to see the real sports).
Sorry, should read " .. while in ideal model worlds (i.e. Newtonian and PDE) all may be well, ..."
LA I'm not sure you should ask how the night shift are paid. Just accept that it leaves them free to vent their bitterness between clients and that they will make work for the cleanup crew in the morning.
The work is, unfortunately, irrelevant. It's because the planet operates thermodynamically to minimise radiation entropy production rate. Hence there is always net mean zero surface IR absorption by atmospheric GHGs.
Take out this 140% increase of heat transfer, surface to atmosphere compared with reality, and the false 'back radiation' concept used to reduce that excess to 40%, another failure to understand the difference between exitance and real EM energy flux, and the maths becomes much easier and controlled mainly in the first ~2.6 km of the atmosphere.
We just know it’s bad practice., or normal pratice in climate 'science '
Disinvest from 97% of climate science. The only losers, are not worth any more money. They have wasted everything they have already taken.
...and NOBODY responds to Harry Huffman. Why is his theory not the subject of intense scrutiny? Why is it constantly overlooked? Why hasn't it had its own post where his theory can be dissected to the degree that the claims - if correct, undoubtedly deserve?
knr, the prats of climate science, have danced on thin ice for too long. Fortunately for them, the ice was thicker than they thought possible, and allowed a greater density of climate scientists to engage in ever more extravagant money flinging routines.
Climate science conferences could be held on sea ice, with great confidence, that unusual weather conditions will not disrupt anything at all. Polar bears would probably turn up to watch, of their own free will, eager to show opportunistic predators how to survive without financial support, in a hostile environment.
Isn't this just a way of saying that modelling (integrating) non linear partial differential equations is beyond our current computational ability,. and the crime of climate modellers is to assume that the differential equations that govern climate are linear, and the non linear stuff can be 'parameterised..'
Its a bit like modelling a sniper where if there is no wind he kills the he president, if there is a 20mph wind he misses the president, and saying of there is a 10mph wind the president is 50% dead.
The point about non linear differential equations is that you can't treat with the variables separately - each one depends on the others for the value of its derivative.
https://en.wikipedia.org/wiki/List_of_nonlinear_partial_differential_equations
Look at how many of those are the physics of natural systems ....
At first glance, Evans' post is extremely confused. At second glance it is just plain wrong, and will leave a lot of scientists scratching their heads, especially if they have worked in CFD, seismic processing or multiphase fluid flow through porous media.
There are numerous numerical models which do an excellent job of modeling physical behaviour in complex systems. where the governing equations have to be expressed as a set of simultaneous partial differential equations. In many applications, the modeler has little choice in the form of governing equation used; this is especially true if he is modeling a well-understood, well-tested physical process; the governing equations are determined by the assumptions made about the physical process being modeled. The modeler does however have a great deal of choice in how the equations are solved, most notably in the form of solution routine used as well as the type of gridding used to characterise the system.
These systems can and do produce multiple interdependent output variables (solution vectors) which are testably correct. i.e. They solve the governing equations within acceptable accuracy, and they can be tested against observational data or analytic solutions of the same problem. THERE IS NO GENERAL PROBLEM HERE, hence the general point which Evans seems to be trying to make is neither correct nor general.
"Let me explain: effectively climate models model a hypothetical world where all things freeze in a constant state while one factor doubles. But in the real world, many variables are changing simultaneously and the rules are different."
The above statement is just silly. Evans seems to be confusing not two things but three things here.
(i) The choice of explicit vs implicit solution routine. Numerical models need to step forwards in time, predicting a state at time n+1 from what occurred at time n. This is called a timestep. If you have multiple dependent outputs as is often the case, then you can solve the equations for a timestep either explicitly or implicitly. Explicit solutions will freeze some of the variables to calculate local partial gradients of the variables of interest, and will then update all of the variables in the solution vector - before moving on to the next timestep. Explicit solutions need very small timesteps (sometimes unacceptably small) to converge to a reasonable solution. On the other hand, implicit solutions will, within each timestep, continuously update the variables until the final solution vector satisfies the input governing equation to some required convergence level, and only then will the algorithm move on to the next timestep. This is far more computationally intensive than an explicit formulation on a per-timestep basis, but sometimes offers a trade-off in terms of more accurate solutions and longer time-steps. Between these two formulations is a whole raft of possible semi-implicit variations. The two GCMs which I have examined both use some form of (not very sophisticated) semi-implicit formulation, but I do not know what other GCMs use.
(ii) The doubling of CO2 (actually quadrupling in the recent suite of CMIP5) is just one of several prescribed numerical experiments which allows a comparison of GCM results. Personally, I am very grateful for these results, since it does allow some very interesting conclusions to be drawn.
(iii) Interpretation of GCM results using simple "emulation" models. These results frequently use a highly simplified forcing feedback formulation.
If Evans wants to present an argument that the formulation is not solving the PDEs correctly or that a particular GCM is using the wrong governing equations, then he can set out his argument and I will listen to him.
If Evans wants to complain about the inclusion of the CO2 quadrupling experiments in the CMIP5 suite, then I profoundly disagree.
If Evans wants to complain about the use of a particular post-hoc emulation model, then he should set out which emulation model he wishes to complain about. (He has a problem with Planck??)
Instead his presentation of the problem of solving PDEs as the discovery of the "misuse of a basic maths technique" just looks plain silly to me.
"As ATTP points out Evans never really says what he is waving his hands about."
"Sep 26, 2015 at 10:36 PM | Eli Rabett"
============================================
"hand waving" is the new insult isn't it? You guys are very tedious. Press the buttons. Watch you go. Hand waving. Baaah
There is clearly a substantive point about the mathematics of partial derivatives. It should be for the modellers to show that their models are free of such flaws.
I have long maintained an interest in chaos theory and complex systems: it seems to me this recent Santa Fé paper is highly relevant to climate modelling:
http://www.santafe.edu/media/workingpapers/15-06-017.pdf
Abstract:
100% of climate models are wrong. Climate science wants to tweak the existing models to improve them, whereas Evans is reviewing some of the assumptions made by all climate models.
In the broader field of 'what went wrongology', those waving their arms the most frantically, claiming there is nothing wrong here, are normally wrong. What they are wrong about, may not be obvious to anyone, but if 100% of experts assume they are 100% right, but evidence proves them wrong, I would not pay the self appointed experts for anymore work.
It might not be CO2 after all, just natural variation. Nothing to destroy mankind about. Keep calm, carry on.
Just to add to what David Young has already written above. Typically, partial derivatives are used as a way of breaking a complicated differentiation down into manageable components. Evans might contemplate the chain rule for partial differentiation, for example. They are seldom of primary interest by themselves, for the reasons Evans notes.
This thread contains the output of some very good minds but sadly many are misdirected. As with many discussions, the usefulness of contributions depends upon the aim of the discussion which in this case is not clear but if the aim was to create a more accurate and useful method of building a model, then the thread will have been of some use.
However unless you are a modeller will this discussion have been of any use or have any relevance for the real problems we face? Even if you are a modeller, is this the right time to be discussing how to construct a good model or not?
The idea of modelling our climate today is a complete joke simply because we lack knowledge about how the climate works and about all the variables involved and their relationships with each other and with the known variables. We do not even know all the relationships between the known variables. To pretend to be able to model our climate is fraudulent (hello Met Office).
Dr Evans claims that using PDE with non-independent variables is bad practice and has unpredictable consequences. Such practice is at the core of climate models. Both of these statements are confirmed in publications unrelated to this discussion. Some have been mentioned above.
I'm not convinced by those seeking to defend the climate models. The two statements are relatively straightforward and require straightforward answers.
The argument that the models are not flawed is not supported by the fact that their output is clearly wrong.
Those who defend the models - and I can't tell whether the defence is viable - could help by discussing the possibility of checkable intermediate results. Surely there is some checkable emergent number we can use? Apart from the tropospheric temps, which, if they fail, show the process is flawed. Anything else we can look at?
Any news on how many runs are aborted and on what grounds? Are any final results excluded on a sanity check? Why aren't bad performers kicked out of the ensemble? Is it all because the models are already producing the required results?
Arent all ODE's in fact PDEs? Its just that we do not know it.
Which should make us worried about solving ODEs with any confifdence
the very first ODE, initiated with an apple falling from a tree, described a particle subject to force, but the particle might
react in many different ways to the force apart from its "anticipated one"(acceleration) eg its "spin" might increase , or it might get "hotter" ??
Anyway DEs , an intersting DE is how for example people in numbers dissolve in a manifestation (Corbyn speech):
an easy DE can be written for that with an exponential solution I guess.Based on the numbers at each time.
However, EVERY single "particle" has many other reasons for deciding whetehr to leave, NONE of them anything to do with the speech and/pr number of people ...eg "my feet hurt", "I have to see the mistress", "need to take the dog out beofre dinner" etc etc
The same, i suspect, is the case with EVERY DE.
What I think the modellers are saying is that the PDEs the models solve are expressed in terms of independent variables, and then the solution produced is examined to obtain some calculated metrics like TOA forcing, which also happen to involve partial derivatives and at least in the formula Evans gives are non-independent.
This says nothing at all about the reliability of the climate models. Evans would have to show explicitly what PDE he was talking about and show that the modellers had built it using non-independent variables.
It might imply that TOA forcing isn't a well-defined concept - it depends on what else you hold constant. I've no doubt that in whatever method is used to calculate it, there is a set of variables chosen to be constant, which means that at least operationally the metric is well-defined. (So long as they always calculate it the same way.) But it might mean the written definition is ambiguous, if Evans is giving the same definition. This problem could be solved by adding a footnote.
TOA forcing isn't part of the model. It's like building a model of a car's engine performance, 'driving' it around for a bit, and then dividing the petrol consumed by the distance traveled to get the mpg figure. The mpg for a car is not a well-defined number - it depends how you drive. But this lack of definition doesn't mean the engine performance model is wrong, unless it does something stupid like working out the distance traveled from the fuel consumed by dividing by it.
It's not been shown that the climate modellers did that.
Models are useful for a variety of reasons but their true utility is where they have been validated and are "skillful" as prominent model man Gavin Schmidt is v. fond of telling people...
Trouble is - the level of skillful used to judge the quality of a climate model seems in much (not all) climate science to be the model's ability to agree with policy rather than rigorous validation against observation.
To some extent the internals can be black boxed since if the model's output (prediction) agrees with subsequent observation repeatedly then it can be used with confidence - that is rather the point of models - accurate mimickry of the natural world?
I am fed up with wild climate projections based on policy / ideology models being wheeled out - from a hard science / engineering perspective the lack of validation and the utterly stubborn defense of dysfunctional models simply stinks.
Why do people think it's OK to release apocalyptic predictions based on models that are self evidently broken ??
rhoda
If you're right then it ought to be very simple to calculate climate sensitivity or at least to work out where we are on the scale.
If x° is the amount by which the temperature will rise for a doubling of CO2 then if the temperature was y°C with CO2 concentrations at 280ppm and if the effect of increased CO2 concentration is logarithmic then what will the temperature be with concentrations at 400ppm.
Except that no-one appears to be able to agree on the value of x. And if there are a number of options for this variable then there is no way to say which if any of those models is right. RG Brown has made the point before: if the models are wrong taking an average of their results is totally meaningless. Leo Smith has said the same thing in different words above.
Which means this is a lovely discussion to distract us from Scotland's 39-16 win over the USA but what does it mean in the real world of climate (non-)warming and the determination of the Warmistas and the Green Blob to have their wicked way with the rest of mankind?
It should be noted that the posting which you are discussing from Jo Nova’s site is one of several offered for criticism (preferably constructive), so enabling Mr Evans to test his theories, and to hone them and pursue them yet further, ultimately to the intended conclusion of unequivocally demolishing all the AGW theories. Discussing its mathematics on this site does seem a bit of a waste of time for the participants in the “He’s right / Oh, no he isn’t” discussions that predominate on this thread – go over to Jo Nova, and talk over its merits and demerits, there.
Cheshire red...and NOBODY responds to Harry Huffman. Why is his theory not the subject of intense scrutiny?
I agree absolutely. I went to his site and there is considerable polite discussion and quite a few people complimenting him, who obviously are switched on to the physics. Is Occam's Razor appropriate, but no-one wants to abandon their pet theories?
My own approach is rather basic, is there anything unique about current climate? Historical, Archaeological and Paleological evidence says no. In fact we have been experiencing a relatively benign climate compared to previous periods. So how come, when this trace gas CO2 is supposed to be cooking us?
The whole thing is political, look at the shenanigans at the UN at the moment, with new development goals to take us on from Agenda 21, known as Agenda 2030, complete with Holy Orders from the Pope. How many people even know it is taking place? General public, close to zero.
If they get Son of Kyoto in Paris then we move closer to the World government envisaged by Maurice Strong, with control of global energy via "carbon" taxation operated by the global financiers on behalf of the UN and the promoters all with their snouts in the trough. Whilst so many people are caught up in this, the juggernaut lumbers on. It doesn't matter if you prove them wrong, they don't care, AGW is a means to an end. Wake up and smell the green tea!
http://www.concordeurope.org/publications/item/460-cso-open-letter-to-first-vice-president-frans-timmermans-on-the-2030-agenda
"Dear First Vice-President Timmermans, We, European Civil Society Organisations working on both international and domestic EU policies across a variety of sectors ranging from Youth, Sustainability, Social Justice, Fair Trade, International cooperation, Health, Culture, Environment, Gender Equality, Migration, Climate Change, Local Democracy, Human Rights and Media Development, are writing to you in your capacity as Vice-President of the European Commission, mandated with the horizontal responsibility for sustainable development.
At the UN Summit in New York taking place from 25-27 of September, Heads of State and Government will adopt the universal, people- and planet-centered 2030 Agenda for Sustainable Development (‘Agenda 2030’). As European civil society, we now expect the EU to match the ambition of this Agenda with a clear EU implementation strategy, which reflects the integrated, interlinked and comprehensive nature of the Agenda, in order to ensure well-being for all within planetary boundaries."
Dung: The reason that the bandwagon keeps rolling is
a) The world is definitely going to get warmer anyway in the long run - we're in an interglacial epoch,
b) There has to be some climate sensitivity to increased CO2, even if it's minimal,
c) All the warmists expect to be dead or at least retired by the time the shit hits the fan.
Uncle Gus
I have pointed out on this blog many, many times that if there is such a thing as a 'normal' temperature for the earth then it is about 10 - 15 deg C higher than today and that we are in an interglacial ^.^
There is no sensitivity right now to CO2 because that is what the ice core records tell us.
I will sadly be dead long before most of the warmists ^.^
Uncle Gus,
Dead and richer.
That about sums it up.
ALL climate variables are "dependent" in the warmish sceance, for they are ALL steered by CO2??
Wise words, Uncle Gus. Wise words.
My only dispute is with part C.
If the poop hits the fan they will definitely be retired. There isn't enough money to feed them all without the certainty that their field is very important.
I found that post plausible. A related (alleged) problem is that (supposedly) some model inputs are judgments, rather than measurements. That practice allows the researcher to try different inputs until he gets a result he wants. Two other problem are that the temperature measurements are uncertain and that many of them have been adjusted. Even if the adjustments were unbiased, the process of adjustment would add more uncertainty. If one truly reflected all these sources of uncertainty, the uncertainty of the final model would be very large. It's no wonder that the models have done a poor job of predicting future temperature anomalies.
dennisa 6:59, I think you are being deeply cynical. I can't help but agree.
Politicians, having achieved high office need a vanity project, so they will be remembered with reverence in their planned retirement, for the benefit of their enlarged egos. None of them want to be remembered, as the loser whose name no one can remember, like old what's his face.
They all dream of being SuperMan or SuperWoman, bounding from tree to tree, with a cheery smile for the cameras and their crucial signature on the Treaty that saved the world.
Climate scientists want to be politicians. They are both good at not admitting getting anything wrong. After that it gets worse, for everyone. Prison sentences remain a good route to posterity's rear of shame.
It has to be cheaper to let Evans develop his ideas further. How many times has the Hockey Team ever admitted a mistake? Zero. How many times has the Hockey Team ever got something right? Zero. What is the total measurable change in the climate? Zero
Of course, not only are the models dodgy, the input data are also somewhat questionable...
At Sept 26, 2015, 11:42 pm, Rud Istvan said, "Even a 'right answer' could just be a fluke.”
Which I think accurately describes the two or three 'outliers' in Roy Spencer’s CMIP5 spaghetti graph that are closest to observed temperatures.
Evans is just wasting our time. As others have patiently explained, those of us who do modeling for a living are not as dumb as he presumes. Much of the techniques involved in climate modeling were actually developed by engineers to solve real world problems and they do really well at that. Climate scientists are just johnny-come-latelies attempting the near-impossible by using techniques meant for highly-constrained, small-scale systems to model the entire climate of the entire planet. The very attempt is a complete joke! However often things are done in numerical analysis that a mathematical purist would be horrified at but very often it works. So deriding general numerical modeling, much of which is highly successful, in order to score a purist point is not going to work because it entirely misses the point - in ways that have been adequately explained above by those of us who know about modeling.
The real problem is that the models are not fit for the purpose in so many other ways; much of the real physics of the climate is just unknown, amplification factors are guessed at (even the signs), complex things like aerosols or clouds or the ocean are grossly simplified, natural variation is largely ignored, some equations are used which apply only to limited situations, outputs are not verified properly and merely reflect the inputs given - which are themselves unverified, the grid size is too crude and obvious numerical instability due to overly large time-steps is laughably assumed to be useful output, convergence is not achieved but rather the solution is just cut short to get an answer, the input error range demands sensitivity analysis which is rarely done, and so many more......... .And then they take these crap model runs and combine them. There is no mathematical or experiential basis for combining 20 crap models and assuming the combination will be somehow better. Proven & tested good models yes can be combined - but not crap models! And then they assess these models with pseudo-Bayesian inputs and data-mined output and subject them to frequentist stats. There are enough real horror stories in the modeling efforts to supercede the irrelevant, minor issue Evans brings up.
Now environmental models are known to be extremely unreliable - often giving the exactly wrong answer. This is known. Also any prediction of the future is well known to be mere unreliable guesswork. Yet climate models get a free ride on these issues where other modelers would be laughed at. Why is this? Apparently due to misplaced morals, zeitgeist, angst, unbalanced funding and sheer downright dishonesty! And like topsy it grew and grew until no dissent was tolerated and no scientist can stick their head above the parapet without it being shot off. There have been other science issues where this happened in the past and continues to happen. Alas none of them affected the communal energy policy that is responsible for making the West affluent and would do the same for the developing world if left alone and none of them became such a battleground for anti-progress progressives versus un-conservative conservatives.
To me there is only truth, which is not ideological but testable. And man-made warming fails all tests. Every single one! If we left our failed ideologies at the doorstep this would be accepted as just another failed enviro-apocalypse prediction along with all the others. Alas we live in a world where these shysters get huge prizes and media kudos despite being proven wrong all the time - apparently because their hearts are in the right place. And skeptics - well we are just beneath the feet of the sanctimonious ignorami that constitute the chattering classes and it doesn't matter how often we are proved right because we just don't share the same blind faith as they do.
@Paul_K & JamesG
It would be far more useful, efficient even, if you left your comments on the actual website thread where Evans is writing, so he has a chance to reply:
http://joannenova.com.au/2015/09/new-science-5-error-2-model-architecture-means-all-feedbacks-work-through-the-surface-temperature/#comments
As it stands, your comments are really just shadow boxing, whether they have substance or not
> These things can make a big difference to the (relatively very small) changes in energy flows in the system [...]
Of course they could. Anything can make a big difference in this supremely chaotic world. The question is if they (assuming we know which) do.
Appealing to ignorance is a common ClimateBall move. Lord Russell might frown upon it. In fact, we can suspect he might have thrown a tea pot toward those who use that move.
PS: When will the recent GWPF announcement be announced at their PA's blog?
I see that Lord Russell's squirrel has appeared. However, there is no appeal to ignorance. Russell's squirrel could find the large effect of turbulence modeling on fluid flows in the literature as there are at least 100,000 papers on it over the last 50 years. This is all really so well known it is embarrassing that the arboreal rodent population remains in the dark.
> Russell's squirrel could find the large effect of turbulence modeling on fluid flows in the literature as there are at least 100,000 papers on it over the last 50 years.
Finding a large effect does not a difference make until a difference is found and shown. Until then, an appeal to ignorance it remains. We enter Russell's tea pot territory when this appeal to ignorance is used within the lukewarm playbook.
***
Using fighting words like "ClimateBaller" is incidentally a ClimateBall move, and comes from the fiercest player in its history:
https://andthentheresphysics.wordpress.com/2014/09/29/the-ghost-of-present-climateball-tm/
This kind of move, along with jabs "powerful green political machine," contrast with the scientific purity in which DY wraps himself.
Anybody care to precis that exchange for someone whose language is limited to English?
Lord Russell' squirrel,
Even the esteemed Dr. Rice would admit that tropical convection and cloud sub grid models DO make a difference. These are also things i mentioned above. In fact, it seems very obvious that something is probably wrong in the tropics with models as they seem to not match the data very well. Even the Trenberth paper highlighted recently by Rice shows exactly this, that these things can make a big difference!! It is so obvious, one marvels at arboreal rodent obfuscation and chirping around the edges about it.
The fact of your ignorance does not mean i am appealing to that ignorance. Subgrid models are hard enough even for real scientists, idle Climateball referees haven't got much of a chance.
> [T]ropical convection and cloud sub grid models DO make a difference.
This adds equivocation to your ClimateBall arsenal, DY.
To see how, try to finish your sentence. Suppose you have access to near-perfect convection and cloud models: what difference would this difference make? If all you can add is that we ought to have better modulz, then your concerns (about which I am neverendingly thankful) get dissolved into triviality.
Between non-linearly praying the Chaos gods and stretching down the limits of lukewarm disingenuousness, you need to choose. You can't have it both ways. That would amount to a Dutch book, and the Tea Pot in Outer Space knows how Lord Russell would disapprove.
AATP and Eli are being their usual rude selves.
It would be desirable for them to point out why Dr Evans has something incorrect instead of calling him pejorative terms like "Confused".
It would also be desirable for them to comment directly rather than here (posting comments of substance here is not as helpful as posting direct on the originating site).
Sep 26, 2015 at 5:16 PM | Unregistered Commenter Harry Dale Huffman
Very interesting stuff and should be more widely discussed. Why not send a reminder on to Anthony Watts and Tony Heller?
I think I've never heard so loud,
The Gods are laughing from the clouds.
===================
Eli Rabbett: Read the posts before you comment, lest you look “confused” eh?
Oh yes I did. Those are the partial derivatives in post 3, which describes the conventional basic climate model, “the basic physics” of climate change according to the establishment, and the real reason most establishment climate scientists believe in the CO2 theory — see post 1.
You bring up Navier-Stokes, which was not mentioned in my posts and is inappropriate at the basic level of climate model. You are clueless.
Eli, your link is irrelevant and already answered by the link I provided in the post. In any case, the limitations and problems with the partial derivatives are clearly and adequately described by my post alone.
If you want to discuss it, come over to the post itself -- I won't be answering everywhere on the web.
Thanks to Rud, Schrodingers Cat, radical rodent and others for putting in some sane words.
And Then There's Physics — Please read post 4 of the series, which is on partial derivatives, and put it in the context of the conventional basic climate model (post 3), noting that the real reason most establishment climate scientists believe in the CO2 theory is because of the basic model (“basic physics”) — post 1.
There is no confusion on my part. I recommend you also follow the MIT link in my post to answer your simplistic query.
As for the feedbacks, you would know if you had read posts 1 or 3 that I am using and accept the values given in AR5 (which come from the CMIP5 models).
Provided you actually read the posts in question, you are welcome to comment underneath them :)
David Young - Since we're giving each other advice, may I recommend you read the series of posts first, especially if you think Eli Rabbett knows what he is talking about. :)
You are welcome to comment on them there (under the correct post, of course) :) I am not able to respond to all queries on other sites.
Paul_K. “AT a glance” – you need to do than “glance” It is not I who am confused, but you from reading the confused and confusing remarks here. Just read the post itself, and while you are at it, read the whole series :) Your criticism is irrelevant and off-beam.
Curious, you can't be bothered to read the original, but can be bothered writing a 670 word comment on it here. Who is a seeker of truth?