Should you, dear readers, doubt that the climate empire strikes back at even the mildest qualifications of greenhouse gas (GHG) orthodoxy, merely consider a recent draft essay by Clifford Asness and Aaron Brown on the recent temperature record and attendant implications for policies to reduce GHG emissions. Titled “It’s Not the Heat, It’s the Tepidity,” the immediate — indeed, Pavlovian — and highly critical responses from various observers were fascinating in their vociferousness and contempt. Truly outstanding in those dimensions was that offered by Mark Buchanan on Bloomberg View, about whose critique I offer some observations below.
A full disclosure first: Asness is a member of the Board of Trustees of the American Enterprise Institute, where I work as the senior scholar in energy and environmental policy.
Asness and Brown at the very outset state clearly that they “are not climate scientists” and that they “are not challenging climate science.” They make two very simple points: Neither the land-ocean warming observed over a longer period since 1880 or a shorter one since 1990, nor the future temperature paths projected by the Intergovernmental Panel on Climate Change support the common predictions of severe climate effects over the course of this century from increasing GHG concentrations. They note, but do not depend upon, the observation that the actual temperature record since 1990 has been lower than the lowest IPCC projections made in its first assessment report, and they concede fully that future warming may prove substantially greater than the recent trend suggests.
Their second point is equally straightforward. “Dangerous” global warming often is defined as warming greater than either 2°C or 4°C by the end of the century. Asness and Brown show that an extrapolation of the actual temperature record suggests that future warming will become dangerous under that definition much later than that, and later even than in the lowest temperature paths projected by the IPCC. In their simple extrapolations, dangerous warming does not occur for 130-530 years. They do not state explicitly the obvious implication: to the extent that the actual temperature record provides important information about future trends, the time available for adaptation to the effects of increasing atmospheric concentrations of GHG is substantially greater than often asserted, even under adverse assumptions about such feedback effects as ocean evaporation and cloud dynamics. More generally, the authors do not delve into the policy implications of the recent temperature trend as compared with the predictions of the climate models, but that is straightforward: Given that the models do not predict the near-term record well, why should policymaking defer uncritically to their longer-term predictions?
The arguments offered by Asness and Brown are more than merely modest: They bend over backward to accept the basics of the conventional view of temperature trends, arguing that “the world has indeed warmed for more than a century, … there has not been a recent hiatus,… [and] the most recent measurements are at a record high.” Indeed, their modest argument stacks the deck in favor of the conventional view of the effects of GHG emissions, and itself can be criticized for several reasons as being too fair.
First: That warming has been observed “for more than a century” actually reveals far less than one might assume. Temperatures have been rising off and on since the end of the little ice age around 1850. After that year, temperatures then began to fall by 1880, and in particular in the wake of the eruption of Mt. Krakatoa in 1883. A warming trend was observed from 1910 through 1940, temperatures then fell on net through the mid- to late 1970s, increased through 1998 (a strong El Niño year), and have been flat or increasing much more slowly than projected by IPCC since about 2002. Accordingly, the longer-term trend may reflect in substantial part a global emergence from the little ice age, as well as many other natural influences that are understood poorly, in addition to the more-recent effects of increasing GHG concentrations. After all, the 1910-1940 warming was very unlikely to have been (mainly) anthropogenic, the cooling through the 1970s is problematic for the orthodox view, and one can believe (as I do) that increasing GHG concentrations are having some effect without knowing their importance relative to natural phenomena.
Second: In their temperature charts, Asness and Brown show temperature deviations (or anomalies) from the 1951-1980 average. Because that period displayed essentially no trend, and because it was followed by the warming period through 1998, their chart shows a more-or-less constant warming trend after the late 1970s as deviations from the 1951-1980 average. But the use of a different base period — say, 1981-2010 — yields a different deviation path, as 1981-2010 comprised a warming period followed by a period with slight warming or no trend. Consider the following figure, in which the deviation data show a temperature anomaly of only 0.30°C in February 2015, and almost no trend from 2002 until now. The point here is that there is no obviously “correct” base period from which to measure temperature anomalies, and the use of 1951-1980 arguably makes the conventional view appear to be stronger than would be the case with other base periods. Note also that the Asness-Brown chart is for the global land-ocean annual means, while the figure below is for the lower atmosphere. The two are correlated, but the surface temperature record is deeply problematic, with heat-island effects difficult to expunge from the data, and poor placement of and shifts in the measurement stations, etc. An example: For many years, “China” was one monitoring station in Shanghai, and as that city grew, “China” warmed. Surprise!
Third: With respect to a recent pause (“hiatus”) in the temperature record: The temperature charts for 1880-2014 presented by Asness and Brown suggest a warming trend of 0.67°C per century, a path substantially lower than the 2.8 °C in 2100 that is the mid-range (emissions scenario A1B) projection made by the IPCC in its 4th assessment report (AR4, 2007, Table SPM.3). In the 5th assessment report (AR5, 2013, Table SPM.2), emissions scenarios largely are replaced with four alternative “representative concentration pathways,” that is, alternative projected time paths of atmospheric concentrations of GHG. The third-highest (often used as the base scenario in policy discussions) is RCP6.0 (a theoretical radiative effect of six watts per square meter), yielding a projected temperature increase of 2.2°C by 2100.
But there has not been much of a trend since 2002, and there are some data showing no temperature increase since 1996. So there may be a problem with the Asness and Brown concession that “there has not been a recent hiatus,” but in any event that is not quite the right question. The real issue, as the authors recognize elsewhere in their analysis, is whether the climate models are making predictions consistent with the temperature record, and the evidence on that question is not salutary, as illustrated in the following figure. As a crude generalization, the temperature record is at the bottom or below the range of the models’ projections.
Fourth: That “the most recent measurements are at a record high” is a concession afflicted with the problem discussed in the first point above: Temperatures appear to have been rising in fits and starts since the end of the little ice age, and so the influence of increasing GHG concentrations must be separated from the effects of other phenomena. That is why the common “warmest year on record” argument is far less revealing than often assumed, as it fails to delineate those separate effects.
In short: I believe that Asness and Brown are being overly kind to the conventional argument. They write, correctly, that they “are not challenging climate science”; but that is based upon a sensible definition of that term. Once we recognize that the climate industry employs a far looser definition of “science,” the issue of precisely what Asness and Brown are challenging becomes far more ambiguous. Certainly they are not challenging the view, accepted by virtually everyone, that increasing GHG concentrations will yield some anthropogenic warming, a view supported by some evidence even given the failings of the climate models. But they are arguing also, correctly, that the temperature record does not support the many claims of imminent disaster, and it is rather obvious that the climate industry views those claims — and the attendant policy prescriptions — as part of climate “science.” In that sense, therefore, Asness and Brown are challenging climate science as that term is defined by the climate industry, which to a very significant degree hates fossil fuels as a matter of ideological principle, and which therefore is tied to a particular set of policies that loosely can be categorized as “decarbonization,” or a reduction in energy use and a substitution of unconventional (that is, expensive) energy in place of conventional fuels. In other words, the climate industry, I believe deliberately, has blurred the distinction between scientific hypotheses that are refutable in the light of evidence, and their preferred policy implications of those hypotheses regardless of the evidence. That those policy preferences are debatable is obvious, independent of climate science defined narrowly; and the climate industry has blurred that distinction precisely so that they can accuse those dissenting from the policy prescriptions as flat-earthers engaged in a denial of “science.”
Thus do we arrive at the reaction on the part of the climate industry to the (again, modest) analysis by Asness and Brown. Mark Buchanan on Bloomberg View presents a short critique titled “Asness Should Manage Money, Not the Planet”; he may or may not have chosen that snarky title, but can there be any doubt that he agrees with it? His arguments, and my reactions, can be delineated as follows.
First, Buchanan criticizes the authors for ignoring “the way carbon emissions interact with the environment.” He simply ignores the fact that Asness and Brown make it clear at the outset that they are writing about neither the science nor the modeling of such interactions; it is the data on temperature trends that is the focus of their essay. And Buchanan seems not to recognize that this very interaction issue — feedback effects — is the central crux of the debate. It is easy to doubt that he perceives the existence of a debate at all; in any event, there exists a lot of evidence suggesting that the IPCC has overestimated such feedback effects. (See also the second figure above.) IPCC’s current (AR5) estimate of “climate sensitivity” (the temperature effect by 2100 of a doubling of GHG concentrations) is 3°C, down from 4.5°C in earlier assessment reports; but there is little recent evidence in support of even the lower estimate. The following chart shows a compilation of recent estimates of climate sensitivity published in the peer-reviewed literature, as compiled by Pat Michaels and Chip Knappenberger. Most of these estimates are around 40 percent below the IPCC GHG climate sensitivity figure. Bloomberg View describes Buchanan as a physicist; can it possibly be the case that he does not understand any of this? Or is he ignoring the scientific literature due to ideological preferences?
And as aside: Notice the ease with which Buchanan uses the word “carbon” in place of “carbon dioxide” or GHG. That represents a substitution of political propaganda in place of analysis: Carbon dioxide is not “carbon” and it is not a pollutant, as a certain minimum atmospheric concentration of carbon dioxide is necessary for life itself. Water vapor and clouds are responsible for 65-85 percent of the radiative (warming) properties of the atmosphere; does Buchanan believe that water vapor too is a “pollutant”?
Second, Buchanan then argues that “extrapolation can be very misleading,” a trivial truism followed by this amazing assertion:
We know that global warming has gained momentum along with the concentration of carbon dioxide in the atmosphere, which has gone from only 280 parts per million in the pre-industrial era to about 400 ppm today. We also know that the concentration will accelerate… [and] as a result, we should expect accelerating temperature change.
Where to begin? Apart from Buchanan’s descent into an obvious extrapolation — which, remember, “can be very misleading” — his argument about temperature “momentum” is false as a factual matter, and his “accelerating” assertion about the effect of rising GHG concentrations is wrong scientifically! With respect to “momentum,” see the first figure above; there has been virtually no temperature trend since 2002 even as GHG concentrations increased from 337 ppm in 1979 to 370 ppm in 2000 to almost 400 ppm in 2014. Or, to put it a bit differently: Notwithstanding Buchanan’s “momentum” rhetoric — an obvious extrapolation model — the correlation between GHG concentrations and temperature trends is weaker than Buchanan would have us believe. This obviously is due to the importance of other factors: feedback effects, the El Niño/Southern Oscillation, the Atlantic Multidecadal Oscillation, ad infinitum. Can Buchanan possibly believe that those phenomena are understood well? Or has he — again, a physicist — never heard of them?
With respect to Buchanan’s “accelerating” argument: It is well-known in the scientific community, but probably less so among journalists, that the relationship between increasing GHG concentrations and temperatures (actually, radiative effects) is logarithmic. (First approximations of the equations are provided here.) As GHG concentrations rise, the temperature impact of a further given increase in concentrations diminishes, a scientific reality about which IPCC is clear. And so after accusing Asness and Brown of “ignor[ing] how physics actually works,” and after having criticized extrapolations, Buchanan criticizes them with an extrapolation argument that is wrong scientifically. Wow.
Finally, Buchanan tells us that “climate scientists… [have] made much more troubling projections of climate change than a simple extrapolation would suggest — projections like those of the IPCC.” Perhaps Buchanan is embarrassed to compare those projections with the actual temperature record, which is what Asness and Brown do, and which is illustrated in the second figure above. But given his assertions, it is reasonable to assume that Buchanan supports policies to reduce GHG emissions: the Obama Climate Action Plan narrowly, and the UN Framework Convention on Climate Change more broadly. And so it is wholly appropriate to ask whether Buchanan has bothered to examine what the future temperature effects of various policies would be. And so let us apply the MAGICC/SCENGEN climate model developed at the National Center for Atmospheric Research — used by both the EPA and IPCC — to the reductions in US and global GHG emissions discussed most prominently. The following table presents those projected temperature effects.
In brief: The prospective temperature reductions would be trivial (and the costs enormous, a topic for another day). Buchanan, in short, is arguing that GHG emissions are creating climate effects that are serious, and therefore (implicitly) that GHG emissions should be reduced, without exploring the likely effects of such policies. The plain reality is that Asness and Brown have offered thinking about this issue far more serious than is the case for Buchanan.
Benjamin Zycher is the John G. Searle scholar at the American Enterprise Institute.