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.
@me
> I posted a comment about this to Nilolov,
Ned, if you’re listening, sorry about the typo, “Nikolov” (Николов, по русски. Читаю немного)
John Day says:
January 28, 2012 at 12:21 pm
No. I haven’t seen a single proposed mechanism other than ghg’s that is feasible. That doesn’t meant that something else isn’t possible, just extremely unlikely.
Also, no. At best it would be a minor correction to the existing theory if it were true, which it isn’t.
John Day said @ur momisugly January 28, 2012 at 12:21 pm
Let’s assume ATE is correct. Now, how does ATE explain the incoming and outgoing spectra that clearly show evidence of emission and absorption?
I’ve got preview and I failed to use it. Doh!
The line: If ATE turns out… should have been in blockquotes.
[Fixed. -w.]
@John. Sorry about the “el”.
@joel Shore. You have the answers – I can see that. All the gaseous planets are shrinking as they lose heat and this gravitational compression causes heat. So Neptune with a radius of 25,000 km emits twice the amount of heat that it receives from the sun, the same amount emitted by Saturn which has a radius of 60,000 km (over two times larger) which emits more heat than Jupiter (1.6 times) with its diameter of 71,500 km. Then there is Uranus with a radius of 25,500 km that emits the same amount of heat as it receives. And I thought Uranus would be shrinking at the mere thought that these numbers don’t make sense. Maybe it’s just puckering.
Neptune. R=24,765 km, Net heat=2 times
Saturn. R=60,268 km, Net heat=2 times
Jupiter. R=71,492 km, Net heat=1.6 times
Uranus. R=25,559 km, Net heat=0
Sorry. Should be:
Uranus. R=25,559 km, Net heat=1.0 times
Uranus, mass = 8.68E+25 kg, net heat = 1.0 times
Neptune, mass = 1.02E+26 Kg, net heat = 2.0 times
Saturn, mass = 5.69+26 Kg, net heat = 2.0 times
Jupiter, mass = 1.90E+27, net heat = 1.6 times
So the net heat that is caused by gravitational compression doesn’t appear to be directly related to either the size of the planet or its mass. Joel? Some help here.
Willis and Joel,
Your criticisms lack merit. If you want a valid basis to criticise the design, conduct or results of my experiment you will need to present your own empirical experiment. BS(blackboard scribbling) does not count.
WUWT readers who have some concerns about Willis and Joel dismissing empirical results when they have none of their own can visit
http://tallbloke.wordpress.com/2012/01/22/konrad-hartmann-experiment-to-determine-the-effect-of-pressure-on-temperature-in-earths-atmosphere/#comments
and read first hand about what was done. The experiment was designed to be replicated by others. Construction instructions and a materials list is provided.
@DeWitt Payne
Yes, exactly what I meant. (I get plausible-feasible-possible mixed up all the time)
> Also, no. At best it would be a minor correction to the
> existing theory if it were true,
Now that I would have to quibble with a bit. Certainly the social impact would be enormous, Al Gore and his minions would be out of a job. Kyoto and other thinly disguised ‘spread-the-weath’ scams would go out the door.
I think it would also have an effect in the [secular] environmental sciences too. Knowing that temperature, on a planetary scale, is independent of the chemical composition of the earth and atmosphere.
> … which it isn’t
Ah, I see your position is already ‘entrenched’. No problem, perhaps that could change as the theory is unfolded. I’m not entrenched either way, but I do find ATE compelling for symmetry reasons (without completely understanding yet how it works).
@git
> Let’s assume ATE is correct. Now, how does ATE
> explain the incoming and outgoing spectra that
> clearly show evidence of emission and absorption?
Let me rephrase that question: “Let’s assume the Copernican theory is correct, Now how does it explain the epicycles which are clearly visible in the heavens?”
So your question has a bit of the same ‘begging the question’ fallacy that Willis’ proof contained.
The answer to the question is that our knowledge of the existence and charertistics of these spectra won’t change at all, in the same sense that we can still “see” the epicycles. It’s just that ATE has eliminated the need to interpret physical chemistry (absorption spectra etc) to explain planetary temperatures. All we need is pressure.
So, you could argue, “But I can predict the planets’ positions accurately using epicyles. That proves Copermicus is wrong” [Maybe your predictions are correct, but Copernicus is still right]
or equivalently “But I can predict planetary temperatures correctly using absorption spectra. That proves N&Z are wrong” [Really? Then why do we need so many fudge factors (aka “sensitivity coeficients”) to make the temps match our expectations. No such ‘fudge factors’ are needed for ATE, what you measure (pressure) is what you get (temperature).]
CAVEAT: Assuming ATE is correct (and pigs have wings) of course.
😐
John Day says:
No…The reason it cannot be true is that it conflicts with well-established Laws of Physics. For example, we have a Law in Physics called “conservation of energy”. If you want to establish some new mechanism, you have to demonstrate how it obeys this law. (Or, make the extraordinary claim that this law is violated, which would require extraordinary evidence.) Nobody has been able to explain how any ATE mechanism that does not rely on the absorption of radiation by the atmosphere (which indeed depends on atmospheric composition) can obey this law.
We don’t just go around accepting any “theory” that comes up just because it agrees with what one’s ideology wants nature to be like. That is not science.
What don’t you understand? It is all very basic stuff…Stuff you certainly have to get under your belt first before you have any ability to say anything intelligent about speculative “theories”. I’d be curious to know what parts of it you find confusing.
Steve from Rockwood says:
Yes, it could easily be more complicated than a simple dependence on radius or mass. I am no expert on the details of the current understanding of the role of gravitational compression in each of the gaseous planets. Here are two wikipedia articles about the basic mechanism though:
http://en.wikipedia.org/wiki/Gravitational_compression
http://en.wikipedia.org/wiki/Kelvin%E2%80%93Helmholtz_mechanism
John Day says:
You seem to have a hard time distinguishing between physics and curve-fitting. You also seem to lack the understanding that N&Z have plenty of “fudge factors”: They have 4 free parameters for a purely empirical equation with no theoretical understanding (except for gobbleygook about things like the Ideal Gas Law and about convection that are either do not support their theory or are demonstrably incorrect).
N&Z are nothing but a litmus test of how silly a notion can be and still attract support from people who want to believe it. It is just a Rorschach test…Useful for distinguishing fake skeptics from people who really exhibit at least a little bit of skepticism. A lot of people who should know better are failing the test.
Two more comments before I clear out of here for the evening:
1) So far we’ve been evaluating the correctness of ATE based on preliminary calculations provided by N&Z in their poster and comments on this blog.
Question: What if those calculations contained arithmetic errors, would that invalidate theory?
In other words, what if Copernicus made a few trivial math errors such that his plotted predictions of the planets were less accurate than ones made with Ptolemaic theory? Would that mean that the Copernican theory was totally false?
Well, the disciples of Ptolemy would certainly argue that. (And I see a some of that going on here in this blog).
Take away: Natural Law is bigger than the men and women who propose it. Errors in presentation of proofs of these laws have no effect on the actual truth of laws of nature.
2) If you believe ATE is false, can you explain why it is not necessary to know the chemical composition of an airless planet to compute its temperature, yet knowledge of chemical composition _is_ necessary if that planet has an atmosphere? Symmetry of natural laws would suggest that if composition was necessary for one, it would be necessary for the other configuration.
And isn’t strange you don’t need to know the composition of the surface, which is very dirty and complicated, yet knowledge of composition is necessary for the relatively pristine atmosphere, even for miniscule whiffs of CO2?
The surface has much greater entropy in an information theoretic sense, than the atmosphere. To see this, consider samples taken at random from the surface and atomsphere and analyzed for composition. A Boolean 92-vector (one Boolean flag for each natural element) reveals the entropy of the sample in terms of the number and variety of bits it would take to describe each sample.
Air samples would tend to have the same small number of bits set for each sample (because the atmosphere is well-mixed with respect to the surface (i.e low entropy, not much information required to describe all the samples).
Whereas the ‘dirt’ samples of the surface would have a much larger number of bits set and much greater variance in which bits were set (i.e. extremly high entropy, a lot of information needed to describe all the samples).
Yet at the planetary scale, the surface requires less information for temperature prediction than the atmosphere (if you don’t believe in ATE).
Why is that? Why the lack of “symmetry”?
You GHG-GHE believers please explain that to me. Thanks.
😐
@joel
> You seem to have a hard time distinguishing between physics
> and curve-fitting. You also seem to lack the understanding
> that N&Z have plenty of “fudge factors”:
I do know the difference quite well. Actually I’m getting the impression from some of you folks (esp. Willis) that you think ATE is entirely about curve fitting.
For these theoretical discussions here I’m not all interested in those curve-fitting exercises. Not al all. Yes, I agree that there must necessarily be a bit of “fudging” in making empirical fits.
But again. At this stage I’m not interested in discussing those emprical equations (unless we talk about how they might be related to actual physics).
So unless you all insist otherwise, let’s start by discuss ATE from principles based on the laws of physics. Hard to get motivated by curve fitting. Agreed?
So, apart from the curve fitting, in the formal ATE theory (as you understand it) what “fudge factors” are you referring to, Joel?
😐
John Day says:
There are not just little errors. Their whole “theory” is just built on fundamental misconceptions. The only reason people in this little section of the universe are not just dismissing it as nonsense is because it is telling people here what they want to believe. But, notice that the most serious scientists amongst the AGW skeptics, e.g., Roy Spencer, Willis, Ira, and Robert Brown are desperately trying to give you their reasoned opinion as to why this is nonsense.
I don’t know what “symmetry of natural laws” means in this context.
In this case, the reason for the different is very simple: Most solids and liquids have a basically continuous spectrum of excitations at infrared wavelengths so that they behave quite close to blackbody radiators and absorbers in this wavelength regime. Gases, on the other hand, are much different because the molecules are so weakly interacting. If a gas molecule does not have an excitation mode that can couple to electromagnetic radiation in a certain wavelength regime, then it simply will not absorb or emit radiation. This means that molecules like N_2 and O_2, which make up most of our atmosphere are transparent to radiation in the mid- and far-infrared. The small amount of interactions that do occur between molecules can allow for a very small amount of absorption and emission as the density of the gas increases, but this is just not enough to be close to relevant for Earthlike atmospheric densities.
@joel
> No…The reason it cannot be true is that it
> conflicts with well-established Laws of Physics.
Here, let me rephrase that for you: “The reason Copernican Theory cannot be true, is that it conflicts with the well-established Laws of Epicycles”
Yes, I know that ATE hasn’t been proven. But you can’t prove a new theory’s falseness merely by claiming that it ‘conflicts’ with current theory. (If by ‘conflict’ you mean it proposes a different mechanism to explain the same results).
It is densely packed. I just didn’t have time to wade through it. Yes, I will go through it and understand it. But my initial impression, from skimming, is that you’re merely presenting the ‘status quo’ again as ‘proof’ that can’t possibly be any other better theory. But I appreciate the time you took to write it out for me. I’ll ask for clarifications if necessary. (Remember: remarkably ignorant me)
BTW, were you able to find that Pierrehumbert reference you made about “page 148”? I would also like to understand the point you were making about that. Thanks.
😐
John Day says:
January 28, 2012 at 2:59 am
Of course you’re not sure …
If you’re not sure about it, John, and I’ve said I don’t give a fig for what you think because I’m waiting for Ned to answer … then what on earth would impel you to share your vague fantasies about what Ned means?
STOP, John. Please do yourself a favor and just stop posting your nonsense about the question for Ned. If Ned answers, he does. If not, he doesn’t. In neither case is your “i’m not so sure about that” guess as to what he means of the slightest use to anyone.
I hate to be so blunt, but my friend, you are clueless in these matters, and from all appearances, you don’t have the faintest idea just what a spectacle you are making of yourself in public. Read and think for a while, do some research, get a textbook, I don’t know how you might go about it but you should truly do your homework before uncapping your electronic pen.
Let me make my position clear. I think you’re capable of learning what you’d need to know to contribute to the discussion. I don’t know if you will do so, but I certainly encourage you to apply yourself.
My best regards to you, and I am totally serious about what I said above. You’re like Wile E. Coyote after he runs off the cliff, you have no idea you are standing on thin air …
w.
@Willis
> Of course you’re not sure …
Sorry my words were indeed ambiguous. What I meant was “I’m not so sure that you know ”
To repeat: it is very clear that the 14,800 watts/m2 refers to the additional warmth that the atmosphere provides compared to an airless planet. Were you aware of that? (That’s what I wasn’t sure about.)
It should be clear from the N&Z writeups that he was trying to impress that the ATE theory will generally come up with a much larger value for this temperature enhancement. If this larger enhancement turns out to be valid (TBD), then it will be more difficult to maintain that it is generated by GHG’s.
1) The fact that warming will occur, qualitatively, is true whether or not you believe in ATE.
2) But the actual amount will vary, because of the way ATE computes the GB temp based on the Hoelder integration.
How did Ned come up with exactly the number “14,800”. I don’t know.
I agree with you, Willis, only Ned can provide that piece of information. I don’t know how he came up with “14,800”.
But even if it represents some miscalculation on Ned’s part, that doesn’t necessary invalidate the whole theory. Errare humanum est
😐
@Kuhnkat
> I think Willis would help the discussion along
> by not heaping the ridicule so quickly and so high.
Thanks for those words of support. His ridicule really doesn’t bother me anymore, because my “anti-venom” is knowing that that he’s wrong on these issues.
😐
I thought this response over at Tallbloke’s was worth bringing over:
New comment on Tallbloke’s Talkshop
Konrad commented on Konrad Hartmann: Experiment to determine the effect of pressure on temperature in Earth’s atmosphere.
in response to tallbloke:
Many thanks for allowing me to present the results of my basic empirical experiment into the Nikolov and Zeller hypothesis at the Talkshop. Introduction The Nikolov and Zeller hypothesis has generated thousands of comments across several climate sites, but sadly little in the way of empirical evidence. After the first few hundred comments I decided […]
@ur momisuglyTilo I have had time to run further tests using the same rig. With regard to allowing the bottles to reach equilibrium away from sunlight, both equalise at the external air temperature. I have run further tests with pure CO2. The first was to run the same experiment again. One bottle at atmospheric pressure, the other maintained at a higher constant pressure. The results were as before, with the higher pressure bottle rising to a higher temperature when exposed to sunlight. The only difference noted was that the higher pressure bottle appeared to cool more swiftly when the rig was removed from sunlight. I conducted a further test in which both bottles were at atmospheric pressure but one contained air and the other 100% CO2. I was unable to measure any temperature differential between the bottles. They both heated at the same rate. I swapped bottles and probe thermometers. I tried sunlight and then a flood lamp (more IR) without success. If CO2 has some effect between 25 and 50C it is below the 0.1C resolution of my thermometers.. @ur momisugly tchannon The hot water bottle and the bricks do ensure a constant pressure in the high pressure bottle. The volume will adjust as the temperature increases. @ur momisuglyP.G. Sharrow Water columns are a good idea, but I would need a U shape over a metre in height to achieve similar pressures. @ur momisuglyTallbloke I’m sure a simple LOL or ROTFL would be an elegant sufficiency 🙂
John Day says:
The point is that it doesn’t just conflict with current theory…It conflicts with things like conservation of energy. And, the way in which it conflicts with current theory is based on huge misunderstandings of basic points. For example, the authors don’t understand that the way one applies conservation of energy to a planet is not by assuming that the planet is isolated from the rest of the universe (and hence its average temperature can’t change) but by assuming it interacts radiatively with the sun and space. Another example is the authors add in convection into a simple radiative model of the greenhouse effect in a way that drives the atmosphere to be isothermal with height, in clear contradiction to reality and then marvel at getting a result (the disappearance of the greenhouse effect) that is well-known to be true for an atmosphere with an isothermal temperature profile.
These are simply elementary bone-headed errors.
Yes, it is in section 3.3 of his textbook (“Radiation Balance of Planets”), on p. 148 of the final printed version. In it, he is discussing the simplest picture of the greenhouse effect…and one thing that emerges from this picture is that “the greenhouse effect only works to the extent that the atmosphere is colder at the radiating level than it is at the ground”.
kuhnkat says:
January 28, 2012 at 2:30 am
Nope. We’re trying to tell you that smart people give much thought to and design their experiments very carefully so that they show something that is useful about their theory.
Who would have guessed?
w.
Willis,
you know I am a pot calling the kettle black don’t you?? It is an issue I need to work on also and is too common generally. As you say, the frustration with people who simply won’t see the TRUTH gets to us.
Please point out where I have argued for a process that would result in perpetual motion. As you point out, I may not have realized that is what I was doing and would appreciate the information even if I don’t understand it.
“Nope. We’re trying to tell you that smart people give much though to and design their experiments very carefully so that they show something that is useful about their theory. Who would have guessed? w.”
Yes, and I am telling you that everyone is not that smart and can still get useful results. Whoda thunk?
John Day says:
January 28, 2012 at 11:20 am
Gosh, John, do you think a clue might be found in the post I was responding to? Let’s see what that post was:
jae says:
January 27, 2012 at 7:10 pm
So I was just messing with Jae, because he was claiming that he had the power to set the rules about what was a win for me and what was not, involving me reading some guy I have no interest in and “refuting” his evidence. You can’t refute empirical evidence, only interpret it, but let that be. I found it pretty funny that some random anonymous internet poster had puffed himself up to where he was dictating what is or isn’t a win for me.
So I thought I’d list a few of my wins. I’ve been on a bit of a roll, and since Jae wanted to dictate what constitutes a win, I thought I’d mention what I consider a win.
I am sorry to see that my listing my wins seems to infuriate you. It’s your choice in your hands, you know. You can just ignore me and continue with your important conversation that you mentioned above.
Or I tell you what. You can list your wins too, and I can ignore them. Then we’ll be even, and we can move on.
All the best,
w.
Konrad says:
January 28, 2012 at 4:24 pm
Thanks for the reply, Konrad. Generally it’s good form to start by quoting the criticism you object to. That establishes what you are referring to, and allows others to see the error you say I’m making.
Then you need to demonstrate, not claim but actually show, that the criticism fails for specific reasons that you list and discuss.
I have said a couple things:
1. If you want to design an experiment to either falsify or support a theory, smart people start by first understanding the theory.
2. Once you understand the theory, then you can design an experiment that will be unambiguous and clear and measurable and with few confounding factors that will provide irrefutable evidence of the theory’s truth or falsity.
Which of those is “without merit”?
That is not true. I can evaluate your experimental design without proposing something better. If I see a flaw, I see a flaw, whether or not I could build it better.
I’m banned from Tallbloke’s Talkshop, so I fear your experiment will go unexamined. If you’d like to give the “elevator speech” for the N&Z theory that you are attempting to either support or falsify, I’d love to hear it. To date, no one has come forward to explain in clear concise terms how their theory works.
w.
kuhnkat says:
January 28, 2012 at 7:10 pm
Yeah, you’re probably right, kuhnkat. I’m working on it, and it’s a work in progress. I’m just tired of often-anonymous posers who admit that they have absolutely no understanding of physics, yet they proudly and loudly and stubbornly declaim that perpetual motion machines are really truly honestly possible if we just look at their beautiful theory in the right light. (Dim lights are greatly preferred. Shining a spotlight is discouraged.)
No.
Perpetual. Motion. Is. Not. Possible.
You and others may not realize that you are arguing for a physical impossibility. But some of us do. My problem is that I find the intransigence hard to take at times, but I’m learning.
Appreciated,
w.