### My model, used for deception

Filed under: — david @ 4 October 2007

Well, not my model exactly. I developed and host a web interface to the modtran model of atmospheric infrared radiation, an early example of a line-by-line code which I downloaded and use to teach and as part of a textbook. Now David C. Archibald from Summa Development Limited, Perth, WA, Australia claims that my “University of Chicago modtran facility” proves that global warming won’t happen.

Archibald begins by discovering that the IR light flux at the top of the atmosphere is more sensitive to changes in atmosphere CO2 when the concentration of CO2 is lower. This will come as no surprise to regular readers of realclimate who will know that the energy flux scales with the logarithm of CO2. The log dependence is why the climate sensitivity parameter is often posed as a temperature change for doubled CO2 concentration; to first order, a change from 10 ppm to 20 would have about as much climate impact as a change from 1000 to 2000 ppm. So Archibald is right on this score, clearly climate is more sensitive to CO2 when levels are lower. However, I think most climate models are aware that atmospheric CO2 is 380 ppm rather than 10 ppm, and they predict global warming anyway. If we were starting out from 10 ppm, the warming would be even worse.

Archibald then takes an atmospheric increase of 40 ppm which he thinks will happen by the year 2030. I’d have guessed 60 ppm by then at least, the way things are going, but whatever, we’ll see. He uses my setup of modtran to calculate that the IR flux to space would drop by 0.4 Watts / m2 as a result of this 40 ppm. Try it yourself. Run the model once with 375 ppm CO2 and another time with 415 ppm, and compare the Iout values in Watts / m2. The exact number you get depends on humidity, setting, clouds, etc. Formulas given in IPCC would say 0.5 Watts / m2; zeroing out water vapor in modtran gets the IR response up to 0.6 Watts / m2 for the default tropical atmosphere case. At any rate Archibald isn’t wildly off here either.

But then Archibald multiplies the radiative forcing by an absurdly low value of the climate sensitivity parameter. In this case he is using the parameter in units of degrees C per Watt / m2. The two forms of the climate sensitivity parameter that we have discussed here are related by a factor of about 4 Watts / m2 for a doubling of CO2. The value Archibald uses is 0.1 degree C per Watt / m2 which was “demonstrated” in a paper entitled “CO2-induced global warming: a skeptic’s view of potential climate change” by Idso, 1998. Translated, Idso’s climate sensitivity winds up to be 0.4 degrees for doubling CO2. IPCC finds it essentially impossible (yeah, I know, highly unlikely or whatever) that the climate sensitivity could be less than 1.5 degrees C for doubling CO2, and 3 degrees C is a best-guess value.

In the end, Archibald concludes that the warming from the next 40 ppm of CO2 rise (never mind the rest of it) will only be 0.04 degrees C. Archibald’s low-ball estimate of climate change comes not from the modtran model my server ran for him, but from his own low-ball value of the climate sensitivity.

### 172 Responses to “My model, used for deception”

1. 151
Alexander Harvey says:

Timothy,

Subsequent to your comments, I have been having a bit of a look at how LW radiation passes through the atmosphere. This has led me to wonder whether people are using the correct interpretation of Beer’s (Beer-Lambert) law given the geometric considerations. Hopefully they are! A naive application would seriously underestimate the extinction of direct earth surface radiation. The geometric considerations I have considered is that the outbound “beam” is isotropic and hence the extinction is much faster than a simplistic application of that law would predict.

I apologise that I have not as yet found the time to answer ,ost of your comments.

Best Wishes

Alexander Harvey

2. 152

LGL, Well, I’ve got to hand it to you. You have come up with a theory even more half-baked than the the Galactic-Cosmic-Ray claptrap. You really need to do some research on ocean circulation. First the upwelling associated with the ENSO is a relatively small portion of the total deep ocean upwelling. It is not driven mainly by wind, but by differences in density and thermal effects. You can read about this here:
http://en.wikipedia.org/wiki/Thermohaline_circulation

Second, Global Climate Models include ocean circulation. Gavin has written on the subject here on RC.
Now, most important, where are you getting these ideas? If you came up with them yourself, then I have to ask you why you reject a perfectly good model in favor of one you haven’t really worked out. What you have done is try to explain something you don’t understand in terms of something else you don’t understand. That’s not how science works.
If on the other hand, you have read this elsewhere, the question is whether they really say this or whether you have misinterpreted the theory. However, the question still stands: Why reject a perfectly good theory in terms of something less than half worked out?

3. 153
Timothy Chase says:

Alexander Harvey (#151) wrote:

Subsequent to your comments, I have been having a bit of a look at how LW radiation passes through the atmosphere. This has led me to wonder whether people are using the correct interpretation of Beer’s (Beer-Lambert) law given the geometric considerations. Hopefully they are! A naive application would seriously underestimate the extinction of direct earth surface radiation.

Beer’s isn’t supposed to apply that well in the atmosphere I believe due to changes in pressure and temperature with altitude.

Alexander Harvey (#151) wrote:

The geometric considerations I have considered is that the outbound “beam” is isotropic and hence the extinction is much faster than a simplistic application of that law would predict.

I suppose so – but I thought that Beer wasn’t dealing with the isotropic inverse square thing, just a straight beam, wasn’t he?

Alexander Harvey (#151) wrote:

I apologise that I have not as yet found the time to answer most of your comments.

With regard to Kirchoff’s, he derived it from thermodynamic equilibrium but it has shown that the law applies quite well within LTE. Doesn’t depend upon radiation equilibrium.

The thought is that one derives Kirchoff’s law:

Emission over absorption is…

εν(T)/κν(T) = Bν(T)

… where εν(T) is the emission coefficient for frequency ν at temperature T, κν(T) the absorption coefficient for frequency ν at temperature T, and Bν(T)

This is done for a thermodynamic equilibrium by means of the thought experiment involving a narrow band passive filter between two chambers. The filter is such that the only radiation between the two chambers is at ν (to ν+dν), but does so without performing any work. (Hence the name.)

Given thermodynamic equilibrium, the chambers are assumed to be the same temperature. Now there can be no net radiation carrying net power as this would result in a rise in temperature in one of the compartments leading to a cooling of the other.

However, the temperature of the radiation field Bν(T) is independent of the materials which the two chambers are made from and the temperatures of those materials, but the emissivity and absorption coefficients depend only upon the materials and their temperatures and are independent of the radiation field. Consequently Kirchoff’s law holds for any emitting and absorbing material so long as it is in thermal equilibrium – without regard to whether it is in equilibrium with the radiation field.

That’s the condition of a Local Thermodynamic Equilibrium. (Of course all bets are off when the two chambers are at different temperatures.)

Anyway, still have more to figure out…

4. 154
Rod B says:

Ray and Hank,

Yes, but as you get to higher energy levels 1) the probability of a particular transition gets a little less, but, more to the point, 2) the now closer spaced lines do in fact form a continuum but this is within a relatively small bandwidth; no where near “continuous” ala Planck function.

I think it is charge acceleration rather than mass movement that generates radiation. But, Hank, your point is well taken. But it is the (odd?) movement/acceleration of a gas molecule, which is not limited in its movement, not its internal jiggling (vibration, rotation, electronic), which is, that generates a continuous Planck function radiation. They just generate teensy tiny amounts if they are sparse and cold.

5. 155
Hank Roberts says:

Gas or liquid or solid, at any temperature, as long as it’s dense enough that interactions modify the movement a solitary atom or molecule can make.

Plasma at high temperatures like the Sun, again because interactions modify the movement a solitary ion or atom or molecule can make.

For a molecule, the bending or rotation or vibration isn’t just in its simple isolated clear form when there are other molecules interacting. Look at how water behaves as a vapor versus as a liquid versus ice — the bending motions are going to be very different when the molecules are interacting. Same for CO2 as a solid or very dense gas, the motions are going to be different than they’d be for a single isolated molecule.

6. 156
Alexander Harvey says:

Timothy,

I think stated, as you have done, in terms of emission and absorption coefficients it is all much clearer. Really all that remains is to relate these terms to the underlying mechanisms. Firstly the Einstein coefficients and then the various forms of spectral line broadening.

Regarding Beer’s Law: Providing you can integrate the absorption coefficient along each path and take account of the densities, pressures, etc. along that path to give the optical depth (d) of that path we have the proportion that is transmitted along the path equalling the exponential of -d

EXP(-d)

One then has to integrate over all the paths, it is this last stage I was referring to.

Geometrically the atmosphere is a thin layer over a large curved surface and it is not a simple matter to perform the integration. At a first approximation I think you get a transmittance equalling

EXP(-d) + d*Ei(-d)

where Ei(x) is the Exponential Integral and is negative for negative x.

This tends to zero much faster than the EXP(-d) term alone.

I think that a naive application of Beer’s Law to the atmosphere whether formally or merely when thinking about it would lead to poor outcomes. That is what I was trying to say.

Which brings me back to the original topic of this thread. It is all too easy for anyone (amateur or professional) to misapply a technique, or a model, or a law, or a result, and come up with misconception.

Personally I should like to see more transparency from all quarters. (Not just the atmosphere ;) )

I think it is important that the details, both method and data, are clearly stated so that they can be checked in every detail. I doubt that peer review stretches to checking the data, identifying the unstated assumptions, checking all the physics and equations and reworking the calculations. I would think that the combined brainpower of the contributors here along with the disciplines we all enjoy would be sufficient to review some of the papers in depth providing the data and methods were provided.

Even the most respected authors allow mistakes to occur in their work and are possibly never the wiser.

There is much that I should like to see gone over with a fine-tooth comb. Not because I expect foul play but that I feel the results are so important that they should be beyond reasonable doubt.

The same is true of models.

I should feel much happier if I knew what assumptions go into MODTRAN or any of the climate models. Are these assumptions written down anywhere?

At this moment I should like to see a list of the assumptions and the formulae that go into the calculation of the forcing due to a doubling of CO2. As I understand it, the figure is based on a lot of assumptions or facts about the state of the atmosphere and the properties of gases. Is this underlying information available? I would be obliged if someone could point me to it.

Just as important is the limits of applicability of laws, methods and values like the CO2 forcing figure.

As I understand it the CO2 forcing value is an ad hoc one. I.E. it applies to the current atmosphere. E.G. it has taken into account current amounts of cloud cover, water vapour, albedo and importantly temperature etc. If so could it be applied to ice age conditions or is a different value used. Are the limits on this values applicability stated.

There are many things I should like to know, and to know in great detail.

I ask this: can anyone shed any light on the applicability of the CO2 forcing value and the assumptions that have gone into its production?

Best Wishes

Alexander Harvey

7. 157

Rod, keep in mind that the blackbody distribution is the equilibrium distribution for a photon gas–it’s what the photon gas will approach provided there is sufficient interaction with the matter in its vicinity. Yes, free charges will radiate if accelerated, but what is doing the accelerating? Also, keep in mind that the energy emitted by an accelerating charged particle also has its own characteristic spectrum (e.g. bremsstrahlung, cerenkov, etc.). The point is that ALL of these processes can push the system toward the blackbody spectrum, and they will keep pushing it until it gets there (or arbitrarily close), because that is the equilibrium condition for the photon gas.
Consider Earth’s atmosphere. To start with, we have more IR photons that would be found at equilibrium for that temperature, so the GHGs absorb IR photons. And preferentially, the relaxation process is via collisions with other molecules, so the gas heats up and the photon gas cools down. If the ghgs srarted out with the same energy as the photon gas, you would also have vibrational modes being excited collisionally and some decaying radiatively–equilibrium, and the number of photons would stay the same, as would the ghg temperature. And if the ghgs had a higher temperature, you’d have net energy flowing from there to the photon gas. This takes place in the absorption bands of the ghgs because that is where the photons can interact. Outside these bands, the atmosphere is inert, and you see the photons radiating from Earth’s surface.

8. 158
lgl says:

Re 150&152#

I’m not rejecting any theory, but the ghg theory is not a perfectly good theory. With ENSO you have a good correlation but not a well understood mechanism. With ghg you have almost no correlation but a understood mechanism. The only correlation you have is that both ghg concentration and temperature is higher now compared to decades ago. I’m just searching for a better correlation. For instance, how can the heat generated by ghg after 1983 have been stored away for 15 years and then suddenly appear in 1998? You probably end up with the ocean currents there too and need to understand those mechanisms to fully explain the ghg theory.
There is also a problem with the arctic amplification. The seaice coverage started to decline way back in the fifties, so it’s more likely a feedback of the solar forcing before 1950 and not a feedback triggered by ghg.
Timothy, I don’t see what you mean. I picked the span from 1975 because it’s the more interesting. If you start from 1950 you will still see a close to perfect correlation, with the decline up to 1975 because in that time span La Ninas were in majority.
And natural variability. So 0.4 oC rise in 8 years is a natural variability but 0.6 in 30 years can not be natural variability?

9. 159
Rod B says:

Hank, I think your 155 is correct. (At least I think I agree, which might not be the same thing…[;-) The internal molecular energies (vibration, rotation, electronic) are still limited by virtue of their quantization. True, those energy levels are far more active if there’s a lot of other molecules in the area that can collide and transfer energy in and out. True, also, that you get magnitudes more molecular activity in solids and liquids that in gases. Do we have a disagreement here?

10. 160
Rod B says:

Ray, I pretty much agree with your 157. I hope I’m not dreaming [;-). One clarification: the example radiation types can also include PVF — plain vanilla flavor. And one nit-pick (to the point of near insignifance): a photon gas has a characteristic temperature, not a “real” temperature, so can’t really “cool down”. Though your point is valid.

11. 161
Hank Roberts says:

Hello, Gavin? Would you ask David Archer to take a look and see if we’re making sense as we seem to be reaching agreement on the words we’re throwing at this question?

[Response: Sorry, I was tuned out for a bit. I think you’re right, that acceleration of the charge by bond vibration drives the absorbtion / emission band, but at high pressure, interactions among molecules tend to broaden the frequencies of light that can be emitted or absorbed. This thread has many fibers in it, however, so I’m not sure if I’m responding to the words that you’re asking about. I should also caveat that I’m an ocean chemist, not an atmospheric radiation physicist. I just posted MODTRAN and use it to teach, at about this level; I’m no authority on the details. David]

12. 162

lgl, do you have any science background? If you don’t have a mechanism, you have precisely butkis! See:
http://www.realclimate.org/index.php/archives/2007/05/fun-with-correlations/

In a system as complicated as Earth’s climate, you of course have to take many factors into account. When you do (for example in global climate models) you get very good agreement with what is occurring. Look at the magnitudes of the energies involved–a simple model like yours does not come close. Again, I ask, where are you getting your information. It sounds vaguely like some of the organic matter a certain Hurricane expert from Colorado has been spewing.

13. 163

Rod B. The temperature of a photon gas is as real as any other temperature thermodynamically–the partial of energy with respect to entropy holding volume and photon number constant (which of course you can’t).

14. 164

lgl, you have also not answered my question as to why you think causality points one way and not the other. Why should the number of El Ninos start changing? What is driving that? I say again, you are explaining something you don’t understand in terms of something else you don’t understand. That is not science.

15. 165

“Bupkes” is the accepted spelling of that particular Yiddishism. :)

16. 166
lgl says:

Ray, I’m obviously not a scientist. I’m getting my information mainly from the GISS web pages, and definitely not from a Hurricane expert from Colorado. To me the correlation between temperatur and ENSO is obvious, when there are more El Ninos than La Ninas (or positive ENSO I guess) the temperature rise rapidly. This is not something I’m making up, the GISS page I linked to clearly shows this.
The mechnism behind I do not know but would very much like to know.
But you have not answered my question.
How can the heat generated by ghg after 1983 have been stored away for 15 years and then suddenly appear in 1998? What mechanism is that in the “perfectly good theory”?
I don’t understand your “why you think causality points one way and not the other”
El Ninos means rise in temp and La Ninas means drop but that can’t be what you refer to.
(And sorry for my bad English, you probably know by now it’s not my first language)

17. 167

Barton,
I stand corrected. Thanks.

18. 168

LGL–I don’t even understand your question. The heat didn’t get stored–it was warming the planet all along–as the temperature rise over that period shows. 1998 was in the middle of a pretty deep El Nino. See:
http://www.globalwarmingart.com/wiki/Image:Instrumental_Temperature_Record_png

Now if you are talking about the seeming hiatus in warming from ~1945 to the mid 70s, that is well understood. The fossil fuels burned in this period generated a lot of aerosols as well as CO2. The aerosols blocked sunlight, so during this period there was little net warming. So what happened during the 1970s? Clean air legislation decreased particulate, sulfate and other aerosols. Since aerosols have a short lifetime in the atmosphere, while CO2 lasts for hundreds of years, the CO2 really kicks in about this time. Global climate models have reproduced this effect, and the eruption of Mt. Pinatubo provided additional constraints.

The issue of causality is important. If as you claim climate change were driven by the relative # of El Nino vs. La Nina, then we have to ask why this changed. There must be some underlying cause. On the other hand, with anthropogenic greenhouse gases, we know what changed–we dumped about 40% more CO2 into the air–we know that has happened because we did it. So if you can’t point to the underlying causes of the changes in the ENSO, isn’t it more likely that climate change could be behind these changes rather than the other way around?
Changes don’t just happen. Energy doesn’t just come from nowhere. The energy flow related to the ENSO does affect the energy in the climate system, but these effects are rather short term. The energies involved aren’t that large. On the other hand, greenhouse gases provide several watts per square meter over the entire surface of Earth.

Other criticism of your theory: Why would it cause night-time temperatures to rise more than daytime temperatures? Why would it cause Winters to become shorter? Why would it have more effect at the poles.

That is why if you don’t understand the mechanism, you have “Bupkes” (as our good friend Barton counsels).

19. 169
lgl says:

Ok, if I’m the only person seeing this correlation we can leave it there. In a few years we will probably see a long lasting negative ENSO, then this will be much debated.
Other criticism. All that you mention here is probably the nature of a warmer planet, no matter cause.

20. 170
Bruce says:

Why is the temperature dropping in the Southern Hemisphere? Why is it approaching only a .2C rise in temps since the 1940s?

Is CO2 dispersed differently throughout the atmosphere?

Why is it still cooler than 1934 in the USA?

21. 171