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.
Richard M
Where have we claimed that we replace the greenhouse effect with gravity?
What we state is that the GH effect, when measured as a dimensionless number (Ts/Tgb), i.e. the relative thermal enhancement, is completely explainable by pressure. Is pressure a gravity? No! Pressure is a FORCE resulting from the atmospheric mass per unit area AND gravity! What is the kinetic energy of a gas that determines its temperature? It is a product of Pressure and Gas Volume (PV), i.e. FORCE x Distance = Joules. In other words, you cannot have kinetic energy and temperature of a gas without a FORCE. On a planetary scale the force of pressure is INDEPENDENT of solar heating, atmospheric volume, or temperature, because we have on average an isobaric thermodynamic process at the surface. So, changing the mass of the atmosphere will change the FORCE generated by gravity at the surface, therefore, changing the temperature. Our non-dimensional NTE factor (the relative thermal enhancement) is a manifestation of that physical characteristic of pressure called FORCE … How is that for a physical explanation? We elaborate more on this in our Reply Part 2 …
The key to grasping our theory is understanding the actual physical meaning of different parameters such as pressure, irradiance, temperature, and energy and the best way to do that is to properly deciphering the units …
Joel Shore says:
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!
Nice try, Joel. The Gas Law does not violate the 1st Law of Thermo, sorry! What violates that law BIG TIME is the idea that changing the concentration of minor gases in the FREE atmosphere can increase the kinetic energy of the troposphere and cause surface warming, or the assertion that some ‘heat-absorbing’ gases which amount to less than 0.5% of atmospheric mass can raise Earth’s surface temperature by 133K above that of an equivalent gray body! That’s more than a violation of the 1st Law, it’s a SUPERSTITION. Do you know how much of addition radiation is needed to actually raise the temperature of the Moon by 133K?? 14,800 W m-2 …
Please, tell your friends at Kodak that you are failing the assigned task here, and they should stop wasting money on you … 🙂
> One other piece of advice, to all: “Don’t feed the trolls!”
But we can ask the Moderators to step in and snip/ban the trolls, when it becomes clear that they are trying to disrupt, not contribute to, the discussion.
😐
Ned Nikolov says:
January 25, 2012 at 11:30 am
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.
A reasonable value for surface heat capacity would bring the temperature of Earth to about 250K and the GHE could do the rest.
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?
We know what the cause of the large discrepancy that you calculate, it is your assumption of zero heat capacity, include that then we can talk
NN> properly deciphering the units …
Good point. So given that Nte is dimensionless, and Ps is in N/m2: in “Nte(Ps) = e^(t1 * Ps ^ t2 + t3 * Ps ^ t4)”, what units are t1 and t3 in?
William M. Connolley says:
See my reply to Richard M above for a physical explanation …
You seem to have limited knowledge about how science works. Do you know that all major theories that gave birth to so called ‘first principles’ have began with fitting empirical data to curves and inferring responses and functions? Do you have a science degree in a non-soft science area?
Willis,
Degrees of freedom vs parameters. End of story. Not to be Shakesperian, but this seems to be much ado about nothing, or perhaps a tempest in a tea pot. Not that you are, of course, not right.
William M. Connolley says:
January 25, 2012 at 12:39 am
Threatened by you, Billy? By YOU??? Oh, stop, please stop, is this your new plan? This is truly beyond evil, I see through your nefarious plot—you intend to kill me by making me laugh myself to death. It’s diabolical, I beg you, please have mercy, don’t threaten me again, my stomach can’t take the laughter and my cheeks are aching from grinning at your antics …
As for “showing my real side”, yes, my real side does indeed think that you are a unpleasant and undesirable person. You are willing to do anything to promote your POV, and it sure isn’t an NPOV. Your claim that we don’t really understand why you were forced out of Wikipedia is pretty funny, though. You were thrown out, but it had nothing to do with your POV and everything to do with you being a POS. You see, Billy, I was one of the nobodies who actually believed Jimmy Wales when he began wikipedia. And I was one of the nobodies who tried to do what Jimmy said, contribute. And I was one of the nobodies who got resounding erased, and re-erased, and re-erased, by the King Of Wikipedia climate censorship. That would be you … it didn’t matter if I posted facts, didn’t matter if God himself said it, if you disagreed with it, it was gone.
So I see you quite clearly and accurately, Billy, and what I see turns my stomach. And yes, that is my real opinion, my “real side” if you will.
Gosh, Billy, did you really think when I said that I was referring to your meaningless opinion on this issue? Your reading comprehension is worse than I thought … no, belay that, actually, given your history at Wiki, your reading comprehension has always been uniformly horrible, so I guess it’s no worse than ever.
And now you announce your intention to school me on computers, clearly another part of your devilish “make him laugh until he croaks” plan. I could point out that I wrote my first computer program when you were in nappies, Billy, but that wouldn’t be true. I wrote my first computer program before you were born.
But please, don’t spare us your hilariously obvious insights about computers, continue with your brilliant comic exposure of your fantasies, prejudices, and personal shortcomings. This is great entertainment, and to date the laughter hasn’t killed me, so do your worst, keep it up, leave’m dying in the aisles.
w.
PS—Some people have said that I’m too crude or too hard on poor William. Quite the opposite. The problem is not that I am too outraged at what Mr. Connolley has done.
The problem is that the rest of you are not outraged enough. The man has ruthlessly censored real, valid scientific opinion for years, and you want to play patty-cake and go all California on him, ask him about his feelings and invite him to share? Not me. He doesn’t deserve to hang out with decent people, I am never happy to see him show up, he is a force for damage and dissension who leaves trouble in his wake, and I think it is important that people know who he is and what he has done.
Is he beyond the pale, beyond redemption? I have no idea. But for him to show up and immediately start, not by adding to the scientific discussion but by attacking me personally, should give you a clue. You can pat his head if you want, I’m kinda attached to having all my fingers, he is unrepentant and he bites. I’ll wait until he shows some sign of at least noticing the trail of damage he has done to free scientific thought before I change my opinion.
Willis Eschenbach,
We are in total agreement on this:
“The problem is not that I am too outraged at what Mr. Connolley has done.
The problem is that the rest of you are not outraged enough.”
@joel Shore
> 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.
Assuming, for the sake of argument, that this ‘extreme error’ is real, that doesn’t rule out that there might be a correct way of adding convection to make this idea
“work” (or not). I suppose N&Z will further address this issue in Part II.
But these convective claims do make sense to me in this qualitative way (waving my hands): the surface of the Earth absorbs most of the heat energy coming from the Sun. The atmosphere much less, especially the bulk of it O2 and N2 being transparent to heat radiation. But the GHG’s, though present in much smaller percentages, do behave like the surface in that they are capable of absorbing radiated heat too.
Here’s the difference: when the GHG’s heat up they can rise up and release this heat much closer to the TOA. (Too bad, Surface, you’re Grounded!) Furthermore, H2O can change states, up in the clouds, and release even more heat (of condensation) up high.
I know this is nothing new, but it sounds like GHG’s comprise a kind of ‘heat engine’ which involves both heating and cooling to me. The trick is to nail down the physics here in a quantitative way, which seems to be a very controversial topic. Beyond my pay grade. I hope N&Z (et alia) will give us further analysis of this and possibly eliminate some the controversy with their new ideas.
@me
> … when the GHG’s heat up they can rise up and release this heat much closer to the TOA…
Forgot to mention that O2 and N2 are also involved, of course, by being warmed by conductive transfer of heat from the surface to the air in direct contact with the ground. So they are also part of this ‘heat engine’.
Nick Stokes says:
January 25, 2012 at 4:09 am
Indeed, that has been asked before and not answered as far as I know. Nor do I know at what level one measures “surface pressure” on a planet with no ocean.
w.
Regarding Connolley, Willis says: “You were thrown out, but it had nothing to do with your POV and everything to do with you being a POS.”
Seconded, doubled and squared. Connolley is a despicable insect who is unable to get traction with his climate alarmism, so he misuses his insider position to censor opposing points of view, no matter how polite and scientifically factual. Connolley is simply dishonest.
The difference between Wikipedia propaganda and WUWT – the internet’s “Best Science” site – is that Connolley is allowed to post his comments here. If equivalent comments were posted on Wikipedia, censor Connolley would arbitrarily delete them if he was not in personal agreement.
That makes William Connolley a despicable human being. Unfortunately for Connolley, he is unable to censor comments here like he does on Wikipedia – a blog that is losing credibility precisely because of Connolley’s dishonest censorship of different points of view.
Ned Nikolov says:
January 25, 2012 at 11:54 am
Where have we claimed that we replace the greenhouse effect with gravity?
Sorry I didn’t specify all the factors. I have mentioned them all in other comments this past few weeks. It was probably due to commenting recently on the Jelbring paper.
However, the problem is and will always be … without changing the effective radiation height nothing can warm the planet above the surface temperature and not violate thermodynamic laws. All of the planets that have atmospheres have radiating gases so there is no problem here.
I believe what you may have found is the GHE ltself is limited by equation 8. That puts just as big a nail in the AGW coffin as anyone needs. It means further additions of GHGs will have no impact. If you can demonstrate the physics that leads to this limit then you truly will have made a huge contribution.
John Day says:
Yes…That can be ruled out. It is ruled out by the fact that when the scientific community has added convection correctly, they do not get the result that N&Z get. Besides which, it is easy enough to correct what N&Z did in adding convection to the toy model of the radiative greenhouse effect and one indeed sees that their claim of the radiative greenhouse effect disappearing is no longer true.
NN> See my reply to Richard M above for a physical explanation …
That doesn’t answer my question. I asked, what units are some of the constants in your equation in? Surely you can answer that simple question?
In your version, it is the “0.233001”. Is it dimensionless (I can’t see how it can be) or if it has dimension, what are the dimensions?
WE>
You’ve gone a bit dull I’m afraid. But I need to make a slight correction: NPOV should have been WP:RS (http://en.wikipedia.org/wiki/Wikipedia:RS). Yes, I know neither you nor any of yours care about wiki’s policies, but those of us who edit have to.
WE> I was one of the nobodies who tried to do what Jimmy said, contribute…
Really? Not under your own name, then. Did you do it secretly?
Ned Nikolov says:
The Gas Law is not what we are talking about violating Thermodynamics. What violates the 1st Law is your misapplication of it to claim that the Earth can emit 390 W/m^2 from its surface but only 240 W/m^2 as viewed from space, and that the discrepancy can be explained anything other than the fact that some of the surface emissions are absorbed by the atmosphere!
(1) Perhaps you didn’t notice but there is an object in the center of our solar system called “the sun”. It emits radiation to the Earth…and the Earth also emits radiation back out into space. Hence, the Earth is not an isolated system and you can’t apply conservation of energy as if it is.
(2) Nobody but you seems to believe that “heat absorbing” gases raise the Earth’s temperature by 133 K. They raise it by 33 K. The other 100 K is accounted for by the fact that different temperature distributions with different average temperatures can yield the same average emission of 240 W/m^2.
No…It is called “science”.
Really…And, you calculated that how exactly, given that a blackbody at 288 K is only emitting 390 W/m^2?
I am not at Kodak anymore. Speaking of failing…I must admit that I expect MUCH higher standards of science and scientific integrity from our public servants than I have seen displayed here!
Ned Nikolov said @ur momisugly January 25, 2012 at 11:30 am
Sorry, wrong! It’s in T R Oke’s Boundary Layer Climates and a host of other places. It’s part of the Received View. That is, it’s what is taught at universities. Do you not understand what “well established” means? Get a grip! If you want to overthrow a paradigm theory, you don’t start by refusing to acknowledge that it is the established paradigm theory. I said earlier that you need to show that you understand what it is that you are claiming to overthrow. Ignorance doesn’t cut it.
Pseudo-scientists do so love to compare themselves to Copernicus and Galileo don’t they? It is not a “superficial observation that Heavens revolve around Earth”. There’s nothing “superficial” about it. I observe it most nights. Yes, I know because I accept the Received View in astronomy that Earth revolves around the sun, but it’s manifestly not what I observe.
It’s worth bearing in mind why Copernicus model* was only slowly accepted by astronomers. It predicted a variation in diameter of 8:1 for Mars and 6:1 for Venus. There is no observable change for Venus and only a ratio of 2:1 when making naked eye observations. This is why Galileo was abjured not to take the model literally, but it was OK to use it mathematically. It wasn’t until that discrepancy was resolved that the Copernican model of the planets orbiting the sun (in perfect circles with even more epicycles than Ptolemy used) was accepted.
“The whole notion about the GHE being a radiative phenomenon is based” on observed emission spectra, not confusion. It may well be “wrong”, but you sure as hell are not making your case for anyone with a grounding in basic physics and philosophy (logic).
* Footnote: I wonder how many “sceptics” are now going to stop believing the Copernican model of the solar system just because it’s a model. Maybe they will just deny that it’s a model…
Richard M:
Please, consider the following important point (, which we have not discussed in the papers released to the public) – There is NO effective radiation height in reality. The concept of an emission height come from the incorrect application of the SB law when estimating the mean temperature of a sphere (see our Reply part 1). That simple equation produces about 255K (-18C) for Earth. This temperature is then erroneously identified with the actual measured temperature at about 5 km in mid troposphere, and a conclusion is drawn that this is where most of the IR radiation had to be emanating from. The reality is that IR radiation is coming out from all heights in the atmosphere simultaneously including the surface. It’s a continuum of IR fluxes in different wavelengths reaching space from various depths of the atmosphere. That continuum of emitted IR radiation with depth is what enables the remote sensing of vertical temperature and moisture profiles (as well as other atmospheric parameters) from satellites.
The ’emission temperature’ of a sphere computed from the simple inversion of the SB law is not really compatible in a physical sense with any palatable real temperatures measured in the atmosphere or at the surface … Hence, the emission height is another fiction of the GH theory with no physical equivalence!
“Nick Stokes says:
January 25, 2012 at 4:09 am
One of the mysteries of this paper is where the observed T_s values (which they claim to predict) came from. For most, no information on sources is given at all. Europa? Triton? What did they use and how did they get a global figure?
Indeed, that has been asked before and not answered as far as I know. Nor do I know at what level one measures “surface pressure” on a planet with no ocean”
As I pointed out if either T or P is specroscopically determined then the other will have a SB distribution.
Dr. Nikolov
You know much more about fysics then i do.
Stilli have this question.
Maybe the word back-radiation is misleading but its meaning does not have to be so.. Maybe we should recall it into “blocking radiation” or something. There is in fact no production of down-welling flux of 343 W m-2.. There is a delay of outgoing radiance.
It’s like my banket in a cold night. It doesn’t warm me but it isolates my body.
With no blanket i schould lose a lot of bodywarmth. So in fact my blanket warms me a lot. What of course is in fact not true.
tallbloke says:
January 25, 2012 at 8:23 am
Thanks, Tallbloke. I guess my point still isn’t clear to you, and it’s not in how I derived density.
What N&Z have done is just curve fitting. You can do it as you have done, with solar and pressure. You can do it as I have done, with Tgb and Gravity and Density. You can do it with solar and pressure using a much simpler equation than you used. You can do it with solar and density and a simple equation.
My point is, it is a painfully easy curve to fit. You can fit it in a host of ways. But that doesn’t mean anything, TB. With four parameters you can fit an elephant. I’m a good mathematician and it’s a simple elephant, so I did it with three, but it’s STILL JUST FITTING AN ELEPHANT!
Using five parameters and a free choice of any equation when there are only eight data points is a laughable newbie mistake, tallbloke. It’s not a MIRACLE. It’s not a blinding insight. It is a trivial equation fit, and I’ve just demonstrated that by fitting it in a variety of ways.
But forget about my demonstrations. Go ask the statistician of your choice. Ask them if fitting eight data points, given a free choice of any equation under the sun and five free parameters, is a valid and meaningful procedure.
They will tell you what I am telling you. It is risible. The fact that N&Z included it in their argument shows that they also are clueless about models and parameters. It should be a sign to anyone considering their theory.
I can’t state this strongly enough, tallbloke. Five parameters, eight data points and a free choice of equations is a no-brainer fit and it means nothing. Zip. Zero. It has no probative value. Ask a real scientist, TB, you’re embarrassing yourself claiming this procedure is valid.
w.
Richard M says:
January 25, 2012 at 8:56 am
No, no, no, and no. Let me say it a different way, as clearly I still have not gotten the point across to you.
I have not “confirmed the relationship may be even stronger”. I’ve just fit the same points with a different curve. So far, N&Z fit it one way, and I’ve fit it three other ways using different variables and different equations and different numbers of parameters.
All that means is that the problem is trivially simple, because there are five tunable parameters and eight data points and I’m given free choice of any equation, no matter how non-physical it might be.
So none of the fits mean anything, Richard.
w.
John Day says:
January 25, 2012 at 9:11 am
My point exactly, John. N&Z are just “gaming the system” as you put it, their fits mean no more than mine do.
w.
Dr. Zeller, as the man who challenged us to look at equation 8, the MIRACLE equation, thank you for returning to defend it.
As you see above, my problem with the equation is that there are five tunable parameters, only eight data points, and you have free choice of any kind of equation to fit to the data points.
Please, please, sir. You are doing your reputation great damage by continuing to insist that fitting a curve to eight data points is somehow meaningful. I have shown three other ways to do it, using different variables, different equations, and different numbers of parameters. What that means is this:
Fitting eight data points when given five free parameters and free choice of equations means nothing because it is so trivially easy to do.
Now, I know you don’t believe me because I don’t have a PhD. So ask your favorite statistician. Ask someone involved with modeling natural phenomena. Ask any hardcore scientist, and they will tell you what I just said. Your results have no meaning because your model could be the poster child for overfitting.
My detailed comments follow:
kzeller says:
January 25, 2012 at 9:51 am
I see no miracle in fitting an elephant with five parameters. Neither did “Johnny” von Neumann.
Dr. Zeller, you should consider the question of how to test and validate a model such as yours. This is not new stuff, the statistics of it are well understood. Your claim is … painful.
I guess you missed the part where I explained that my equation actually provides a better fit than yours … or maybe not. But like I said, you’re welcome to use it …
My friend, what part of my demonstration above did you not understand? I just fit your elephant using an exponential equation, here it is again, see Figure 3 for the results:
See the exponent part on the right?
Er … um … you might want to research the whole topic of “overfitting” before sending out the obituary notices for AGW …
w.
Ned Nikolov says:
January 25, 2012 at 1:55 pm
Please, consider the following important point (, which we have not discussed in the papers released to the public) – There is NO effective radiation height in reality. The concept of an emission height come from the incorrect application of the SB law when estimating the mean temperature of a sphere (see our Reply part 1). That simple equation produces about 255K (-18C) for Earth. This temperature is then erroneously identified with the actual measured temperature at about 5 km in mid troposphere, and a conclusion is drawn that this is where most of the IR radiation had to be emanating from. The reality is that IR radiation is coming out from all heights in the atmosphere simultaneously including the surface. It’s a continuum of IR fluxes in different wavelengths reaching space from various depths of the atmosphere. That continuum of emitted IR radiation with depth is what enables the remote sensing of vertical temperature and moisture profiles (as well as other atmospheric parameters) from satellites.
The ‘emission temperature’ of a sphere computed from the simple inversion of the SB law is not really compatible in a physical sense with any palatable real temperatures measured in the atmosphere or at the surface … Hence, the emission height is another fiction of the GH theory with no physical equivalence!
Yes, it is a continuum. That’s why it’s called an effective radiation height. No one believes all the radiation comes from a single altitude. The point is it does not all come from the surface. Without radiating gases it would have to come from the surface and the surface would have to average out to 33C cooler than it is. I don’t see anyway to avoid it.
I can see how the atmosphere could limit the GHE based on mass, gravity and solar input. Those items are what creates the structure of the atmosphere. All the gases work within that structure and it may be possible that the overall GHE is independent of the exact concentration of the gases and adhere to the Miskolczi constant optical depth concept. However, the atmosphere still requires a minimum amount of radiating gases.