Friday, June 3, 2011

Radiation physics constraints on global warming: CO2 increase has little effect

By Denis G. Rancourt
Former physics professor, University of Ottawa, Ottawa, Canada.

Abstract – I describe the basic physics of planetary radiation balance and surface temperature, using the simplest model possible that is sufficiently realistic for correct evaluations of predicted surface temperature response sensitivities to the key Earth parameters. The model is constrained by satellite absolute integrated intensity and spectroscopic measurements and the known longwave absorption cross section of CO2 gas. I show the predicted Earth temperature for zero atmospheric resonant absorption of longwave radiation (no greenhouse effect in the otherwise identical atmosphere) to be –46 C, not –19 C as often wrongly stated. Also, the net warming effect from the atmosphere, including all atmospheric processes (not just greenhouse forcing), without changing anything else (except to add the removed atmosphere) is +18 C, not the incorrect textbook value of +33 C. The double-layer atmosphere model with no free parameters provides: (a) a mean Earth surface temperature of +17 C, (b) a post-industrial warming due only to CO2 increase of δT = 0.4 C, (c) a temperature increase from doubling the present CO2 concentration alone (to 780 ppmv CO2; without water vapour feedback) equal to δT = 1.4 C, and (d) surface temperature response sensitivities that are approximately two orders of magnitude greater for solar irradiance and for planetary shortwave albedo and longwave emissivity than for the atmospheric greenhouse effect from CO2. All the model predictions robustly follow from the straightforward underlying assumptions without any need for elaborate global circulation models. The same longwave optical saturation that provides such a large radiative warming of the planet surface also ensures that the warming effect from increasing CO2 concentration is minimal. I conclude with suggested implications regarding warming alarmism, errors by sceptics, research funding, and scientific ignorance regarding climate feedbacks.

Historically, the greatest ability of the physicist has been to perform simple calculations that capture the essential features of a physical phenomenon in order to correctly elucidate the underlying principal causes. This is the ultimate “What is going on?” challenge.

Too many practicing physicists, in the many areas where physics is applied, have lost or never had this ability. Instead, they have been incorporated into the enterprise of using computers to simulate reality using questionable selections of approximate or invalid algorithms in large simulation programs.

These programs develop lives of their own, as careers and reputations are invested in their incremental development and in their predictions. A pathological optimism envelops the practitioners with the illusion that their algorithms capture complex features in some “average” or “effective” way and efforts are made to demonstrate this in so-called “validation” exercises rather than perform simple calculations that would demonstrate the algorithms themselves to be wrong for the intended application.

Physicists have largely abandoned their gadfly role of fundamentally challenging broad interpretive schemes in order to serve and benefit from career-enhancing collective enterprises, often dressed in elaborate conceptual edifices and often supported by computer simulations.

I believe this situation is playing itself out today in climate modelling science. As a physicist, if on close examination I can’t understand what the CO2 warming alarmism is about and I can’t get any of my colleagues to explain it – without computer-black-box magic, in published papers or elsewhere – then I am not going to believe it.

At its core, planetary surface temperature is a macroscopic radiation balance phenomenon that has been understood for one hundred years or so. If global warming alarmism is justified then it must be possible to explain why it is justified in simple terms and without appealing to faith or authority for any essential point in the argument.

I’ve tried to do this, as honestly and openly as possible, and I have asked my peers to find any errors. I believe the present article to be error-free and to conclusively show that we should not be focussed on CO2 if we are concerned about the planet’s surface temperature. I am additionally of the opinion that we should not be concerned about the planet’s surface temperature.

Regarding the sceptic-warmist debate, my conclusion is: The sceptics say many incorrect things but they are right whereas the warmists say many correct things but they are wrong. The skeptics appear to be motivated by skepticism whereas the warmists appear to be motivated by conformism. The incorrect things have been used to discredit the sceptics and the correct things have been used to mask a lie.

MORE... (alternative link)(With small clarifications added and some typos corrected on December 3, 2011.)


Anonymous said...

It seems that the warmists have been increasingly saying wrong things. The recent hype over tornadoes is among them. At a campus debate where I work (Triton College at the planetarium theater named for the last astronaut to walk on the moon) back in 2007, I asked the following question:
Given the publicity generated by Hurricane Katrina and the record-setting 2005 Atlantic Basin tropical storm season, why was there no attention given to the unprecedented tornado season that same year despite the alleged increase in intensity and frequency? My opponents (the Geography Department) were completely perplexed. Of course they had no chance - here's why:

With the exception of 1992 (another "hot" year) for the first time since 1875 there were no fatalities from tornadoes in the months of April and May anywhere in the country.

There were no deaths in June or July of that year either, meaning that there had been no precedent for zero fatalities anywhere in the nation through the months of April, May, June and July since 1854.

Despite always having tornadoes in the month of April since its inception in the late 1800's, Oklahoma had not one single tornado in April of 2005.

For the first time since assessments of this nature have been applied (including retroactively), there was no tornado of Fujita Force scale five for a six year period - in fact, it would take another two years for another to occur (Greensburg), meaning that the record span for highest-end events was stretched from five years to eight years.

There was a multitude of other reasons, but I suspect the reader may tire of them. However, this can serve to underscore the fatal flaw in the idea of invoking hurricanes as well.

Certainly there had been no lack of hurricanes, including very strong events, up until at least 2009. But last year presented a dilemma for the warmists. Despite the relative intensity of Atlantic Basin activity, there was only one U.S. landfalling event and the earth as a whole had a record setting low value for overall tropical storm intensity.

Daniel P. Joyce
Cernan Earth & Space Center

Anonymous said...

I think any model that does not include simple fact that Earth is essentially one large piece of molten rock (at very high temperature) should go to rubbish bin, regardless who put it together -- skeptic or optimist.

Denis Rancourt said...


oops, did i forget to say that?

Here goes: Below the surface, the Earth is essentially one large piece of molten rock (at very high temperature).

Indeed, the inner core of Earth is believed to be molten iron with some nickel. Iron meteorites can be used to size the primordial planets based on their microstructures that allow estimates of core cooling rates during planet maturation...

I could go on and on because I have written papers about it. Anyway, so yes the Earth has a hot core but that's not why the beach sand burns your feet!


Anonymous said...

You claim that "The sceptics say many incorrect things but they are right whereas the warmists say many correct things but they are wrong."

Sorry, but I call BS. The "concensus" warmers have lied from Day 1 and that is why I'm a sceptic. Tell me, what data and conclusions in the IPCC summary reports are accurate? From sea levels rising, to the Brazxilian rain forests, to the Himalayan glaciers, to SCRU-U Dr. Phil "no you can't see my data / oops the only data we have has been adjusted and the originals deleted" Jones, to the reports written by green activists and referenced in the IPCC AR reports. All BS. The hockey stick graph and Hide-the-decline are obvious frauds. (

Let me fix your sentence for you.

"the warmists say many POLITICALLY correct things but they are wrong."

Anonymous said...


Can you go thruough the numbers for me? You say the starting temp without atmosphere is - 46C and the atmosphere generates +18C. Wouldn't that mean an average temp of the earth at -28C? You say you calculate the temp at +17C so that would mean an additional +45C is needed to get from -28 to + 45.

I'm missing something so please let me know where it is.


Denis Rancourt said...


Just read it again. Maybe more slowly?

-- atmosphere effect not equal to greenhouse effect in atmosphere...

It is explained as clearly as I can spell it out. I don't see how I could be more explicit than what I wrote in the article.

magellan said...

Your paper was mentioned at Dr. Roy Spencer's weblog.

A poster responded:
Denis Rancourt intends to do the right thing by using the year-round variation of the insolation, but in the event he does not use it correctly. He assumes that the cyclic change in insolation does not cause a corresponding cyclic change in the behaviour of the atmosphere. So what he does here cannot give a useful answer. Christopher Game

Glenn Tamblyn said...


I have read vsn 3 of your paper. I have the following comments.

I tend to agree roughly with your figure of –46C for a GH free atmosphere – I have calculated something similar. Two caveats however. The values for Thermals (TH) and EvapoTranspiration (ET) would be much much lower in such a cold world. So Surface Radiation would need to be a larger proportion of total heat loss from the surface so surface temps might be higher as a result. Working against this however would be the massive increase in ice cover producing a substantial cooling due to increased reflection. I believe this Albedo change would be the stronger factor so –46, or even colder is a reasonable figure. So the GH Effect really does have a huge impact when all the feedbacks from it are followed through.

However I would take some issue with your criticism of the 33C cooler figure. I have never heard it referred to as the temperature of the earth without an atmosphere. Rather it is described as being the temperature of the earth without the LW Absorption GH properties of the GH gases. And it is still a simplification based just on albedo and energy balance. It doesn’t take into account the fact that some of the other energy entering the atmosphere – your B, C & D flows – will be returned to the Earth as back radiation. Because the GH Effect is made up of three aspects. 1. Absorption of LW by GH gases in the atmosphere. 2. Re-radiation of ALL energy added to the atmosphere. 3. That the re-radiation is not symmetrical to Earth vs to Space due to the presence of GH gases. Aspect 2 still functions in an atmosphere even without the absorptive properties of the GH gases.

Your criticism raises an important point in how AGW science is often criticised. Much of the criticism is not directly of the science. Rather it is often criticism of the REPORTING AND COMMENTARY on the science for not being completely accurate. So, Al Gore didn’t mention 800 year lags, 33 C is used as an illustrative figure for how much the GH Effect contributes, etc. The problem with this type of criticism is that often it is objecting to the fact that indirect, and ultimately simplified, descriptions of the science, aimed at a, to varying degrees, lay audience is not completely accurate – of course its not, it’s a simplification – and that therefore the underlying science is at fault. That is like criticising a science teacher for teaching junior grade science students that electrons ‘orbit’ around the nucleus of an atom; Obviously in inaccurate statement. But good enough for that audience. Those junior students aren’t ready for calculating the shape of electron orbitals, the exclusion principle or probability density functions. But if I felt that that this simple level description is at fault, the appropriate course for me to take is not to say – ‘No, that is BS, these scientists don’t know anything!’. It is to dig deeper to the next level of complexity, past the simplifications, and the next until either my criticisms are no longer justified OR I have reached the deepest level of the science. If I still think there is a problem, I am then justified in raising an objection. Don’t criticise the simplification; criticise the unsimplified version.

Much of AGW scepticism in my observation is about poor quality doubts about the simplified commentary on the science, rather than the rigorous doubts about the real science itself.

To your revised 2-layer atmosphere model. This is closer to reality and as you point out, extending it to an n-layer model would bring you closer still. And as a simplistic reality check on the detailed models it ain’t bad. However there are 2 assumptions you make in your model that would need to be addressed to extend it further.

Glenn Tamblyn said...

A larger issue is your use of a single Aat value for atmospheric ‘albedo’ (actually transmissivity) in each layer. Since the value you are using is based on total transmissivity for the atmosphere as observed from space this value overlooks the way this value actually varies with Altitude (Temperature, Pressure etc) and Frequency. At the central absorption frequency for CO2 for example it is fully saturated and transmissivity is actually 0. What is observed in space at this frequency is actually re-radiation from higher in the atmosphere. Since the degree of saturation at any frequency will vary with altitude and also temperature and pressure broadening of absorption lines will change as well, a realistic analysis needs to allow for the properties of each gas varying with altitude and frequency. So a realistic n-layer model needs a large number of layers, it needs to define the temperature & pressure profile through the air column and then needs to evaluate each absorption line separately based on the properties of each gas at that T & P. At that point this is no longer soluble as a set of simultaneous equations. Thus the need to use some form of full radiative transfer program to do this line-by-line evaluation. So your simplified model has probably gone about as far as it can.

And you have reproduced an approximate value for warming based on doubling CO2 this that isn’t far from the accepted value. In fact your value is somewhat higher at 1.4 C whereas the value usually cited is around 1.2 C.

I am not certain that this result directly reflects the accuracy of your simple model or whether the several different sources of inaccuracy – not enough layers, Aat fixed, B,C,D evenly split, Aat not varying with frequency, have not just fortuitously combined to give an approximately correct value. At the end of the day that is all rather moot unless you plan to develop your model further and compare its results to those produced by the major Radiative Transfer Codes.

However, your next step, trying to show that the effect of CO2 that you have just demonstrated is not significant is risible. The idea of using comparisons based on the 6.7% variation in the ‘solar constant’ is quite bizarre. This statement “The radiation balance steady state temperature of Earth’s surface is approximately two orders of magnitude more sensitive to changes in solar constant and planetary albedo than
to changes of atmospheric concentration of greenhouse effect CO2.”!

What exactly is your definition of Steady State? Over what time period? The steady state variation of the ‘solar constant’ over the course of a year is 0%. If we look at how the solar constant varies between 1 July 2011 and 1 July 2031 the variation is nearly 0%. The only variation we see in the solar constant that impacts on climate is the 11 year solar cycle. And this only varies by around +/- 0.1%.

So what do you mean by “Regarding the above relative sensitivities, … that more research funding should go into studying solar irradiation variations”?

If you mean the 11 year solar cycle then there is already an entire discipline in science devoted to studying this +/- 0.1% variation. However if you mean the 6.7% variation you describe, then the research program into why this occurs was completed a long time ago. The Sun’s output doesn’t vary by 6.7%. The Earths orbit VARIES over the seasons since it is ELLIPTICAL! And the ‘research program’ into this was completed several centuries ago. By Johannes Kepler!

So this argument is a giant red herring. And as a professor of Physics I would have expected you would understand this.

Glenn Tamblyn said...

Next, Albedo change. What will change the surface albedo of the Earth? Well since 70% of the Earth is oceans, and water’s albedo doesn’t vary much at all, the only meaningful source of albedo change there is changes in ice cover. And what drives changes in ice cover – temperature change. Something has to produce a temperature change before ice change induced albedo change can contribute more change.

What changes albedo on land? Well, again changes in ice/snow cover are the big impacts. Lesser impacts would be changes in vegetation type, desertification, soil moisture. And what drives these changes? Temperature and moisture changes again.

So what can lead to albedo change? Something else that changes temperature and moisture levels.

The other form of albedo change is changes in cloud cover and type. What direction the net effect of this change will take may be harder to discern but again what is needed to trigger such a change is a change in temperature and moisture.

So all the albedo change effects that might occur have to be caused by something else. Then the Albedo change might dampen or amplify the change. But Albedo change needs to be instigated by some other change. So we can validly say that this other cause leads to the Albedo change. So the relative magnitude of Albedo change to CO2 change isn’t the issue. The question is whether the CO2 change is the instigator of the Albedo change. If it is, then CO2 change LEADS to Albedo change – CAUSES IT.

Now to your final comments about water vapour feedbacks and clouds. You seem to blur these two together somewhat. Lets start with the first feedback – increases in water vapour content in the atmosphere. As is well accepted (and which Peter Laux has acerbically noted), water vapour is a larger contributor to the existing GH effect than CO2 – approximately 2.6 times. Surely this is the first piece of evidence in support of the Water Vapour +ve feedback argument. IT ALREADY IS.

Next is the question of what will happen to water vapour levels in the atmosphere if the atmosphere has warmed. Consider the following. Water is evaporated from the surface – overwhelmingly from the oceans - moves into the atmosphere, and then returns to the surface as precipitation. Averaged over the planet and the seasons, evaporation must equal precipitation or the atmosphere would either dry out or fill with water. So what will the water vapour content of the atmosphere be? That level which is sufficient to produce enough precipitation to match evaporation rates. Water vapour levels in the atmosphere will rise to the level needed to cause this.

What determines the amount of precipitation that can occur from a parcel of air?

It’s Relative Humidity. Until the RH is high enough to start producing condensation precipitation cannot occur. This does not mean it has to reach the Dew Point since various particles that can act as Cloud Condensation Nuclei (CCN) can cause condensation below the Dew Point. But it is still RH dependent. Just watch a light cloud swirling. Wisps of cloud – condensed water – appear and disappear as the turbulence in the air produces momentary regions of higher and lower temperature and pressure; all changing the saturation point of the air and thus the relative humidity.

So what happens if we increase the average temperature of a parcel of air while the amount of water vapour remains constant? The relative humidity drops! So the amount of condensation, and thus precipitation, that is possible from that parcel of air drops. As long as it’s temp’ is higher, precipitation is lowered. The only way we can return to the same level of precipitation from that parcel of air is to increase the total amount of water vapour in the parcel until the relative humidity is restored. So increasing the aggregate temperature of the atmosphere MUST increase the water vapour content of the atmosphere. Until it is, precipitation rates would fall.And evaporation would continue charging the air with more water vapour.

Glenn Tamblyn said...

And since the GH effects of water vapour are based on total mass of H2O in the atmosphere rather than Relative Humidity, this MUST lead to a water vapour feedback. It is physically impossible for it not to. The question of how much feedback depends of how this additional water vapour is distributed around the world and the radiative physics of its impact.

Glenn Tamblyn said...

But higher temps due to CO2, Methane etc MUST lead to SOME H2O feedback.

So in the event of a temperature changes due to CO2 etc, Water vapour MUST magnify this. And this agrees with observations. Satellite measurements of water vapour quantities in the atmosphere have reported increase of ~ 0.41 kg/M2 per decade since 1988.

So the water vapour feedback is real and observed. This results in a temperature forcing 2-3 times greater than CO2 alone.

Next we come to the impact on cloud cover. If we have increased Temps, must we automatically have increased cloud cover, averaged over the planet? No. We may, but we don’t HAVE to. If Temps have increased but evaporation rates haven’t, then ultimately cloud cover needs to remain at a level at which precipitation rates match evaporation rates. So it doesn’t automatically have to increase. However, if evaporation rates increase then cloud cover needs to adjust to produce increased precipitation to balance this. So the key question is what will happen to evaporation?

If temps increase at sea level then relative humidity will drop. This will enhance evaporation, assuming nothing impacts on solar radiation levels. It actually isn’t temperature that drives evaporation directly. It is relative humidity. The closer to saturation the air is, the less evaporation can occur. Temperature increases evaporation by lowering the relative humidity of the air, enhancing evaporation. So if temp increase is wide spread enough then enhanced evaporation can occur. And resulting in enhanced precipitation.

So what happens to cloud cover. The assumption Denis has made in his paper is that since clouds are a significant part of the planets Albedo, more clouds will cause more reflection and thus have a cooling effect.

However, depending on the type and altitude of the clouds they can have a warming OR cooling effect. Low level clouds tend to have a cooling effect while high level clouds tend to have a warming effect. So in order to CHANGE the impact that clouds have we would need to see a change in the RATIO of cloud types, not just an increase in the total amount of them.

If low level clouds increase disproportionately then this will be a cooling effect. If high altitude clouds increase disproportionately then this will be a warming effect. If the proportions don’t change then the net effect of clouds doesn’t change, even if the total volume of clouds increases.

So we already have a warming due to CO2 (and the other GH Gases), a +ve feedback due to H2O. And then there is the question of whether cloud changes will counteract this (or possibly add to it). This is obviously complex, but for the combined effects of CO2, the other GH gases and water vapour to be negligible, we actually NEED clouds to provide a negative feedback. Clouds just having a neutral impact means we have real warming. +ve feedback adds more.

So the key question is the change in distribution between high & low clouds.

There is no simple answer to this question, comparable to Denis’s simple atmosphere model. What we can say is that there are three possible scenarios: Proportionally more high clouds, proportionally more low clouds, or the same ratio.

Only ONE of these three scenarios provides a negative feedback to counter the warming effect of GH gases & water vapour. We either have less warming, the same warming, or more warming.

How do we evaluate this?

Well we have a reasonable theoretical understanding of the behaviour of clouds, but not exact. We also have decades of observational data on actual cloud behaviour – what sorts of clouds you get under what circumstances. So a starting point in evaluating this is simply to say – in this type of atmospheric conditions - pressure, temperature, moisture content, wind speed, wind shear, latitude, aerosol content etc - what does our observational database says you will tend to get as a mix of cloud types and quantities there.

Glenn Tamblyn said...

This is the basis of what climate models do when modelling clouds. What sort of clouds do we KNOW we get in this situation? This is the basis for how we model the effects of clouds.

So the models predict a certain type of cloud response based on observed cloud behaviour. And this predicts that clouds won’t be a major negative feedback – with a margin of error in that prediction. But don’t forget, in order to say that AGW will be minor we actually NEED clouds to be a negative feedback. Just being neutral means AGW is real. Negative Cloud feedback is our last Get-Out-Of-Jail-Free card. And none of the data/analysis to date suggests a strong negative cloud feedback. It is only a few people with outlier views such as Lindzen who suggest otherwise. And Lindzen’s IRIS hypothesis hasn’t received much observational support to date.

What other sources of insight could we use to evaluate the Cloud feedback question?

Well the other major area is studies into the so-called Climate Sensitivity. Literally how much the climate will change due to a forcing from a doubling of CO2 .

There are a range of sources of information that have been used to try and evaluate CS. Climate models obviously. But also climate responses to volcanic eruptions, analyses of current climate state, recent climatic changes (1000 year type time scales), analyses of changes over the 100,000 year glacial cycle, modelling applied to the glacial cycle, deep paleoclimate analyses (multi-million year time scales). These various studies have returned values within a fairly common range 2.0 to 4.5 with the possibility of higher values.

If cloud feedbacks were going to have a significant negative feedback in our current climate, why didn’t they have a significant negative impact in other climates?

Since these CS studies encapsulate the impact of water vapour feedbacks, cloud feedbacks, and albedo feedbacks over varying time scales, and are producing consistent results using a wide variety of analytical techniques, looking at a wide range of time-scales, it is reasonable to conclude that the 3 combined feedbacks will produce understandable and comparable impacts today. Otherwise we are saying that there is something special, exceptional and unique about the current configuration of our climate that bears no relation to any past climate. And that is a huge claim to make!

So for all these reasons I feel that your arguments and analysis of the immediate impact of CO2 is reasonably valid, but your argument trying to then justify a low contribution from CO2 relative to other factors not INSTIGATED by CO2 are faulty.

CO2 may not be the biggest single factor in the climate. But it is certainly the biggest control knob, controlling the other factors.

Denis Rancourt said...


I agree that CO2 "is certainly the biggest control knob", but not in the sense that you intend it.


Thank you for pointing this out. The comment on the Spencer weblog seems to be about something I never said or implied? Probably from a parallel universe where they use a faulty translation algorithm?

Denis Rancourt said...

The PDF file of this article "Radiation Physics ... CO2 increase has little effect" has to date been download 237 times from

blouis79 said...

BTW, it is plain that if one considers the "earth" within a vacuum as a "body" in radiative equilibrium with the sun, then the correct view of earth is the earth as seen from space and that the "surface" in equilibrium is the integrated visual/radiating surface including atmosphere.

socold said...

don't we know from GCMs, and even radiative-convective models, that doubling CO2 does have a big effect on temperature?

You say it's saturated. How come none of those models find that?

And why should we take your model for the answer and not those more complete ones?

Jose_X said...

Denis, I have gotten a bit distracted in the "critique" I was writing for this "December 3, 2011, pdf version".

In case, I don't get back to this soon, let me just point out a significant problem I found (and I have to recheck the math).

eq.23 is derived using a bit of missing information. That missing information is the relationship between delta of alpha (change in flux at ground) and delta of a_at (change in albedo of atmosphere). "Delta" is in the sense of eq.19,20,21,22,and 23.

Now, it appears you derived some of these aforementioned equations by taking "differentials". So when I did so to eq.13, I got:

δ(alpha)/alpha = -4/((1+a_at)(3-a_at)) δ(a_at) ...eq.23-pre

Then when I use this to construct δ(Tα) of eq.23, I get:

δ(Tα) = (1/4) Tα [(4)/((1+a_at)(3-a_at))] [ 2 ln(σm/σe) ]^-1 Fco2 δ(Cco2)/Cco2 ... eq.23b

You can note that this eq.23b is not the same as your eq.23 for δ(Tα). The difference is the part that matches δ(alpha) to δ(a_at).. which you did not post in the paper and which I had to guess at by deriving from eq.13.

Now, I have various criticisms before we get to this final point, but accepting everything in the paper up to this point and then substituting the values as you did: Tα = 290 K, aat = 0.10, σm/σe ~ 104, and Fco2 ~ 0.26, δ(Cco2)/Cco2 = 1

instead of getting 1.4 K as we get using your eq.23 (actually I got 0.71), I get, using the above eq.23b, a whopping 10.8!

Anyway, I'll try to do a better write-up later, but you might want to check on this, as I had been able to derive just about all your eq.xx except this eq.23, which happens to be the one with the results upon which the majority of the conclusion of the paper is based.

I hope this comment is useful.

Anonymous said...

You are an utter and complete fraud. You claim to be a "former professor." I doubt that. Your PDF article from 2011 is full of un-substantiated assertions and ad-hominem attacks. I can see why you "published" it on your blog. No real journal, and even no conference-proceedings journal, would ever publish such purely non-scientific crap.