The Monckton Files: Solar Variation

Recently, Prof. John Abraham criticized Lord Christopher Monckton for citing scads of scientific papers to back up his opinions about climate change, but when Abraham actually looked into those papers, it often turned out they didn’t support Monckton’s conclusions, or they even contradicted those conclusions.  Prof. Abraham also criticized Monckton for improper citation of others’ work and data, often making it difficult to figure out where he was getting his information.

Given his rap sheet (including numerous infractions mentioned on this blog), I thought it would be fun to start examining Lord Monckton’s recent testimony before a committee of the U.S. Congress.  What if I were to scan through the document, randomly pick one of Monckton’s claims that I don’t know much about, and start investigating the literature he cites?  Would I find that he makes reasonable points, or that he has continued his nearly unblemished record of propagating scientific-sounding nonsense? Tim Lambert has already shown that Monckton’s testimony was flamboyantly incompetent about three issues (solar brightening, ocean acidification, and Snowball Earth), so I picked another topic that has to do with variations in the radiation output of the Sun.  Here’s what Monckton said about it.

The “global warming” that ceased late in 2001 (since when there has been a global cooling trend for eight full years) had begun in 1695, towards the end of the Maunder Minimum, a period of 70 years from 1645-1715 when the Sun was less active than at any time in the past 11,400 years (Hathaway, 2004). Solar activity increased with a rapidity unprecedented in the Holocene, reaching a Grand Solar Maximum during a period of 70 years from 1925-1995 when the Sun was very nearly as active as it had been at any time in the past 11,400 years (Hathaway, 2004; Usoskin, 2003; Solanki, 2005).

The point of rattling off these statistics is obvious, and it has to do with something called Total Solar Irradiance (TSI), which is essentially the amount of radiation from the Sun reaching the Earth’s upper atmosphere..  If we have moved from a period with exceptionally low TSI into a period of exceptionally high TSI, then maybe humans pumping massive amounts of greenhouse gases into the atmosphere aren’t really to blame for the recent warming trend.  If it’s all Nature’s fault, why bother trying to fix it?  Monckton urged members of the U.S. Congress to have “the courage to do nothing” about climate change.

In the remainder of this post, I’ll take apart Monckton’s argument and point out a few things he got right, and others he got wrong.  [SPOILER ALERT:  The final result is classic Monckton.]  😉

1. Monckton cites three scientific papers to back up his claims, but makes it unreasonably difficult to find those papers.

At the beginning of every talk I’ve seen Lord Monckton give, he tells his audience, “You must not believe a word I say.”  Instead, they should meticulously check for themselves on all his claims!  Well, to do that we have to be able to find the literature he cites.  Even if we admit that standards for literature citations in oral presentations are generally a little loose, even among scientists, shouldn’t we expect Monckton to include at least the journal name, volume, and first page number when he submits a written testimony to the Congressional Record?  Unfortunately, all he gives us is:  “(Hathaway, 2004; Usoskin, 2003; Solanki, 2005)”.  He doesn’t provide the traditional reference list at the end of the paper, or even footnotes.

Well, I’m a trained scientist, used to tracking down scientific literature, so I took the time to hunt them down.  Here are the papers that seem to be the most likely candidates.  (Lord Monckton is welcome to correct me if I’m wrong).

Hathaway, D.H. and Wilson, R.M. (2004) What the sunspot record tells us about space climate, Solar Physics, v. 224, pp. 5-19.

Usoskin, I.G., Solanki, S.K., Schüssler, M., Mursula, K., and Alanko, K. (2003) Millennium-scale sunspot number reconstruction:  Evidence for an unusually active sun since the 1940s, Physical Review Letters, v. 91, article # 211101.

Solanki, S.K., Usoskin, I.G., Kromer, B., Schüssler, M., and Beer, J. (2005) Reply to R. Muscheler et al., Nature, v. 436, pp. E4-E5.

There was another 2005 paper with Solanki as the first author (Solanki et al., Irradiance models, Advances in Space Research, v. 35, pp. 376-383.)  However, it doesn’t seem to provide direct support for any of Monckton’s points.

2. Monckton says that recent “global warming” began over 300 years ago, but that the globe has been cooling for “eight full years”. There’s no way around it–this is just a stupid comment, even if it is factually correct-ish.

Let’s look at factual correctness, first.  Thermometers haven’t been around for more than a few hundred years, and people have been taking regular temperature readings all around the globe for even less time.  Therefore, we can only reliably reconstruct an “instrumental” (thermometer-based) global average temperature since about 1850.  To estimate temperatures back farther than that, we have to use temperature “proxies,” i.e., other types of time-calibrated data that we can use to estimate what temperatures must have been like in the more distant past.  There are a number of different temperature proxies, including tree ring densities and borehole temperature profiles.

And what do we get from these temperature proxies?  Figure 1 shows a graph that is part of Fig. 6.10 from the latest IPCC report (AR4 WG1).  The top graph in the figure shows the instrumental temperature record for the Northern Hemisphere, and the bottom graph shows a number of different proxy temperature reconstructions (using different types of data, different statistical techniques, and so on) for the last 1300 years.

Figure 1. Part of Fig. 6.10 from the 2007 IPCC report (Working Group 1). a) Instrumental temperature records. b) Proxy temperature records of various types with the instrumental temperature record overlain as the dark black line.

Obviously, the proxy temperature reconstructions back before the instrumental record don’t give exactly the same answers, but they all show roughly the same picture of how temperatures evolved over that time period.  Furthermore, several of the proxy records show that the temperature was pulling out of a dip around the end of the 17th century, just like Monckton said.  Fixing the start of the recent warming trend to the year 1695 is ridiculous, however, because the data just aren’t that precise!  And furthermore, the instrumental record shows that there were multi-decade dips in the temperature during the last half of the 19th century and the mid-20th century.

So I guess one could say that we’ve generally been warming since 1695 (or thereabouts,) but if there have been multi-decade cooling trends since then, it makes absolutely no sense to contrast “eight full years” of cooling with a 300+ year warming trend.  In fact, even over the past 30 years, during which climate scientists agree there has been a pretty consistent warming trend, there have been other 8-year periods of cooling.  Figure 2 shows a graph of the slope of the global average temperature (from NASA’s GISTEMP and the UK’s HadCRUT3 data sets) over 8-year periods starting with the year plotted.  That is, the point plotted above the year 2002 is the slope over the 8-year period 2002-2009, and so on.  All the points that lie below zero on the y-axis represent 8-year periods in which there was an overall cooling trend.  Note that several points (not just the last couple) have negative slopes, and each of those represents an 8-year period during which there was net global cooling, even though most of the time the overall trend from 1979-2009 was warming.

Figure 2. 8-year slopes in the global average temperature over the past 30 years, for the GISTEMP and HadCRUT3 data sets. Each data point represents the slope in the data during an 8-year period starting with the year plotted. E.g., the point for the year 2002 represents the slope of the temperature data during 2002-2009.

Monckton’s comparison may be a bit weird, and he definitely gives undue precision to the dates he recites, but what about his main reason for bringing it up?  If the recent warming trend began before the Industrial Revolution, then it can’t be due to humans burning fossil fuels, right?

Well of course that’s right–at least partially–and certainly no self-respecting climate scientist would deny it.  Consider Figure 3, which shows a graph taken from Fig. 1 of FAQ 9.2 in the 2007 IPCC Report (Working Group 1).  The black line in the figure shows the actual evolution of the global mean temperature over the 20th century. The blue band shows the range of predictions for the temperature evolution given by the IPCC’s climate models if you only drive the model system with natural factors like changes in solar output, volcanic activity, and so on. The pink band shows what the models predict if you add in human factors like increased greenhouse gas concentrations from fossil fuel burning.

Figure 3. Taken from Fig. 1 of FAQ 9.2 in IPCC AR4 (Working Group 1). The black line shows the actual evolution of the global mean temperature over the 20th century. The blue band shows the range of predictions for the temperature evolution given by the IPCC's climate models if you only drive the system with natural factors like changes in solar output, volcanic activity, and so on. The pink band shows what the models predict if you add in human factors like increased greenhouse gas concentrations from fossil fuel burning.

You should notice a couple things about the graph in Figure 3.  First, given all the known climate drivers (the pink band), the IPCC climate models do a decent job of mimicking the temperature evolution over the 20th century.  But they don’t do such a good job over the last half of the 20th century when you don’t include human causes (the blue band).  Second, the blue and pink bands stop overlapping one another after about 1975, so the predictions for all natural vs. natural + human causes have really diverged by then.  I think it’s fair to say, therefore, that the climate scientists who put this graph in the IPCC report would have no problem attributing climate change before about 1950-1975 mainly to natural factors like variations in TSI.

So really, Monckton’s point is worthless unless he can show that the IPCC climate models can’t adequately reproduce the global temperature evolution over the past 300 years.  Since he didn’t show that, I can only assume that Lord Monckton was playing to the large segment of the public that erroneously believes the IPCC says humans are the ONLY thing driving climate change.

3. Lord Monckton next says that during 1645-1715 the sun was less active than at any other time in the last 11,400 years, and cites “Hathaway, 2004” to support this claim.  This paper (see above for the full reference) actually talks about what the sunspot record tells us about TSI over the last 400 years, not 11,400.

The number of sunspots (averaged over a number of years) has been shown to correlate pretty well with TSI.  Since we haven’t had satellites to measure TSI for very long, we can extend our record of TSI back further by looking at records of sunspot observations, which have been carried out for about the last 400 years.  The article by Hathaway and Wilson does, in fact, show that TSI was very low during the Maunder Minimum (1645-1715), but since their paper is all about sunspot numbers they only discuss the past 400 years, not the past 11,400 years.  Here are a couple quotations from the abstract of the paper to illustrate my point.

The records concerning the number, sizes, and positions of sunspots provide a direct means of characterizing solar activity over nearly 400 years….

The sunspot record shows… there has been a significant secular increase in the amplitudes of the sunspot cycles since the end of the Maunder Minimum (1715).

Let’s not be too hard on His Lordship, though.  It turns out that Lord Monckton was possibly right about the Maunder Minumum being a low point in the last 11,400 years (see below.)  He just didn’t cite the right paper to back up his claim.

4. Monckton claims that during 1925-1995 solar activity was almost as high as it has ever been in the last 11,4oo years, citing three different papers as evidence (Hathaway, 2004; Usoskin, 2003; Solanki, 2005).  If you dig hard enough in that list, you can actually find evidence for Monckton’s claims!

Once again, it’s a little strange to cite Hathaway’s paper when making a point about the last 11,400 years of solar activity, when the paper only covers the last 400.  It also turns out that the paper by Usoskin et al. only covers the last 1150 years!  But at least the the paper by Solanki et al. does, indeed, refer to solar activity during the last 11,400 years.  And to be fair, I should acknowledge that all of these papers do at least support the point that solar activity has gotten abnormally high over the last hundred years, or so.

The wackiness of Monckton’s citations continues, though, because it turns out that Solanki et al. (2005) is simply a short response to a comment some other scientists made on a paper Solanki’s group published the year before!  You can’t find anything remotely like the statistics Monckton cites in Solanki et al. (2005)  So… if you want to finally track down the evidence that, compared to the last 11,400 years TSI is abnormally high now, and was abnormally low in the Maunder Minimum, you will need to find the following paper:

Solanki, S.K., Usoskin, I.G., Kromer, B., Schüssler, M., and Beer, J. (2004) Unusual activity of the Sun during recent decades compared to the previous 11,000 years, Nature, v. 431, pp. 1084-1087.

Here’s what Solanki’s paper is about.  We have only had satellites to directly measure TSI for a few decades, and people have only been counting sunspots for about the last 400 years.  To get estimates of TSI before 400 years ago, scientists turn to cosmogenic isotopes, which are isotopes generated in the atmosphere due to interactions with incoming cosmic rays.  Since TSI affects the amount of incoming cosmic rays, and the amount of cosmic rays affects the amount of cosmogenic isotopes generated, we can estimate variations in TSI by looking at the amounts of various cosmogenic isotopes in features of known age, like tree rings or annual ice layers in glaciers.  Solanki’s group used the amount of C-14 (a cosmogenic isotope) in tree rings of known age to estimate the sunspot number (and hence the TSI) over the last 11,400 years.  The graph in Figure 4 (taken from Fig. 3 of Solanki et al., 2004) summarizes their results.  Here the red line represents the observed sunspot number over the last 400 years, and the blue line represents the reconstructed sunspot number from their tree ring data over the last 11,400 years.

Figure 4. This graph was taken from Fig. 3 of Solanki et al. (2004), shows. The red line represents the observed sunspot number over the last 400 years, and the blue line represents the estimated sunspot number from their C-14 tree ring data over the last 11,400 years.

5. The graph above shows that even if Monckton’s citation style is completely wacky, it turns out that the statistics he quoted were essentially correct!  We are in a period of exceptionally high solar activity, right after a period of exceptionally low activity.

Point conceded, even if I had to go to extraordinary lengths to find out the origin of Monckton’s statistics.

6. However, Monckton has often used the standard climate skeptic’s argument that the “hockey stick” temperature graph used by the IPCC (see the first figure above) was fraudulently transmogrified so that the Medieval Warm Period (about AD 950-1250) is no longer shown to have been warmer than today.  (Click here to read more about the “Hockey Stick Controversy.”)  Solanki’s data stomps that argument into the ground.

Just look at the graph in Figure 5, which is taken from Fig. 2 of Solanki et al. (2004).  It shows measured and estimated sunspot numbers for a little over the last 1000 years, which is about the same time period found in the temperature plots in Figure 1.  And guess what?  Both the temperature plots and the sunspot plots are shaped like… hockey sticks.  This was discussed further by:

Usoskin et al. (2004) Solar activity over the last 1150 years:  Does it correlate with climate? Proceedings of the 13th Cool Stars Workshop, pp. 19-22.

Figure 5. Graph of real and estimated sunspot numbers for the last 1000+ years, taken from Fig. 2 of Solanki et al. (2004).

I find it interesting that someone like Monckton would argue BOTH that the hockey stick temperature graph is a fraud, AND that variations in TSI over the last few hundred years (which look like a hockey stick when graphed) have been the main driver of temperature change.

7.  And yet, even if we grant that all of Monckton’s statistics are factually correct, natural variations in TSI cannot explain the last few decades of global warming, because over that time TSI has been steady while temperature has gone up.  This has been specifically discussed by the very scientists from whom Monckton collected his factoids.

Remember how Monckton cited the 2005 paper by Solanki’s group, but the information Monckton referred to was really in their 2004 paper?  I see two possibilities–either it was a simple typo, or Lord Monckton didn’t actually read Solanki et al. (2004).  If he had even read the abstract of the paper, he would have noticed the following statement.

Although the rarity of the current episode of high average sunspot numbers may indicate that the Sun has contributed to the unusual climate change during the twentieth century, we point out that solar variability is unlikely to have been the dominant cause of the strong warming during the past three decades.  (Solanki et al., 2004)

If Monckton had read on in the paper, he would have found out that Solanki’s group had published a paper in 2003 in which they had analyzed the question of the extent to which variation in TSI can explain the last few decades of climate change.

In ref. 3, [Solanki, S. K., and Krivova, N. (2003) Can solar variability explain global warming since 1970? Journal of Geophysical Research, 108, doi: 10.1029/2002JA009753.] reconstructions of solar total and spectral irradiance as well as of cosmic ray flux were compared with surface temperature records covering approximately 150 years. It was shown that even under the extreme assumption that the Sun was responsible for all the global warming prior to 1970, at the most 30% of the strong warming since then can be of solar origin.

Supposing Monckton didn’t bother to actually read the literature he cites, he could have appealed to Google to find out what Sami Solanki thinks about how his work relates to the climate change debate.  That’s what I did, and I found the following passage in this article from the BBC.

Even though misguided journalists have sometimes mistaken his work as implying a solar cause to modern-day warming, Sami Solanki agrees with the IPCC verdict.

“Since 1970, the cosmic ray flux has not changed markedly while the global temperature has shown a rapid rise,” he says. “And that lack of correlation is proof that the Sun doesn’t cause the warming we are seeing now.”

8. Although Monckton’s statistics were supposed to show us why the IPCC is wrong about humans being responsible for recent global warming, the data from which he got those statistics actually agree quite well with the picture painted by the IPCC.

As Dr. Solanki said, variations in TSI cannot explain much of the warming trend since 1970.  And remember my discussion of Figure 3, which was taken from the IPCC report?  I said that the graph shows the IPCC is only claiming humans have been the main drivers of climate change since sometime around 1950-1975.

So what was the point of Monckton’s recitation of statistics about TSI?  It seems to me that he chose statistics that would sound problematic for the IPCC to laypersons who don’t clearly understand what the IPCC actually claims.  In fact, his statistics are essentially meaningless in that context.

The Bottom Line

I have examined a single paragraph in Lord Monckton’s recent testimony to a committee of the U.S. Congress, in which Monckton cited some statistics about natural variations in incoming solar radiation (TSI) to show that recent global warming can be explained by natural causes.  I show that Monckton’s statistics are correct, but the scientists who originated the data he cites say that the data actually agrees with the IPCC.  Furthermore, I show that Monckton made it exceptionally difficult to check his claims by citing the wrong papers and failing to provide full citations.  These are exactly same kinds of problems John Abraham found in his recent examination of one of Lord Monckton’s presentations.

[UPDATE (June 26, 2010):  I didn’t realize it when I posted this, but it turns out that John Abraham has already beat me to the punch on some of my points, because Monckton brought up something similar in his Minnesota presentation that Prof. Abraham critiqued.  Prof. Abraham did a much more thorough literature review than I did–I mainly stuck to the papers Monckton cited in his testimony before Congress–and he went to the trouble of contacting David Hathaway of “Hathaway, 2004” fame.  Here’s what Hathaway told him:  “I did not then, nor did I ever, suggest that solar variability plays a dominant role in climate change.”  In any case, in his response to Abraham’s critique, Lord Monckton castigated people who wrote with approval about the critique without personally checking all of Abraham’s claims.  “As usual though, none of these silly bloggers make any attempt actually to verify whether what poor Abraham is saying actually has the slightest contact with reality.” Well, I’d say that I spent considerable time checking this particular point, and even before I realized the overlap, I had verified Abraham’s assessment in spades.  I get the impression that Lord Monckton adjures his audiences to check his claims for themselves simply for the dramatic effect.  He doesn’t actually expect anyone to do it.]

Advertisement