Guest commentary by Barry R. Bickmore, Brigham Young University
If you look around the websites dedicated to debunking mainstream climate science, it is very common to find Lord Christopher Monckton, 3rd Viscount of Brenchley, cited profusely. Indeed, he has twice testified about climate change before committees of the U.S. Congress, even though he has no formal scientific training. But if he has no training, why has he become so influential among climate change contrarians? After examining a number of his claims, I have concluded that he is influential because he delivers “silver bullets,” i.e., clear, concise, and persuasive arguments. The trouble is his compelling arguments are often constructed using fabricated facts. In other words, he makes it up. (Click here to see a number of examples by John Abraham, here for a few by myself, and here for some by Tim Lambert).
Here I’m going to examine some graphs that Lord Monckton commonly uses to show that the IPCC has incorrectly predicted the recent evolution of global atmospheric CO2 concentration and mean temperature. A number of scientists have already pointed out that Monckton’s plots of “IPCC predictions” don’t correspond to anything the IPCC ever predicted. For example, see comments by Gavin Schmidt (Monckton’s response here,) John Nielsen-Gammon (Monckton’s response here,) and Lucia Liljegren. Monckton is still happily updating and using the same graphs of fabricated data, so why am I bothering to re-open the case?
My aim is to more thoroughly examine how Lord Monckton came up with the data on his graphs, compare it to what the IPCC actually has said, and show exactly where he went wrong, leaving no excuse for anyone to take him seriously about this issue.
Atmospheric CO2 Concentration
By now, everyone who pays any attention knows that CO2 is an important greenhouse gas, and that the recent increase in global average temperature is thought to have been largely due to humans pumping massive amounts of greenhouse gases (especially CO2) into the atmosphere. The IPCC projects future changes in temperature, etc., based on projections of human greenhouse gas emissions. But what if those projections of greenhouse gas emissions are wildly overstated? Lord Monckton often uses graphs like those in Figs. 1 and 2 to illustrate his claim that “Carbon dioxide is accumulating in the air at less than half the rate the UN had imagined.”
Figure 1. Graph of mean atmospheric CO2 concentrations contrasted with Monckton’s version of the IPCC’s “predicted” values over the period from 2000-2100. He wrongly identifies the concentrations as “anomalies.” Taken from the Feb. 2009 edition of Lord Monckton’s “Monthly CO2 Report.”
Figure 2. Graph of mean atmospheric CO2 concentrations contrasted with Monckton’s version of the IPCC’s “predicted” values over the period from Jan. 2000 through Jan. 2009. Taken from the Feb. 2009 edition of Lord Monckton’s “Monthly CO2 Report.”
It should be noted that Lord Monckton faithfully reproduces the global mean sea surface CO2 concentration taken from NOAA, and the light blue trend line he draws through the data appears to be legitimate. Unfortunately, nearly everything else about the graphs is nonsense. Consider the following points that detail the various fantasies Monckton has incorporated into these two graphics.
Lord Monckton claims the light blue areas on his graphs (Figs. 1 and 2) represent the IPCC’s predictions of atmospheric CO2 concentrations.
The IPCC doesn’t make predictions of future atmospheric CO2 concentrations. And even if we ferret out what Lord Monckton actually means by this claim, he still plotted the data incorrectly.
The IPCC doesn’t really make predictions of how atmospheric CO2 will evolve over time. Rather, the IPCC has produced various “emissions scenarios” that represent estimates of how greenhouse gas emissions might evolve if humans follow various paths of economic development and population growth. The IPCC’s report on emissions scenarios states, “Scenarios are images of the future, or alternative futures. They are neither predictions nor forecasts. Rather, each scenario is one alternative image of how the future might unfold.” Lord Monckton explained via e-mail that he based the IPCC prediction curves “on the IPCC’s A2 scenario,which comes closest to actual global CO2 emissions at present” (2). In his “Monthly CO2 Report” he added, “The IPCC’s estimates of growth in atmospheric CO2 concentration are excessive. They assume CO2 concentration will rise exponentially from today’s 385 parts per million to reach 730 to 1020 ppm, central estimate 836 ppm, by 2100,” which is consistent with the A2 scenario. In other words, Monckton has picked one of several scenarios used by the IPCC and misrepresented it as a prediction. This is patently dishonest.
Monckton’s misrepresentation of the IPCC doesn’t end here, however, because he has also botched the details of the A2 scenario. The IPCC emissions scenarios are run through models of the Carbon Cycle to estimate how much of the emitted CO2 might end up in the atmosphere. A representative (i.e., “middle-of-the-road”) atmospheric CO2 concentration curve is then extracted from the Carbon Cycle model output, and fed into the climate models (AOGCMs) the IPCC uses to project possible future climate states. Figure 3 is a graph from the most recent IPCC report that shows the Carbon Cycle model output for the A2 emissions scenario. The red lines are the output from the model runs, and the black line is the “representative” CO2 concentration curve used as input to the climate models. I digitized this graph, as well, and found that the year 2100 values were the same as those cited by Monckton. (Monckton calls the model input the “central estimate.” )
Figure 3. Plot of atmospheric CO2 concentrations projected from 2000-2100 for the A2 emissions scenario, after the emissions were run through an ensemble of Carbon Cycle models. The red lines indicate model output, whereas the black line represents the “representative” response that the IPCC used as input into its ensemble of climate models (AOGCMs). Taken from Fig. 10.20a of IPCC AR4 WG1.
Now consider Figure 4, where I have plotted the A2 model input (black line in Fig. 3), along with the outer bounds of the projected atmospheric CO2 concentrations (outer red lines in Fig. 3). However, I have also plotted Monckton’s Fantasy IPCC predictions in the figure. The first thing to notice here is how badly Monckton’s central tendency fits the actual A2 model input everywhere in between the endpoints. Monckton’s central tendency ALWAYS overestimates the model input except at the endpoints. Furthermore, the lower bound of Monckton’s Fantasy Projections also overestimates the A2 model input before about the year 2030. What appears to have happened is that Lord Monckton chose the correct endpoints at 2100, picked a single endpoint around the year 2000-2002, and then made up some random exponential equations to connect the dots with NO REGARD for whether his lines had anything to do with what the IPCC actually had anywhere between.
Figure 4. Here the black lines represent the actual A2 input to the IPCC climate models (solid) and the upper and lower bounds of the projected CO2 concentrations obtained by running the A2 emissions scenario through an ensemble of Carbon Cycle models. This data was digitized from the graph in Fig. 3, but a table of model input concentrations of CO2 resulting from the different emissions scenarios can be found here. The red lines represent Monckton’s version of the IPCC’s “predicted” CO2 concentrations. The solid red line is his “central tendency”, while the dotted lines are his upper and lower bounds. Monckton’s data was digitized from the graph in Fig. 1.
[Nielsen-Gammon] says my bounds for the 21st-century evolution of CO2 concentration are not aligned with those of the UN. Except for a very small discrepancy between my curves and two outliers among the models used by the UN, my bounds encompass the output of the UN’s models respectably, as the blogger’s own overlay diagram illustrates. Furthermore, allowing for aspect-ratio adjustment, my graph of the UN’s projections is identical to a second graph produced by the UN itself for scenario A2 that also appears to exclude the two outliers.
It is fair enough to point out that Fig. 10.26 in IPCC AR4 WG1 has a plot of the projected A2 CO2 concentrations that seems to leave out the outliers. However, Monckton’s rendition is still not an honest representation of anything the IPCC ever published. I can prove this by blowing up the 2000-2010 portion of the graph in Fig. 4. I have done this in Fig. 5, where I have also plotted the actual mean annual global CO2 concentrations for that period. The clear implication of this graph is that even if the A2 scenario did predict atmospheric CO2 evolution (and it doesn’t,) it would actually be a good prediction, so far. In Figures 1 and 2, Lord has simply fabricated data to make it seem like the A2 scenario is wrong.
Figure 5. This is a blow-up of the graph in Fig. 4 for the years 2000-2010. I have also added the annual global mean atmospheric CO2 concentrations (blue line), obtained from NOAA.
Monckton claims that “for seven years, CO2 concentration has been rising in a straight line towards just 575 ppmv by 2100. This alone halves the IPCC’s temperature projections. Since 1980 temperature has risen at only 2.5 °F (1.5 °C) per century." In other words, he fit a straight line to the 2002-2009 data and extrapolated to the year 2100, at which time the trend predicts a CO2 concentration of 575 ppm. (See the light blue line in Fig. 1.)
It is impossible to distinguish a linear trend from an exponential trend like the one used for the A2 model input over such a short time period.
I pointed out to Lord Monckton that it’s often very hard to tell an exponential from a linear trend over a short time period, e.g., the 7-year period shown in Fig. 2. He replied,
I am, of course, familiar with the fact that, over a sufficiently short period (such as a decade of monthly records), a curve that is exponential (such as the IPCC predicts the CO2 concentration curve to be) may appear linear. However, there are numerous standard statistical tests that can be applied to monotonic or near-monotonic datasets, such as the CO2 concentration dataset, to establish whether exponentiality is being maintained in reality. The simplest and most direct of these is the one that I applied to the data before daring to draw the conclusion that CO2 concentration change over the past decade has degenerated towards mere linearity. One merely calculates the least-squares linear-regression trend over successively longer periods to see whether the slope of the trend progressively increases (as it must if the curve is genuinely exponential) or whether, instead, it progressively declines towards linearity (as it actually does). One can also calculate the trends over successive periods of, say, ten years, with start-points separated by one year. On both these tests, the CO2 concentration change has been flattening out appreciably. Nor can this decay from exponentiality towards linearity be attributed solely to the recent worldwide recession: for it had become evident long before the recession began.
In other words, the slope keeps getting larger in an exponential trend, but stays the same in a linear trend. Monckton is right that you can do that sort of statistical test, but Tamino actually applied Monckton’s test to the Mauna Loa observatory CO2 data since about 1968 and found that the 10-year slope in the data has been pretty continuously rising, including over the last several years. Furthermore, look at the graph in Fig. 5, and note that the solid black line representing the A2 climate model input looks quite linear over that time period, but looks exponential over the longer timeframe in Fig. 4. I went to the trouble of fitting a linear trend line to the A2 model input line from 2002-2009 and obtained a correlation coefficient (R2) of 0.99967. Since a perfectly linear trend would have R2 = 1, I suggest that it would be impossible to distinguish a linear from an exponential trend like that followed by the A2 scenario in real, “noisy” data over such a short time period.
Atmospheric CO2 concentration wouldn’t be treated as such a big deal if it didn’t affect temperature; so of course Lord Monckton has tried to show that the Fantasy IPCC “predictions” of CO2 concentration he made up translate into overly high temperature predictions. This is what he has done in the graph shown in Fig. 6.
Figure 6. Lord Monckton’s plot of global temperature anomalies over the period January 2002 to January 2009. The red line is a linear trend line Monckton fit to the data, and the pink/white field represents his Fantasy IPCC temperature predictions. I have no idea what his base period is. Taken from the Feb. 2009 edition of Lord Monckton’s “Monthly CO2 Report.”.
FANTASY #3. Lord Monckton uses graphs like that in Fig. 6 to support his claim that the climate models (AOGCMs) the IPCC uses to project future temperatures are wildly inaccurate.
Monckton didn’t actually get his Fantasy IPCC predictions of temperature evolution from AOGCM runs. Instead, he inappropriately fed his Fantasy IPCC predictions of CO2 concentration into equations meant to describe the EQUILIBRIUM model response to different CO2 concentrations.
Monckton indicated to me (5) that he obtained his graph of IPCC temperature predictions by running his Fantasy CO2 predictions (loosely based on the A2 emissions scenario) through the IPCC’s standard equation for converting CO2 concentration to temperature change, which can be found here.
The problem is that the equation mentioned is meant to describe equilibrium model response, rather than the transient response over time. In other words, they take the standard AOGCMs, input a certain stabilized CO2 concentration, and run the models until the climate output stabilizes around some new equilibrium. But it takes some time for the model systems to reach the new equilibrium state, because some of the feedbacks in the system (e.g., heat absorption as the ocean circulates) operate on fairly long timescales. Therefore, it is absolutely inappropriate to use the IPCC’s equation to describe anything to do with time evolution of the climate system. When I brought this up to Lord Monckton, he replied that he knows the difference between equilibrium and transient states, but he figures the equilibrium calculation comes close enough. But since the IPCC HAS published time-series (rather than just equilibrium) model output for the A2 scenario (see Fig. 7,) why wouldn’t he just use that?
Figure 7. Ensemble AOGCM output for the A2 emissions scenario, taken from Fig. 10.5 of IPCC AR4 WG1.
The answer is that if Lord Monckton had used the time-series model output, he would have had to admit that the IPCC temperature projections are still right in the ballpark. In Fig. 8, I have digitized the outer bounds of the model runs in Fig. 7, and also plotted the HadCRUT3 global annual mean temperature anomaly over the same period. The bottom line is that Monckton has put the wrong data into the wrong equation, and (surprise!) he got the wrong answer.
Figure 8. The blue and green lines represent the upper and lower bounds of the global average temperature anomaly from AOGCM output for the A2 emissions scenario during the 2002-2010 period. The black line represents the HadCRUT3 global temperature anomalies for that timeframe, normalized to the same base period.
I have shown here that in order to discredit the IPCC, Lord Monckton produced his graphs of atmospheric CO2 concentration and global mean temperature anomaly in the following manner:
- He confused a hypothetical scenario with a prediction.
- He falsely reported the data from the hypothetical scenario he was confusing with a prediction.
- He plugged his false data into the wrong equation to obtain false predictions of time-series temperature evolution.
- He messed up the statistical analyses of the real data.
These errors compound into a rather stunning display of complete incompetence. But since all, or at least nearly all, of this has been pointed out to Monckton in the past, there’s just no scientifically valid excuse for this. He’s just making it up.