Roy Spencer’s Great Blunder, Part 2

The following is reposted from Barry Bickmore's blog - it's PART 2 of my extended critique of Roy Spencer’s The Great Global Warming Blunder:  How Mother Nature Fooled the World’s Top Climate Scientists(New York:  Encounter Books, 2010).  If you haven’t read Part 1, you should probably do so before reading this.  See also Part 3.

Summary of Part 2:  Roy Spencer repeatedly claims that most of the rest of the climate science community deliberately ignores natural sources of climate variation, but then contradicts himself by launching an inept attack on the standard explanation for climate change during the glacial-interglacial cycles of the last million years (i.e., they are initiated by Milankovitch cycles).  The problems Spencer identifies are either red herrings or have been resolved, however, and he proposes no other explanation to take the place of the standard one.  In fact, climate scientists have used paleoclimate data such as that for the ice ages to show that climate sensitivity is likely to be close to the range the IPCC favors.  Therefore, it appears Roy Spencer is the one who wants to sweep established sources of natural climate variation under the rug.

The Mantra

It wasn’t easy slogging through Roy Spencer’s latest book, The Great Global Warming Blunder, because although it’s only 176 pages, it’s incredibly repetitive.  There is page after page of carping about how dense and corrupt his colleagues and the IPCC are, how hypocritical Al Gore is, and so on.  Most of this is just mildly annoying, but in my opinion, the language he uses in some of the messages he repeats ad nauseum is patently dishonest.  One such mantra is the claim that the climate science community has donned ideological blinders that prevent them from investigating natural sources of climate change.  Here are a few examples.

Aside from their almost total neglect of the role of nature in climate change, the scientists supporting the IPCC effort have done a pretty good job of summarizing the science of global warming, along with many of the uncertainties.  (p. xv)

We will see that researchers have reasoned themselves in a circle by first assuming that natural climate change does not exist, and then building climate models suggesting that only human pollution is needed to explain global warming.  (p. xxiii)

At this point you might be thinking, “Well of course natural climate change happens.”  But this has been surprisingly difficult to prove scientifically.  The IPCC avoids the subject because it detracts from the claim that humans are now the main driver of climate.  (p. xxvi)

The United Nations’ Intergovernmental Panel on Climate Change does acknowledge that there is natural climate variability on a year-to-year basis, and maybe even decade-to-decade.  After all, we have clear evidence that events like El Niño and La Niña cause some years to be warmer than others.  Yet the IPCC refuses to accept that the global warming (or cooling) on time scales of thirty years or more can also be caused by Mother Nature.  That, apparently, is humanity’s job.  (p. 1)

The IPCC has taken for granted that there are no natural variations in global average temperatures once one gets beyond a time scale of ten years or so.  (p. 16)

If Dr. Hansen is correct and humans are responsible for the recent warming, then what caused earlier periods of dramatic warming–and cooling?  Has natural climate change now ended, having been replaced by human-caused climate change?  This seems unlikely.  (p. 28)

Oh, there are many more–I just got bored of looking for them after about p. 30.  No matter how many times Roy Spencer says it, however, it is flatly untrue–and he knows it.  The fact is the he is the one that wants to ignore the evidence for past climate change, not the scientists associated with the IPCC.

Roy Spencer, Meet Sigmund Freud

How do I know Roy Spencer is aware of the truth-deficient nature of the above statements?  Because he says so in his book.  Check out his discussion of the standard explanation of what caused the glacial-interglacial cycles (i.e., intermittent ice ages) of the last million years or so, which involved huge swings in the global average temperature of about 4-7 °C.

The argument goes something like this:  There are long-term cycles, called Milankovitch cycles, in the Earth’s tilt and orbit around the sun.  These cycles cause small fluctuations in how much sunlight reaches different parts of the Earth.  The prevailing opinion is that the resulting variations in sunlight are not strong enough to have caused the ice age cycles shown in Fig. 6 unless there are positive feedbacks amplifying that small amount of forcing.  That is, unless the climate system is very sensitive.  When the Milankovitch cycles cause a small amount of warming, it leads to an increase in the CO2 content of the atmosphere.  But since more CO2 also causes warming, this sets up a vicious cycle of Warming, then more CO2, then more warming.  The process supposedly reverses when the Milankovitch cycles switch to causing a small decrease in the sunlight reaching the Earth.  A vicious cycle then occurs in the opposite direction, with decreasing CO2 and falling temperatures plunging the Earth into an ice age. (pp. 29-30)

So the other climate scientists do acknowledge the existence of large changes in the Earth’s past climate?  Changes so large that they can be described by phrases such as “vicious cycle” and “plunging the Earth into an ice age”?  And now Roy is telling us that not only do the other scientists acknowledge the existence of these changes, but they think they can actually explain them?

But wait!  Most of Roy’s statements I collected above apply specifically to the IPCC.  E.g., “The IPCC has taken for granted that there are no natural variations in global average temperatures once one gets beyond a time scale of ten years or so” (p. 16).  Could it be that climate scientists in general don’t ignore past climate changes, but the IPCC does?  All one has to do to pop that bubble of hope is to check out Chapter 6 of the Working Group 1 volume of the most recent IPCC report, which is entitled “Paleoclimate”.  It even includes a FAQ called, “Is the Current Climate Change Unusual Compared to Earlier Changes in Earth’s History?”  Their answer?  In some ways, yes, in other ways, no.  The chapter also includes another FAQ called, “What Caused the Ice Ages and Other Important Climate Changes Before the Industrial Era?”  The answer for the ice ages is… Milankovitch cycles.

I obviously can’t know what Roy’s motivations are, but this looks suspiciously like a textbook case of Freudian projection, because the fact is that information regarding large climate changes in the past is regarded by most climate scientists as an essential check on their projections of future climate, whereas it is extremely inconvenient for Roy Spencer’s ideas.  In the next section, for instance, I’ll show how he botches his discussion of the glacial-interglacial cycles so he can sweep them under the rug.

 Roy Spencer’s War on Data

Spencer’s central idea is that the Earth’s climate is fairly insensitive to external forcing.  But an “insensitive” climate means that changes in the incoming solar radiation, greenhouse gases, and so on, aren’t going to have a big effect.  If so, how can he explain the large climate changes that have happened in the past?  He doesn’t have a clue.  No, really.

But I don’t believe we have a clue what the governing factors were for these events.  As it is, our best Earth-observing satellites covering the globe every day are providing information that leads various scientists to different conclusions.  How can we hope to know what, if anything, the conditions on Earth in the distant past have to do with how the climate system operates today?  (p. 31)

What about the standard explanation for the glacial-interglacial cycles he mentioned?  He lists two main objections.  First, he argues that since ice core records show that temperature generally started changing before CO2 concentrations by several hundred years, CO2 can’t be a major cause of warming.

But if the major forcing of temperature really is carbon dioxide,… then the observed time lag either should be reversed or should not be there at all.  Therefore, the fact that the temperature changes preceded the CO2 changes in the ice core record is, to me, sufficient evidence that CO2 was not the forcing of, but instead the response to, the temperature changes.  (p. 30)

According to Spencer’s account of the standard explanation (see above,) the changes in CO2 during the glacial-interglacial cycles are regarded as a feedback, rather than the primary forcing, so it’s difficult to fathom with whom he thinks he’s arguing.  More CO2 can be dissolved in cold water than warm water, and there are a number of carbon sequestering and releasing processes involving ocean life.  Since it takes several hundred years for the deep ocean water to cycle up to the top, where it can be warmed up and lose CO2, it makes sense to suppose that if a warming event is initiated by something else (like changes in the amount and spatial distribution of incoming solar radiation,) the concomitant rise in atmospheric CO2 (which would enhance the initial warming) might lag behind by several hundred years.  There may also be other long-term feedbacks in the carbon cycle.  And while we can’t know for sure the reason for the lag time, it’s not as if it’s some great mystery for which nobody has come up with any plausible explanations.  There are actually a number of plausible explanations that are simply hard to test.  You can read all about different ideas regarding what governed CO2 concentrations during the glacial-interglacial cycles in (you guessed it) Section 6.4 of the last IPCC report (WG 1).

Now let’s move on to Spencer’s second objection to the standard explanation for the glacial-interglacial cycles.

But the biggest objection to the theory that the Milankovitch cycles caused the ice ages is that there is no statistically significant connection between the two!  A careful analysis has shown that the timing of the Milankovitch cycles relative to the ice ages is no closer than what would be expected by chance.  (p. 30)

In support of this claim, Spencer cites a paper by Carl Wunsch, an oceanographer at MIT (Wunsch, 2004).  Wunsh examined temperature records from several individual ice cores, and did a statistical analysis to show that very little of the temperature variation recorded could be explained by Milankovitch cycles.  This isn’t terribly surprising, since we’re talking about local records, which are much more prone to large, random fluctuations than the global average, but Wunsch’s point was important, because a number of previous studies had pointed to these local records as evidence of Milankovitch forcing.  In fact, they weren’t very good evidence for that.

When Roy Spencer saw Wunsch’s paper, he apparently glommed onto it as the last word on Milankovitch forcing–i.e., that we have no clue what caused the glacial-interglacial cycles, and we never will (even though Wunsch didn’t go that far).  But other scientists took the interesting questions Wunsch brought up and tried to address them.  Gerard Roe (U. Washington) showed that the rate of change of global ice volume correlates beautifully with the changes in incoming solar radiation due to Milankovitch cycles (Roe, 2006).  Peter Huybers and George Denton went on to show that the glacial-interglacial climate near the North Pole varies with the intensity of incoming solar radiation at those latitudes, while variations in Antarctic climate seem to be governed by changes in the duration of summer (Huybers and Denton, 2008).  So it appears that the Milankovitch theory is in better shape than ever.

Here’s the basic idea, once again, but explained a little more precisely.  1) Milankovitch cycles in the Earth’s orbit and tilt with respect to the Sun produce small variations in the amount and spatial distribution of sunlight hitting the Earth.  2) When these are on the increase in areas covered by ice sheets, small temperature changes are initiated that, in turn, start the ice sheets melting.  3) Since ice is very reflective, their melting causes a decrease in the albedo (i.e., the fraction of sunlight reflected back into space) of the Earth, which enhances the initial temperature change.  4) The oceans gradually release CO2 due to the warming, which further enhances the trend.  5) All of this reverses when the Milankovitch forcing starts pushing the system back the other way.

The great thing is that, since we can make good estimates of the changes in solar radiation, changes in the Earth’s albedo due to melting ice, and changes in atmospheric CO2 concentration during the ice ages, scientists can directly calculate the sensitivity of the climate to changes in the atmospheric CO2 concentration.  Obviously, there are uncertainties involved, but the result is that if you double the CO2 concentration, it’s likely to raise the temperature about 3 °C, plus or minus a degree or two.  Using models that reproduce more recent temperature data, the IPCC concluded that climate sensitivity is likely between 2 and 4.5 °C, with a best estimate of about 3 °C, and very unlikely less than 1.5 °C.  Scientists have come up with a number of other ways to estimate climate sensitivity, as well, sometimes involving paleoclimate data over widely divergent timescales, from hundreds of years to hundreds of millions.  And guess what?  They keep coming up with about the same answers.

This really is the crux of the whole matter.  If we’re going to go along with Roy Spencer, we have to reject a basic model of climate change that explains the data over widely divergent timescales, and replace it with… a big, fat nothing.  It’s not an argument about whether “Mother Nature” or mankind controls the climate, but over how the climate responds to changes in things like solar radiation and greenhouse gases, no matter what is governing them during a particular time period.  Spencer’s inept attack on the field of paleoclimatology is just his way of trying to sweep inconvenient data under the rug, all the while projecting his own rejection of established natural drivers of climate change onto the rest of the climate science community.

A Developing Theme?

If you read Part 1, do you notice a developing theme?  Roy Spencer finds a result he likes, and then stops looking.  In the next installment, I’ll reveal the worst example of all.


Huybers, P., and Denton, G. (2008) Antarctic temperatures at orbital timescales controlled by local summer durationNature Geoscience, 1, 787-792.

Roe, G. (2006) In defense of MilankovitchGeophysical Research Letters, 33, L24703.

Wunsch, C. (2004) Quantitative estimate of the Milankovitch-forced contribution to observed Quaternary climate changeQuaternary Science Reviews, 23, 1001-1012.

Posted by bbickmore on Tuesday, 1 March, 2011

Creative Commons License The Skeptical Science website by Skeptical Science is licensed under a Creative Commons Attribution 3.0 Unported License.