söndag 8 mars 2015

Correspondence with Pierrehumbert on Back Radiation

Here is a response on my letter to Raymond Pierrehumbert:

Dear Claes,

“Back radiation” is simply the name given to the downward infrared flux evaluated at the surface.  The flux computation (in any direction) proceeds directly from the Planck distribution, Kirchoff’s Laws, and measured emissivities of the atmospheric constituents.  These have been established beyond doubt, both by theory and measurements, so if you have come to a conclusion that “back radiation” doesn’t exist, there is clearly an error in your calculation, and I’m afraid I have other things I need to be doing and can’t take the time to plow through it and see where it is.  “back radiation” doesn’t exist as a concept distinct from radiation fluxes in general, so it is impossible to “disprove” its existence without throwing out the whole of radiative transfer theory as well — which is about as well established as the Newton’s Laws of Motion.  

There have been a great many fallacious arguments about back radiation that have appeared in the past.  The following blog post has useful citations to the literature on radiative transfer, and also some discussion of where some of the other attempts to “disprove” back radiation have gone wrong. 

Perhaps studying those will help you spot the error in your own derivation.


—Ray Pierrehumbert

Here is my response and continued correspondence:

Dear Raymond:

Thanks for your answer. I understand that you base your climate science on a non-physical incorrect form of Planck's and Stefan-Boltzmann's radiation laws, where physical net heat transfer from warm-to-cold is artificially decomposed into two opposite transfers warm-to-cold and cold-to-warm. This is artificial as a procedure performed by symbols on a piece of paper, because heat transfer cold-to-warm violates the 2nd law of thermodynamics. If you do not make a clear distinction between symbolic manipulation and physics, gross mistakes can be made. 

The correct forms of Planck's and Stefan-Boltzmann's laws can only be understood by properly understanding the proofs of these laws. I thus ask you about your proof of these laws? What is it based on? What are the main arguments? I don't ask you to read my proof. I only ask you about your own. Since this is central to climate science and you represent the expertise, you must be willing to give an answer. 

Looking forward to your reply,

Sincerely, Claes


If you think you have a theory of radiative transfer that gives completely different results from that developed by Planck, Einstein and Chandrasekhar, you’d better be ready to show that it can account for the way stars shine, their photospheric temperature, the cosmic microwave background radiation, the successful retrieval of temperature in the atmosphere by satellite measurements of outgoing infrared and microwave radiation, the successful match of the spectrum of OBSERVED downward radiation flux to the ground and about a hundred thousando other ways that radiative transfer is applied throughout physics — and not just in climate problems.  Again, I urge you to study the basic physics and take another look at your derivation to see where you have gone wrong. Until you do, there is no point in continuing this discussion.

—Ray Pierrehumbert

Dear Raymond

I asked you about your proof of Planck's and Stefan-Boltmann's laws, but did not get any answer.
So I repeat my question. I don't think that you as leading climate scientist are in a position, where you can simply close the discussion without giving a proper answer to my question.

Looking forward to your reply.

Sincerely, Claes

From Pierrehumbert:

Oh for goodness sakes, “my” proof of Plank’s radiation law (and Kirchoff’s law) is no different from Einstein’s or the proof you will find in any elementary statistical physics textbook. Let me put this in the simplest terms I can:  a body with a temperature radiates.  In local thermodynamic equilibrium the radiation emitted is isotropic. That means some of the radiation flux is downwards. The sum of all the downward radiation flux reaching the ground from each layer of the atmosphere is what is commonly called “back radiation” it depends only on the temperature and composition of the atmosphere, and is completely independent of how the atmospheric temperature is maintained.  In particular, it  doesn’t require you to keep track of what happens to the radiation emitted upwards from the ground (that’s part of the separate computation of what maintains the atmospheric temperature.  

If you are underneath something with a nonzero temperature that has a nonzero emissivity, it is going to warm you. You will experience the downward flux. This is routinely measured at any number of observing stations around the globe (for references see the link I sent you earlier).  If you think you have a way of making the emitted thermal radiation anisotropic, good luck with making that consistent with the observed properties of the Universe and the objects of which it is composed.  

For the record, while back-radiation is important in determining the surface energy balance of a planet, the greenhouse effect can be understood more simply in terms of the top-of-atmosphere energy budget (which is purely radiative), and the vertical structure of the atmosphere as determined by the adiabat. (As explained in Ch. 3 and 4 of my book, Principles of Planetary Climate).  

Please do not expect any more correspondence from me. I’ve already put more time into this than I can spare.

My answer to Pierrehumbert:

OK Raymond you refer to Einstein, but he did not give any proof of Planck's and Stefan-Boltzmann's laws (including back radiation). You say: A body with temperature radiates. I ask you: Radiates into what? Into an environment at 0 K? Into an environment at higher temperature? After insisting heavily that "back radiation" is real, you take a step back and say that after all the "greenhouse effect", does not depend on any "back radiation". What is the logic? 

Raymond, what you are saying does not make any sense, and I understand that you cannot continue the discussion. But you cannot stop the world nor scientific evidence:  "Back radiation" is non-physical illusion. How come you invest all your money into such a hopeless non-physical idea, which not even Einstein did embrace?  

It may well be that we will meet in person in connection with the new statement on the scientific basis of climate change to be issued by KVA/Henning Rodhe, and then the discussion will continue.  

Sincerely, Claes


Pierrehumbert does not answer my question. The only reference given is to the blogosphere http://scienceofdoom.com/roadmap/back-radiation/. The derivation of Planck's law by statistical physics says nothing about back radiation. I think that it is possible to properly understand the meaning of a mathematical theorem only if you know a proof and a physical law only if you know the arguments behind the law. To read a physical law like a lawyer reads a judiciary law seeking the meaning by inspecting the symbolic wording, can easily lead to misinterpretation confusing symbols with physics. This mistake is done in the case of "back radiation". A related logical fallacy is to say that if Einstein's or Planck's proofs say nothing about the non-existence of "back radiation" as physical reality, then it follows that "back radiation" is a real phenomenon. This is an elementary form of "argumentum ad ignorantiam"; the less you know, the more sure you can be that you are right.

PS See later post on phlogiston theory and "back radiation".

20 kommentarer:

  1. OK Raymond you refer to Einstein, but he did not give any proof of Planck's and Stefan-Boltzmann's laws.

    That is not true. Einstein wrote papers in 1916-1917 addressing Planck's law.

    Have a look at "On the Quantum Theory of Radiation":


  2. Yes he did, but the proof is unphysical formal and says nothing about back radiation. The physics present in Planck's proof was eliminated by Einstein.

  3. Einstein states in the article: "The weakness of the theory lies on the one hand in the fact that it does not get us any closer to making a connection with wave theory..."
    Yes, that is the essence of my criticism: Radiative heat transfer is an electromagnetic wave phenomenon and if you do not consider the wave character, then you miss the physics.

  4. Yes, that is the essence of my criticism: Radiative heat transfer is an electromagnetic wave phenomenon and if you do not consider the wave character, then you miss the physics.

    This is a non-issue. That was sorted out with the birth of quantum fields, then you have both wave character and quantization.

  5. I don't understand what you mean with "proof" of physical laws. For me, proof of laws is for pure math. But here we focus on physics, where laws can be verified by observations. Claes, please tell me why you don't believe in the observations Mr Pierrehumbert refer to.

  6. You are fooling yourself: QED says nothing about blackbody radiation. I ask where in Planck's proof of his law "back radiation" is introduced? If it is not the proof, it is a free invention without connection to Planck's law.

  7. I don't understand what you mean with "proof" of physical laws.

    I agree, "proof" is a very strange word in this setting.

    Formal manipulation in themselves are of course meaningless in connection with physics if no feedback with measurements and observations are done. Definitions are by themselves meaningless. Feynman touches upon this many times in his Lectures on Physics (http://www.feynmanlectures.caltech.edu).

    That's why Einsteins papers are so important, they introduced the Einstein coefficients that are rates that actually can be measured, representing real physics.

  8. The "proof" of a physical law consists of the arguments motivating or justifying the law. A physical law without any form of justification is simply a mystery and not physics.

  9. An example is Newton's inverse square law which results from a gravitational potential connected through Poisson's equation to mass density, and Poisson's equation can be motivated by a conservation principle and conservation can be motivated by non-loss.

  10. And why are you defining something you call "non-loss"? You are dangerously close to something circular here now...

    From chapter 3-7 (How did it get that way?) in the first The Feynman Lectures of Physics (http://www.feynmanlectures.caltech.edu/I_03.html#Ch3-S7)

    There is another kind of problem [...] which does not exist in physics; we might call it, for lack of a better term, the historical question. How did it get that way? [...]

    There is no historical question being studied in physics at the present time. We do not have a question, “Here are the laws of physics, how did they get that way?”

  11. I am just helping you to understand some of the essence of physics based on math.

  12. What is non-loss here?

  13. It is just another way of saying that if nothing is lost then all is conserved. Conservation of energy as the 1st law of thermodynamics says that energy cannot simply disappear. Conservation of mass means that mass cannot be lost. What is lost at one place shows up at another. Nothing strange. Just rational.

  14. Claes, please answer my question (13 mars 2015 09:17). Why don't you believe in the observations Mr Pierrehumbert refer to.

  15. Which observations? DLR by pyrgeometer? See previous posts on these topics.

  16. Hi

    I got curious here and thought I'd take a look on you black body radiation model that you present in "Mathematical Physics of Blackbody Radiation". The equations I address here are eq (8.1) p. 41 and eq (11.2) p. 64.

    If we first look at your earlier comment.

    "I am just helping you to understand some of the essence of physics based on math."

    Ok, so if we try and understand the essence based on physics, instead of math ;-) Let's see what we can find.

    If we have a real crystalline material, a typical lattice has dominant force constants around 1 eV/Å^2 for nearest neighbours. The oscillations in a lattice is with certainty bounded by a maximal frequency of 100THz (typically 1/10th of this).

    Using this to estimate the ratio between the Abraham-Lorentz force and the forces between atoms in the lattice, we get that the lattice forces are about 10^15 times stronger than the Abraham-Lorentz forces in the present situation.

    How on earth could the term including the gamma times jerk, in equations (8.1) and (11.2), influence the dynamics of those equations at all, as long as we are dealing with a realistic material that physically exist? The force related to that term is out-powered some 15 orders of magnitudes in strength.

  17. This is a good question. The answer is probably connected to the fact that the size of the constant gamma of the jerk force is irrelevant as long as it is small. The constant gamma nominally appears in Planck'a law but it is normalized by the temperature scale set for bodies with different composition by radiative equilibrium. I have writt

  18. Continued: I have written about this on the web site Computational Blackbody Radiation under Universality.

  19. "The answer is probably connected to the fact that the size of the constant gamma of the jerk force is irrelevant as long as it is small."

    Excuse me for saying so, but that sounds very vague. Have you showed that the jerk term is relevant at all? I mean, if not, the model may not be valid other then producing a qualitative argument that there should be a cut off. But that is already evident from assuming an existing material, the phonon dispersion is necessarily bounded since the lattice has a limitation in how it can vibrate.

    I also have a preliminary question regarding p. 63, how do you quantify coherent emission? Thermal sources are inherently incoherent with very low coordination both in time and space.

  20. No, it is not vauge: it is an important result of the analysis that the absolute size of gamma is irrelevant as long as it is small and positive. The positivity as absolutely essential as well as the near resonance aspect: perfect resonance with gamma = 0 gives blow-up. The essence is near resonance at small damping where the near resonance is up to 1 Hz and the size of the small damping is cancelled in radiative equilibrium of different bodies with different gamma = universality.