By Christopher Monckton of Brenchley
Joel Shore, who has been questioning my climate-sensitivity calculations, just as a good skeptic should, has kindly provided at my request a reference to a paper by Dr. Andrew Lacis and others at the Goddard Institute of Space Studies to support his assertion that CO2 exercises about 75% of the radiative forcings from all greenhouse gases, because water vapor, the most significant greenhouse gas because of its high concentration in the atmosphere, condenses out rapidly, while the non-condensing gases, such as CO2, linger for years.
Dr. Lacis writes in a commentary on his paper: “While the non-condensing greenhouse gases account for only 25% of the total greenhouse effect, it is these non-condensing GHGs that actually control the strength of the terrestrial greenhouse effect, since the water vapor and cloud feedback contributions are not self-sustaining and, as such, only provide amplification.”
Dr. Lacis’ argument, then, is that the radiative forcing from water vapor should be treated as a feedback, because if all greenhouse gases were removed from the atmosphere most of the water vapor now in the atmosphere would condense or precipitate out within ten years, and within 50 years global temperatures would be some 21 K colder than the present.
I have many concerns about this paper, which – for instance – takes no account of the fact that evaporation from the surface occurs at thrice the rate imagined by computer models (Wentz et al., 2007). So there would be a good deal more water vapor in the atmosphere even without greenhouse gases than the models assume.
The paper also says the atmospheric residence time of CO2 is “measured in thousands of years”. Even the IPCC, prone to exaggeration as it is, puts the residence time at 50-200 years. On notice I can cite three dozen papers dating back to Revelle in the 1950s that find the CO2 residence time to be just seven years, though Professor Lindzen says that for various reasons 40 years is a good central estimate.
Furthermore, it is questionable whether the nakedly political paragraph with which the paper ends should have been included in what is supposed to be an impartial scientific analysis. To assert without evidence that beyond 300-350 ppmv CO2 concentration “dangerous anthropogenic interference in the climate system would exceed the 25% risk tolerance for impending degradation of land and ocean ecosystems, sea-level rise [at just 2 inches per century over the past eight years, according to Envisat], and inevitable disruption of socioeconomic and food-producing infrastructure” is not merely unsupported and accordingly unscientific: it is rankly political.
One realizes that many of the scientists at GISS belong to a particular political faction, and that at least one of them used to make regular and substantial donations to Al Gore’s re-election campaigns, but learned journals are not the place for über-Left politics.
My chief concern, though, is that the central argument in the paper is in effect a petitio principii – a circular and accordingly invalid argument in which one of the premises – that feedbacks are strongly net-positive, greatly amplifying the warming triggered by a radiative forcing – is also the conclusion.
The paper turns out to be based not on measurement, observation and the application of established theory to the results but – you guessed it – on playing with a notorious computer model of the climate: Giss ModelE. The model, in effect, assumes very large net-positive feedbacks for which there is precious little reliable empirical or theoretical evidence.
At the time when Dr. Lacis’ paper was written, ModelE contained “flux adjustments” (in plain English, fudge-factors) amounting to some 50 Watts per square meter, many times the magnitude of the rather small forcing that we are capable of exerting on the climate.
Dr. Lacis says ModelE is rooted in well-understood physical processes. If that were so, one would not expect such large fudge-factors (mentioned and quantified in the model’s operating manual) to be necessary.
Also, one would expect the predictive capacity of this and other models to be a great deal more successful than it has proven to be. As the formidable Dr. John Christy of NASA has written recently, in the satellite era (most of which in any event coincides with the natural warming phase of the Pacific Decadal Oscillation) temperatures have been rising at a rate between a quarter and a half of the rate that models such as ModelE have been predicting.
It will be helpful to introduce a little elementary climatological physics at this point – nothing too difficult (otherwise I wouldn’t understand it). I propose to apply the IPCC/GISS central estimates of forcing, feedbacks, and warming to what has actually been observed or inferred in the period since 1750.
Let us start with the forcings. Dr. Blasing and his colleagues at the Carbon Dioxide Information and Analysis Center have recently determined that total greenhouse-gas forcings since 1750 are 3.1 Watts per square meter.
From this value, using the IPCC’s table of forcings, we must deduct 35%, or 1.1 Watts per square meter, to allow for negative anthropogenic forcings, notably the particles of soot that act as tiny parasols sheltering us from the Sun. Net anthropogenic forcings since 1750, therefore, are 2 Watts per square meter.
We multiply 2 Watts per square meter by the pre-feedback climate-sensitivity parameter 0.313 Kelvin per Watt per square meter, so as to obtain warming of 0.6 K before any feedbacks have operated.
Next, we apply the IPCC’s implicit centennial-scale feedback factor 1.6 (not the equilibrium factor 2.8, because equilibrium is thousands of years off: Solomon et al., 2009).
Accordingly, after all feedbacks over the period have operated, a central estimate of the warming predicted by ModelE and other models favored by the IPCC is 1.0 K.
We verify that the centennial-scale feedback factor 1.6, implicit rather than explicit (like so much else) in the IPCC’s reports, is appropriate by noting that 1 K of warming divided by 2 Watts per square meter of original forcing is 0.5 Kelvin per Watt per square meter, which is indeed the transient-sensitivity parameter for centennial-scale analyses that is implicit (again, not explicit: it’s almost as though They don’t want us to check stuff) in each of the IPCC’s six CO2 emissions scenarios and also in their mean.
Dr. Lacis’ paper is saying, in effect, that 80% of the forcing from all greenhouse gases is attributable to CO2. The IPCC’s current implicit central estimate, again in all six scenarios and in their mean, is in the same ballpark, at 70%.
However, using the IPCC’s own forcing function for CO2, 5.35 times the natural logarithm of (390 ppmv / 280 ppmv), respectively the perturbed and unperturbed concentrations of CO2 over the period of study, is 1.8 Watts per square meter.
Multiply this by the IPCC’s transient-sensitivity factor 0.5 and one gets 0.9 K – which, however, is the whole of the actual warming that has occurred since 1750. What about the 20-30% of warming contributed by the other greenhouse gases? That is an indication that the CO2 forcing may have been somewhat exaggerated.
The IPCC, in its 2007 report, says no more than that between half and all of the warming observed since 1950 (and, in effect, since 1750) is attributable to us. Therefore, 0.45-0.9 K of observed warming is attributable to us. Even taking the higher value, if we use the IPCC/GISS parameter values and methods CO2 accounts not for 70-80% of observed warming over the period but for all of it.
In response to points like this, the usual, tired deus ex machina winched creakingly onstage by the IPCC’s perhaps too-unquestioning adherents is that the missing warming is playing hide-and-seek with us, lurking furtively at the bottom of the oceans waiting to pounce. However, elementary thermodynamic considerations indicate that such notions must be nonsense.
None of this tells us how big feedbacks really are – merely what the IPCC imagines them to be. Unless one posits very high net-positive feedbacks, one cannot create a climate problem. Indeed, even with the unrealistically high feedbacks imagined by the IPCC, there is not a climate problem at all, as I shall now demonstrate.
Though the IPCC at last makes explicit its estimate of the equilibrium climate sensitivity parameter (albeit that it is in a confused footnote on page 631 of the 2007 report), it is not explicit about the transient-sensitivity parameter – and it is the latter, not the former, that will be policy-relevant over the next few centuries.
So, even though we have reason to suspect there is a not insignificant exaggeration of predicted warming inherent in the IPCC’s predictions (or “projections”, as it coyly calls them), and a still greater exaggeration in Giss ModelE, let us apply their central estimates – without argument at this stage – to what is foreseeable this century.
The IPCC tells us that each of the six emissions scenarios is of equal validity. That means we may legitimately average them. Let us do so. Then the CO2 concentration in 2100 will be 712 ppmv compared with 392 ppmv today. So the CO2 forcing will be 5.35 ln(712/392), or 3.2 Watts per square meter, which we divide by 0.75 (the average of the GISS and IPCC estimates of the proportion of total greenhouse forcings represented by CO2) to allow for the other greenhouse gases, making 4.25 Watts per square meter.
We reduce this value by about 35% to allow for negative forcings from our soot-parasols etc., giving 2.75 Watts per square meter of net anthropogenic forcings between now and 2100.
Nest, multiply by the centennial-scale transient-sensitivity parameter 0.5 Kelvin per Watt per square meter. This gives us a reasonable central estimate of the warming to be expected by 2100 if we follow the IPCC’s and GISS’ methods and values every step of the way. And the warming we should expect this century if we do things their way? Well, it’s not quite 1.4 K.
Now we go back to that discrepancy we noted before. The IPCC says that between half and all of the warming since 1950 was our fault, and its methods and parameter values seem to give an exaggeration of some 20-30% even if we assume that all of the warming since 1950 was down to us, and a very much greater exaggeration if only half of the warming was ours.
Allowing for this exaggeration knocks back this century’s anthropogenic warming to not much more than 1 K – about a third of the 3-4 K that we normally hear so much about.
Note how artfully this tripling of the true rate of warming has been achieved, by a series of little exaggerations which, when taken together, amount to a whopper. And it is quite difficult to spot the exaggerations, not only because most of them are not all that great but also because so few of the necessary parameter values to allow anyone to spot what is going on are explicitly stated in the IPCC’s reports.
The Stern Report in 2006 took the IPCC’s central estimate of 3 K warming over the 20th century and said that the cost of not preventing that warming would be 3% of 21st-century GDP. But GDP tends to grow at 3% a year, so, even if the IPCC were right about 3 K of warming, all we’d lose over the whole century, even on Stern’s much-exaggerated costings (he has been roundly criticized for them even in the journal of which he is an editor, World Economics), would be the equivalent of the GDP growth that might be expected to occur in the year 2100 alone. That is all.
To make matters worse, Stern used an artificially low discount rate for inter-generational cost comparison which his office told me at the time was 0.1%. When he was taken apart in the peer-reviewed economic journals for using so low a discount rate, he said the economists who had criticized him were “confused”, and that he had really used 1.4%. William Nordhaus, who has written many reviewed articles critical of Stern, says that it is quite impossible to verify or to replicate any of Stern’s work because so little of the methodology is explicit and available. And how often have we heard that before? It is almost as if They don’t want us to check stuff.
The absolute minimum commercially-appropriate discount rate is equivalent to the minimum real rate of return on capital – i.e. 5%. Let us oblige Stern by assuming that he had used a 1.4% discount rate and not the 0.1% that his office told me of.
Even if the IPCC is right to try to maintain – contrary to the analysis above, indicating 1 K manmade warming this century – that we shall see 3 K warming by 2100 (progress in the first one-ninth of the century: 0 K), the cost of doing nothing about it, discounted at 5% rather than 1.4%, comes down from Stern’s 3% to just 0.5% of global 21st-century GDP.
No surprise, then, that the cost of forestalling 3 K of warming would be at least an order of magnitude greater than the cost of the climate-related damage that might arise if we just did nothing and adapted, as our species does so well.
But if the warming we cause turns out to be just 1 K by 2100, then on most analyses that gentle warming will be not merely harmless but also beneficial. There will be no net cost at all. Far from it: there will be a net economic benefit.
And that, in a nutshell, is why governments should shut down the UNFCCC and the IPCC, cut climate funding by at least nine-tenths, de-fund all but two or three computer models of the climate, and get back to addressing the real problems of the world – such as the impending energy shortage in Britain and the US because the climate-extremists and their artful nonsense have fatally delayed the building of new coal-fired and nuclear-fired power stations that are now urgently needed.
Time to get back down to Earth and use our fossil fuels, shale gas and all, to give electricity to the billions that don’t have it: for that is the fastest way to lift them out of poverty and, in so doing, painlessly to stabilize the world’s population. That would bring real environmental benefits.
And now you know why building many more power stations won’t hurt the climate, and why – even if there was a real risk of 3 K warming this century – it would be many times more cost-effective to adapt to it than to try to stop it.
As they say at Lloyds of London, “If the cost of the premium exceeds the cost of the risk, don’t insure.” And even that apophthegm presupposes that there is a risk – which in this instance there isn’t.
The Viscount Monckton of Brenchley
Part 1 of Sense and Sensitivity can be found here