I’ve been looking at the Nikolov and Zeller paper again. Among other things, they claim to be able to calculate the surface temperature Ts of eight different planets and moons from knowing nothing more than the solar irradiation So and the surface pressure Ps for each heavenly body. Dr. Zeller refers to this as their MIRACLE equation. He says:
Why aren’t you all trying to disprove our MIRACLE equation rather than banging your heads against walls trying to prove or disprove who knows what and exclaiming you have problems with this or that? The question is how can we possibly have done it – there is no question that our equations work – if you haven’t verified that it works, why haven’t you? […] Why aren’t you thinking: “hmmmm, N&Z have given us an equation that lo-and-behold when we plug in the measured pressures and calculate Tgb as they suggest, gives us a calculated Ts that also matches measured values! You can’t disprove the equation? So maybe we are cooking the data books somehow, but how?
This is supposed to be evidence that their theory is correct, and people keep telling me ‘but they’ve got real evidence, they can make predictions of planetary temperatures, check it out”. Plus it’s hard to ignore an invitation like Dr. Zellers, so I checked it out.
Figure 1. These are not the equations you are looking for.
They first postulate something called the “Near-surface Atmospheric Thermal Enhancement” or “ATE” effect that makes the earth warmer than it would be without an atmosphere.
The “ATE effect” is measured by something called Nte(Ps), which is defined and estimated in their paper as follows.
where Nte(Ps) is a measure of the “Near-surface Atmospheric Thermal Enhancement” effect.
Nte(Ps) is defined as the actual average surface air temperature of the planet Ts divided by the theoretical “graybody” temperature of the planet Tgb calculated from the total solar insolation So of the planet. Nte(Ps) is estimated using a fitted function of the surface pressure of the planet Ps.
Let me simplify things a bit. Symbolically, the right part of equation (7) can be written as
Nte(Ps) = e^(t1 * Ps ^ t2 + t3 * Ps ^ t4) (7Sym)
where “e” is the base of natural logs and Ps is the surface pressure on the planet or moon. There are four tunable parameters (t1 through t4) that are “fitted” or tuned to the data. In other words, those values are repeatedly adjusted and tuned until the desired fit is obtained. This fitting can be easily done in Excel using the “Solve…” menu item. As you’d expect with four parameters and only eight datapoints, the fit is quite good, and their estimate is quite close to the actual value of Nte(Ps).
Amusingly, the result of equation (7) is then used in another fitted (tuned) equation, number (8). This is:
where So is total solar irradiation.
This is their piece de resistance, their MIRACLE equation, wherein they are saying the surface temperature of eight different planets and moons can be calculated from just two variables— Pr, the surface pressure, and So, the total Solar irradiation. This is what amazes the folks in the crowd so much that they write and tell me there is “evidence” that N&Z are right.
Obviously, there is another tuned parameter in equation (8), so we can rewrite this one symbolically as:
Ts = t5 * (Solar + adjustment ) ^ 1/4 * Nte(Ps). (8Sym)
Let me pause a minute and point something out about equation (8). The total solar irradiation Solar ranges from over 9,000 W/m2 for Mercury down to 1.51 W/m2 for Triton. Look at equation 8. How will adding the adjustment = 0.0001325 to any of those values before taking the fourth root make the slightest bit of difference in the result? That’s just bizarre, that is. They say they put it in so that the formula will be accurate when there is no solar, so it will give the background radiation of 3 Kelvins. Who cares? Truly, it changes Ts by a maximum of a thousandth of a degree for Triton. So for the moment let me remove it, as it makes no practical difference and it’s just confusing things.
Back to the tale. Removing the adjustment and substituting equation 7 into equation 8 we get:
Ts = t5 * Solar^0.25 * e^(t1 * Ps ^ t2 + t3 * Ps ^ t4) (eqn 9)
This is amazing. These guys are seriously claiming that with only eight datapoints and no less than five tunable parameters , they can calculate the surface temperature of the eight planets knowing only their surface pressure and solar irradiation. And with that many knobs to turn, I am sure they can do that. I did it on my own spreadsheet using their figures. I get about the same values for t1 through t5. But that proves nothing at all.
I mean … I can only stand in awe at the sheer effrontery of that claim. They are using only eight datapoints and five tunable parameters with a specially-designed ad-hoc equation with no physical basis. And they don’t think that’s odd in the slightest.
I will return to this question of the number of parameters in a bit, because even though it’s gobsmacking what they’ve done there, it’s not the best part of the story. Here’s the sting in the tale. We can also substitute equation (7) into equation (8) in a slightly different way, using the middle term in equation 7. This yields:
Ts = t5 * Solar^0.25 * Ts / Tgb (eqn 10)
This means that if we start out by knowing the surface temperature Ts on the right side of the equation, we can then calculate Ts on the left side … shocking, I know, who would have guessed. Let’s check the rest of the math in equation (10) to see why that works out.
Upon inspection it can be seen that the first part of the right side of equation (10),
t5 * Solar^0.25
is an alternate form of the familiar Stefan-Boltzmann equation relating temperature and radiation. The S-B equation can be written as
T = (Solar / c1) ^ 0.25.
where T is temperature and c1 is a constant equal to the S-B constant times the emissivity. We can rewrite this as
T = 1/(c1^0.25) * Solar^0.25
Setting another constant c2 equal to 1 / (c1^0.25) gives me the Stefan-Boltzmann equation as:
T = c2 * Solar^0.25
But this is exactly the form of the first part of the right side of equation 10. More to the point, it is an approximation of the graybody temperature of the planet Tgb.
We can check this by observing that if emissivity is .9 then constant c1 is 5.103E-8, and c2 is therefore about 66. However, that value will be reduced by the rotation of the planet. Per the N&Z formula in their latest post, that gives a value of about 27.
Their fitted value is 25, not far from the actual value. So curiously, what it turns out they’ve done is to estimate the Stefan-Boltzmann constant by a bizarre curve fitting method. And they did a decent job of that. Actually, pretty impressive considering the number of steps and parameters involved.
But since t5 * Solar^0.25 is an estimation of the graybody temperature of the planet Tgb, that means that Equation 10 reduces from
Ts = t5 * Solar^0.25 * Ts / Tgb (eqn 10)
to
Ts = Tgb * Ts / Tgb.
and finally to
Ts = Ts
TA-DA!
CONCLUSION
Let me recap the underlying effect of what they have done. They are looking at eight planets and moons.
1. They have used an equation
e^(t1 * Ps ^ t2 + t3 * Ps ^ t4)
with four free parameters to yield an estimate of Ts/Tgb based on surface pressure. As one would expect given the fact that there are half as many free parameters as there are data points, and that they are given free choice to pick any form for their equation without limit, this presents no problem at all, and can be done with virtually any dataset.
2. They have used an equation
t5 * Solar^0.25
with one free parameter in order to put together an estimate of Tgb based on total planetary insolation. Since Tgb does depend inter alia on planetary insolation, again this presents no problem.
3. They have multiplied the two estimates together. Since the result is an estimate of Tgb times an estimate of Ts/Tgb, of course this has the effect of cancelling out Tgb.
4. They note that what remains is Ts, and they declare a MIRACLE.
Look, guys … predicting Ts when you start out with Ts? Not all that hard, and with five free parameters and a choice of any equation no matter how non-physically based, that is no MIRACLE of any kind, just another case of rampant curve fitting …
Finally, there is a famous story in science about this kind of pseudo-scientific use of parameters and equations, told by Freeman Dyson:
We began by calculating meson–proton scattering, using a theory of the strong forces known as pseudoscalar meson theory. By the spring of 1953, after heroic efforts, we had plotted theoretical graphs of meson–proton scattering. We joyfully observed that our calculated numbers agreed pretty well with Fermi’s measured numbers. So I made an appointment to meet with Fermi and show him our results. Proudly, I rode the Greyhound bus from Ithaca to Chicago with a package of our theoretical graphs to show to Fermi.
When I arrived in Fermi’s office, I handed the graphs to Fermi, but he hardly glanced at them. He invited me to sit down, and asked me in a friendly way about the health of my wife and our newborn baby son, now fifty years old. Then he delivered his verdict in a quiet, even voice. “There are two ways of doing calculations in theoretical physics”, he said. “One way, and this is the way I prefer, is to have a clear physical picture of the process that you are calculating. The other way is to have a precise and self-consistent mathematical formalism. You have neither.
I was slightly stunned, but ventured to ask him why he did not consider the pseudoscalar meson theory to be a selfconsistent mathematical formalism. He replied, “Quantum electrodynamics is a good theory because the forces are weak, and when the formalism is ambiguous we have a clear physical picture to guide us. With the pseudoscalar meson theory there is no physical picture, and the forces are so strong that nothing converges. To reach your calculated results, you had to introduce arbitrary cut-off procedures that are not based either on solid physics or on solid mathematics.”
In desperation I asked Fermi whether he was not impressed by the agreement between our calculated numbers and his measured numbers. He replied, “How many arbitrary parameters did you use for your calculations?”
I thought for a moment about our cut-off procedures and said, “Four.”
He said, “I remember my friend Johnny von Neumann used to say, with four parameters I can fit an elephant, and with five I can make him wiggle his trunk.” With that, the conversation was over. I thanked Fermi for his time and trouble, and sadly took the next bus back to Ithaca to tell the bad news to the students.
The Nikolov and Zeller equation contains five parameters and only eight data points. I rest my case that it is not a MIRACLE that they can make the elephant wiggle his trunk, but an expected and trivial result of their faulty procedures.
My regards to everyone,
w.
PS—There is, of course, a technical term for what they have done, as there are no new mistakes under the sun. It is called “overfitting”. As Wikipedia says, “Overfitting generally occurs when a model is excessively complex, such as having too many parameters relative to the number of observations.” Five parameters is far, far too many relative to eight observations, that is a guaranteed overfit.
PPS—One problem with N&Z’s MIRACLE equation is that they have not statistically tested it in any way.
One way to see if their fit is even remotely valid is to leave out some of the datapoints and fit it again. Of course with only eight datapoints to start with, this is problematic … but in any case if the fitted parameters come out radically different when you do that, this casts a lot of doubt on your fit. I encourage N&Z to do this and report back on their results. I’d do it, but they don’t believe me, so what’s the point?
Aother way to check their fit is to divide the dataset in half, do the fit on one half, and then check the results on the other half. This is because fitted equations like they are using are known to perform very poorly “out of sample”, that is to say on data not used to fit the parameters. Given only eight data points and four parameters for equation 7, of course this is again problematic, since if you divide the set in half you end up with as many parameters as data points … you’d think that might be a clue that the procedure is sketchy but what do I know, I was born yesterday. In any case I encourage N&Z to perform that test as well. My results from that test say that their fit is meaningless, but perhaps their test results will be different.
[UPDATE] One of the commenters below said:
Willis – go ahead – fit an elephant. Please!
Seriously N&Z are only demonstrating in algebra what has been observed in experiments, that heating a gas in a sealed container increases both pressure and temperature.
OK, here’s my shot at emulating the surface temperature using nothing but the data in the N&Z chart of planetary body properties:
Figure 1. Willis’s emulation of the surface temperature of the planetary bodies.
My equation contains one more variable and two less parameters than the N&Z equation. Remember their equation was:
Ts = 25.3966 * Solar^0.25 * e^(0.233001 * Pressure ^ 0.0651203 + 0.0015393 * Pressure ^ 0.385232)
My equation, on the other hand, is:
Ts = 0.8 * Tgb + 6.9 * Density + 0.2 * Gravity)
Note that I am absolutely not making any claim that temperature is determined by density and gravity. I am merely showing that fitting a few points with a few variables and a few parameters is not all that difficult. It also shows that one can get the answer without using surface pressure at all. Finally, it shows that neither my emulation nor N&Z’s emulation of the planetary temperatures are worth a bucket of warm spit …
[UPDATE 2] I figured that since I was doing miracles with the N&Z miracle equation, I shouldn’t stop there. I should see if I could beat them at their own game, and make a simpler miracle. Once again, their equation:
Ts = 25.3966 * Solar^0.25 * e^(0.233001 * Pressure ^ 0.0651203 + 0.0015393 * Pressure ^ 0.385232)
My simplified version of their equation looks like this:
Ts = 25.394 * Solar^0.25 * e^(0.092 * Pressure ^ 0.17)
Curiously, my simplified version actually has a slightly lower RMS error than the N&Z version, so I did indeed beat them at their own game. My equation is not only simpler, it is more accurate. They’re free to use my simplified miracle equation, no royalties necessary. Here are the fits:
Figure 2. A simpler version of the N&Z equation 8
Again, I make no claim that this improves things. The mere fact that I can do it with two less tuned parameters (three instead of five) than N&Z used does not suddenly mean that it is not overfitted.
Both the simplified and the complex version of the N&Z equations are nothing but curve fitting. This is proven by the fact that we already have three simple and very different equations that hindcast the planetary temperatures. That’s the beauty of a fitted equation, if you are clever you can fit a lot using only a little … but THAT DOESN’T MEAN THAT PRESSURE DETERMINES TEMPERATURE.
For example, I can do the same thing without using pressure at all, but using density instead. Here’s that equation:
Ts = 25.491 * Solar^0.25 * e^(0.603 * Density ^ 0.201)
And here’s the results:
Figure 3. An emulation of the planetary temperatures, using density instead of pressure.
Does this now mean that the planetary temperature is really controlled by density? Of course not, this whole thing is an exercise in curve fitting.
w.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
> I really think Willis comments need moderation.
Agreed.
Willis, among your many eccentric talents, you are pretty good at heaping out verbal scorn and abuse. In the case of Stoat, the blast he got from you was well deserved. I’ll have to give you an ‘Attaboy!’ for that one.
But in the case of N&Z, your words have directly and indirectly brought down a ton of scorn and derision on their work which is completely undeserved. You owe them a big apology.
Their thesis is not at all about curve-fitting data to 8 points. You’ve got it backwards. In those 8 planetoids they noticed a remarkable correlation between pressure and temperature. Using those 8 planets as a ‘training set’ they were able to come up with an empirical power law which gave a ‘good fit’ between pressure and temperature.
But that was just a preliminary step in their research. Such empirical power laws don’t explain why they work. Zipf’s Law predicts word frequencies, city sizes, personal incomes etc but even Zipf couldn’t come up with a good reason why it worked so well. All he could say, waving his hands, was it due to some ‘principle of least effort’.
N&Z have done better than Zipf. They have started a new theory, based in classical physics, which explains _why_ you only need to know pressure to determine temperature.
Does the pressure somehow ‘create’ the temperature? Of course not. It’s plain-old-physics that’s been around for a century: the Ideal Gas Law.
Think of it as a ‘dual solution’ in the abstract problem space. The theory doesn’t add any new laws or change any of the old laws of physics. It just ‘inverts’ the way we look at them such that radiation it put in the background and pressure is brought out as the ‘explanatory mechanism’.
So the Sun is still pouring out radiated energy, and the planets still absorb that energy as heat. Mostly the surface, but GHG’s are still absorbing too (as much as they always have). What’s different is that part of the physics is all subsumed as an assumption in the Ideal Gas Law.
As for curve fitting, you (and others) are totally mistaken in the idea that you can claim the sin of ‘overfitting’ by merely counting the numbers of parameters in the model.
In the first place, the 8 planets in question are merely the ‘training set’ used to train the model. Obviously the ‘test set’ comprises all of the other planets in the Universe, which I don’t have a number for, but I don’t think 5 or 6 parameters would be considered ‘overkill’ for that job.
In the second place ‘overfitting’ is not always bad. When you use trig functions on your computer you are invoking polynomial models because there are no ‘closed form’ solutions for these functions. So the models are carefully crafted to fit the expected values of the function over as many ‘useful’ ranges as possible. ‘Overfitting’ (implying perfect match) would be welcome here.
Overfitting can be bad when you train models using subsets of the data you want to model but not always. Most of the time you can get a good fit on the training set. Depending on the model used, the results on the test set can diverge greatly from the expected values. Polynomial fits (without regularization) are notorious for this. Also, outliers in the training set can throw the models off. But not always, support vector machines are surprisingly robust in this respect. You can ‘overfit’ the SVM’s training data all you want, and test sets still fit good too. (But they don’t explain anything about the features used. Bummer.)
So we need to stay tuned to the N&Z Show, Part II. Keep our scornful thoughts down to a minimum and learn some new ideas about old planetary physics.
Ned and Karl, keep up the good work! And keep those papers coming!
😐
I might be an idiot (please do point out if so) but to me it occurs that the atmospheric pressure has a lot to do with how heavy the celestial body is and how fast it rotates on its own axis?
venus has a thick atmosphere because it hardly rotates (243 earth days for 1 venus rotation)
mars has no atmosphere and it rotates like earth but has 1/10th the mass.
the physical explanation is the atmospheric molecules stick around due to gravitation , but might be spun off if spinning too fast. lol.
as we cannot change mass and rotation speed of earth (unless we build too many wimmills which will make earth lose its kinetic energy) earth atmosphere will remain at , say, 1 bar.
So, for every CO2 molecule that goes up in the air one has to drop down (or we would be violating an important rule here)
scf (Jan. 24, 2012 at 3:45 pm):
Further to my remarks of Jan. 24, 2012 at 9:21 PM, as the IPCC’s models are not testable, the IPCC’s inquiry into anthropogenic global warming cannot have had a scientific methodology but the IPCC has represented it as having had a scientific methodology. This is a really colossal error in the construction of the IPCC’s inquiry but few bloggers or climatologists exhibit an awareness of this error. If we could get people thinking in terms of statistical populations and samples rather than the nebulous term “datapoints,” we could turn this situation around!
What the h*** is the point of this post???
So you prove that Ps = Ps and that makes it wrong?
This post is worthy of RC, orwelian doublespeak at its best.
So you got better results by using eq. Ts = 0.8 * Tgb + 6.9 * Density + 0.2 * Gravity
Ok, now show us how you can just insert the expressions for the variables and get the result Ts = Ts. If you cant, your nothing but a POS with a personal agenda.
LucVC: “Take the Conolley comment. Even though he erased many of my contributions he’s not beyond repair as noboddy is.”
William Connolley is a political crusader Luc. Truth and science have no meaning for him except as potential vehicles for advancing a political agenda. Where he can use science for that purpose he will do so. Where the science goes against him he will suppress it. Nothing he does here will ever change that. In this Willis post he saw an opportunity to cheer lead something that he felt was good for his agenda, and that is really all that his comment consisted of. Willis was right to distance himself from Connolley’s intentions, even if he did it a little more emphatically than necessary.
Willis Eschenbach says:
January 25, 2012 at 2:09 am
I have added an additional update, showing that I can beat their emulation of the planetary temperatures (lower RMS error) using only three parameters instead of five … take a look. I’ve done it with density instead of pressure as well.
Harry Dale Huffman beat you to it. Anyway, show us your working so we can see how you derived density without using pressure. Two sides of the same coin.
DRE says:
January 23, 2012 at 4:50 pm
“The right answer for the wrong reason is ALWAYS the wrong answer. (Used to teach cal. based Physics)”
This entire exercise reminds me of the oposite of the above which I experienced, ie getting the wrong answer for the right reason. In dynamic physics sophomore year where the problem involved James Bond shooting a bullet and calculation of the terminal effects, I got an answer of .006 jules or whatever. The professor took time to note that though he gave me almost full credit for the wrong answer since my error was simply in rounding, he should have given me no credit for my stupidity in not realizing the answer was supposed to be.007.
Willis Eschenbach says:
January 25, 2012 at 2:09 am
Very nice, Willis. However, you did not create an elephant as requested. 😉
What you did is confirm the relationship may be even stronger than N&Z have found. I don’t think that was your intent. It still may be junk but it would take investigations of the values you used to determine that as a fact.
Did you read the Sorokhtin paper?
http://translate.google.com/translate?sl=ru&tl=en&js=n&prev=_t&hl=en&ie=UTF-8&layout=2&eotf=1&u=http%3A%2F%2Ffiz.1september.ru%2Farticlef.php%3FID%3D200501111
It’s a little beyond my abilities but it appears he uses real physical values in comparing Earth and Venus (which are probably the only two planets worth comparing).
@Willis
> I can beat their emulation … using only three parameters instead of five …
There are at least a zillion planets (maybe more) in the Universe. Does your 3-parameter model work better for those too? Do you understand why your parameters work as well as they do in terms of real-world physics?
If not, then you are just “gaming the system”.
When are you going discuss the science behind N&Z, instead of playing with models (something we “skeptics” normally scorn)?
@Willis
> Please confine yourselves to the topic of the thread …
Oops sorry, I forgot that is against the rules you have arbitrarily set up for this
gamepost.😐
Willis says …. “they claim to be able to calculate the surface temperature Ts of eight different planets and moons from knowing nothing more than the solar irradiation So and the surface pressure Ps for each heavenly body. Dr. Zeller refers to this as their MIRACLE equation…” …..”My simplified version of their equation looks like this: Ts = 25.394 * Solar^0.25 * e^(0.092 * Pressure ^ 0.17)” = A SIMPLER MIRACLE
You folks just don’t get it do you, you’re not seeing the forest for the trees: Willis’ rendition of our MIRACLE is also a MIRACLE!!!!!!! What is the Miracle you don’t see? We calculate the average global equilibrium surface temperature on any planet/moon using only Solar input and surface pressure! Why is this a miracle? Because it implys that the AGW theory is bogus. Why does it do that? Since the average global surface temperature of any planet/moon IS the basic bottomline determinator of that planet/moon’s climate and our Eq 8 accurately calculates this temperature without using greenhouse gas information.
Dr. Nikolov & I have been working on this for over 2 years, our first attempts looked like Willis’ simpler miracle, and we’ve played with density also, but we are trying to get it exact, currently Eq 8. You CAN NOT fit an elephant with an exponential equation, you can with a polynominal. The argument about the number of constants in our equation 8 would be valid it it were a polynominal – it’s not.
We are handing WUWT ‘THE NAIL’ to the AGW coffin and you guys have forgotten about the coffin and are fixated on the details of the nail! Is it galvinized? Why isn’t it a wooden spike? They need 2 more nails. Wonder what kind of hammer they plan to use?
AusieDan: “Carbon dioxide, by its molecular weight and atomic structure, expands more rapidly under heat than does an average sample of air. This causes increased pressure in the sealed container holding the carbon dioxide sample, relative to the container holding just air. Increased pressure results in the temperature of the carbon dioxide to rise higher than the air in the other container. That is the so called greenhouse effect.
HOWEVER, when the pressure in the two vessels are allowed to equate, such as by providing a hole in the top for the excess gas to escape to the outside air, then the temperature in each vessel rises to a lower level than before and the temperature in both containers remains the same. Bye Bye, greehouse effect. This has also been demonstrated on a number of occasions, but with far less noise and propaganda than the false greehouse effect.”
Dan, I had read this earlier and had meant to get back to it. This is new information for me, and I’d like to know a little more about it. So basically, we have been arguing about climate sensitivity from the perspective of about 1C of warming per CO2 doubling, plus some undefined feedback. The 1C of warming for 2 X CO2 has been regarded as a given, due to the lab experiments with enclosed CO2. Now, if I understand your argument correctly, you are saying that the 1C is also false, and that the extra CO2 in the atmosphere will simply cause atmospheric expansion, not temperature increase. Do you have a link where I can learn more about the work you are refering to?
Dr. Zeller: “You folks just don’t get it do you, you’re not seeing the forest for the trees: ”
LOL. I’m very sympathetic with your frustration here. My thought is that people are picking on your method, “galvanized nails or wooden spikes”, because the underlying physics isn’t clear to them. Hopefully your part 2 explanation will help with that. Also, remember that skeptics to AGW really are skeptics that will try to pick apart everything, even in cases where it is true and to their liking. Think of it as process rather than final conclusion. I think your ideas are basically sound, so please don’t be despirited because you have run into a dustup. In most cases in the past when scientists have produced new ideas the rest of the scientific community didn’t simply throw up their hands, acknowledge the brilliance, and embrace the change. Much more often there was a fight with the people who’s ideas were being displaced and with those that had come to accept those ideas. So buckle your chin strap and explain, explain, explain – then I think you will prevail.
I agree with Karl!
The major point that Willis totally missed (but follows from his own regression analysis) is that the planetary surface temperature CAN be explained in a straightforward fashion over a broad range of atmospheric conditions using ONLY pressure and solar irradiance . No other model can currently do that! This is the main point of our theory! The grand implications of the strong relationship shown in Fig. 5 (the NTE – Pressure curve) is that the so-called GH effect (we call it ATE) is a pressure phenomenon that is completely unrelated to radiative transfer! Hence, the GH effect in its physical nature is NOT a reduction of the surface IR cooling to space as currently assumed, but is a Pressure-induced Thermal Enhancement (PTE) … It is that simple!
kzeller says: “We are handing WUWT ‘THE NAIL’ to the AGW coffin and you guys have forgotten about the coffin and are fixated on the details of the nail! Is it galvinized? Why isn’t it a wooden spike?”
Great analogy, Dr. Zeller. A wooden spike to kill the AGW vampire that wants to suck the blood out of the US.
Our official response to Willis’ ‘analysis’ of Eq. 8 will be coming very shortly….
We through we should not engage in correcting high-school math errors such as those made by Willis above, but it looks like we have to it at least once to set the record straight, so we do not have to degrade this discussion anymore in the future … Stay tuned!
KZ> “Why is this a miracle? Because it implys that the AGW theory is bogus”
Ah, you mean that only supernatural intervention could render the well-established greenhouse effect physics wrong. At last, something we can agree on.
John Day says: “@Willis > ‘Please confine yourselves to the topic of the thread …’
Oops sorry, I forgot that is against the rules you have arbitrarily set up for this game post.”
Since the objective was to focus on the two equations in question without going off on a million tangents, I think the rules are less than arbitrary. As a result of them, we only have to wade through half a million rants, maunderings, troll castings, nit-pickings, and irrelevancies.
kzeller says:
January 25, 2012 at 9:51 am
We are handing WUWT ‘THE NAIL’ to the AGW coffin and you guys have forgotten about the coffin and are fixated on the details of the nail! Is it galvinized? Why isn’t it a wooden spike? They need 2 more nails. Wonder what kind of hammer they plan to use?
Dr. Zeller, you guys need to consider that you’re presenting your ideas to a bunch of skeptics. You should use that fact to hone your own views. Yes, a lot of comments will be useless but that does not mean there isn’t good information available. And, if you can’t argue your points here what do you think will happen when you present them to AGW believers?
Personally, I think you are wrong when you believe you can replace the GHE with gravity. And, as I’ve said before, I think the relationship you found in the planets might be very important. That’s where you should try and improve your paper. You need to highlight the physical factors that lead to the miracle equation in more detail.
Don’t get discouraged.
William M. Connolley:
There is no ‘well established’ GHE physics! The whole notion about the GHE being a radiative phenomenon is based on a confusion that down-welling IR radiation warms the surface on a global scale. This confusion in turn comes from the fact that radiative transfer is being solved in climate models decoupled from convection. Once these two heat transport mechanisms are robustly coupled (i.e. solved simultaneously as a part of the same system of equations) as it happens in the real atmosphere, the convective cooling completely offsets the warming effect of ‘back’ radiation. This can be easily demonstrated mathematically.
What needs be realized here is that the long-wave ratiadive transfer in the atmosphere is only a RESULT (a BYPRODUCT if you will) of temperature, NOT a cause for it! Temperature, on the other hand, is a function of solar heating and pressure!
The situation with the current GH theory is one of confusion between cause and effect. It resembles quite the paradigm we had in medieval times prior to Copernicus, when it was believed that the Earth was at the center of the Universe based on the superficial observation that Heavens revolve around Earth. Similarly, the fact that we measure a significant LW radiation coming down from the atmosphere, it does not mean that that flux is actually responsible for the atmospheric thermal effect!
As we have shown in our Reply Part 1 through analysis of Moon temperature data here:
http://wattsupwiththat.com/2012/01/22/unified-theory-of-climate-reply-to-comments/
the actual magnitude of the atmospheric GH effect is about 133K. It requires many thousands of Watts of additional radiation (about 14,800 W m-2 to be exact) to raise the temperature of Earth from the gray-body value of about 155K to the present temperature of 287.6K. The observed average downward LW flux is only about 343 W m-2. So, even from the standpoint of this piece of evidence, it is clear that the GH effect CANNOT be possibly caused ‘back radiation’. Our interplanetary analysis explains the GH effect completely through pressure … What can be more clear than that?
Mr Lynn says:
January 25, 2012 at 5:18 am
LucVC says:
January 25, 2012 at 1:59 am
One of the hallmarks of WUWT has been a policy of eschewing ad hominem attacks on other commenters. From Anthony’s Policy page:
Yeah, you’re likely right, Mr. Lynn, and I’m working on it. However, I don’t take well to being insulted by such a charming, intelligent, witty fellow as Mr. Connolley. He and I have a bit of history.
w.
Tilo Reber:
God…Now, THAT’S funny!!!!! Thanks for giving me my daily belly-laugh!
Richard M says:
No…He isn’t. He is presenting it to people who call themselves “skeptics”. There is clearly a huge difference as illustrated by the fact that many people here are still taking Nikolov & Zeller’s work seriously.
John Day says:
Yes…Willis and my point exactly. The difference is that he is telling you outright that he is gaming the system whereas Nikolov and Zeller are not.
The science has already been discussed…and shredded! The “theory” does not even obey the 1st Law of Thermodynamics…I.e., it violates conservation of energy!
The authors claim that convection makes the radiative greenhouse effect has been demonstrated to be due to an extreme error in how they added convection.
The fact that average temperatures on various planets are lower than the conventionally-calculated blackbody temperature has been explained as a well-understood consequence of Holder’s Inequality applied to non-uniform temperature distributions.
What more do you want?
djorgekafkazar says: Great analogy, Dr. Zeller. A wooden spike to kill the AGW vampire that wants to suck the blood out of the US.
Congratulations dj… , you’ve read the blog words, found & tasted some candy, now read the original post and part 1 for the hidden dishes making up our Sublime Smorgasbord. 🙂
John Day says:
January 23, 2012 at 9:25 pm
@Willis
> >There was an option to say “Sorry I accused you of
> >calling N&Z dishonest when you hadn’t”.
>Read my remarks carefully. I didn’t “accuse you of calling N&Z dishonest”. You’re blowing my >statements out of proportion. I said it suggested dishonesty.
I went back and reread your original comment wherein you make the “dishonesty” statement.
I guess it depends on what your definition of “is” is.
@jorge
> Since the objective was to focus on the two equations in question
> without going off on a million tangents, I think the rules are less
> than arbitrary. As a result of them, we only have to wade through
> half a million rants, maunderings, troll castings, nit-pickings,
> and irrelevancies.
With respect, I disagree with that approach. All it does is encourage the meanderings and useless “dancing around the fire”.
The addition of planetary physics to the discussion would have, IMHO, a focusing effect to gauge and filter out the chaff and gaming.
One other piece of advice, to all: “Don’t feed the trolls!”
😐
> He and I have a bit of history.
Do we? I’d forgotten you I’m afraid. And you’ve forgotten me enough to get my title wrong, but then people often do. Or perhaps you’re just being impolite again; it can be hard to tell lack of politeness from lack of competence sometimes.
NN> we should not engage in correcting high-school math errors such as those made by Willis above
Interesting! I haven’t spotted any maths errors by WE, though I didn’t look that closely. I shall watch out for your interesting analysis. I do hope you’ll provide the promised errors in maths, rather than interpretation.
NN> The major point that Willis totally missed (but follows from his own regression analysis) is that the planetary surface temperature CAN be explained…
Err no. That is the point WE is desperately trying to make (indeed has made, quite convincingly) but you haven’t yet understood: that it is all to *easy* to explain the temperatures, using curve-fitting. Explaining it with physics is much harder, you should try it sometime.