Skip to content

The dire results of anthropogenic global warming have become passé. Treated by the news media and climate alarmists as established scientific fact, the IPCC’s vision of a dystopian future, a world ravaged by global warming, is fed to our children in school, TV shows and Hollywood movies. What is never mentioned is that even the IPCC’s predictions encompass several ranges of possible outcome, all predicated on a seemingly simple but mysterious factor called climate system sensitivity. A recent study, published in the journal Science, used spatially more complete paleoclimate data for the Last Glacial Maximum (LGM) in an effort to improve previous estimates of climate sensitivity. The new results have not been widely reported in the news media because, according to the researchers, “these results imply a lower probability of imminent extreme climatic change than previously thought.”

According to a team of researchers led by Andreas Schmittner from Oregon State University, “climate sensitivity is the change in global mean near-surface air temperature ΔSAT caused by an arbitrary perturbation ΔF (radiative forcing) of Earth’s radiative balance at the top of the atmosphere with respect to a given reference state.” More simply put, sensitivity to a forcing—carbon dioxide (CO2) for example—is based on measured change from a base equilibrium state to a new equilibrium state. The equilibrium climate sensitivity (ECS) for a doubling of atmospheric CO2 concentrations, denoted ECS2xC, has been estimated at 3 ± 1.5 °K, an estimate that has remained unchanged for the past three decades. Noting that this value suggests “a large uncertainty,” Schmittner et al. set out to improve that estimate. As described in an accompanying perspective article, by Gabriele C. Hegerl and Tom Russon, the study work was described this way:

Schmittner et al. present estimates of ECS based on recent compilations of spatially reconstructed surface ocean and land temperatures at the LGM. The LGM occurred around 21,000 years ago, when the Northern Hemisphere was strongly glaciated, atmospheric dust levels were higher than today’s, and atmospheric CO2 was ∼100 parts per million by volume lower than preindustrial levels; all of these factors contributed to climate cooling. It is difficult to estimate ECS by directly comparing the reconstructed global average temperature response to the total climate-forcing change, because the feedbacks that enhance the direct temperature response in a much colder climate are different from those that operate during increasing greenhouse gas conditions. Instead, Schmittner et al. performed a large range of simulations of LGM climates using multiple versions of the University of Victoria (UVic) Earth System Model of intermediate complexity; each version assumed a different model ECS value. They then compared the simulations with the LGM climate reconstructions to determine which ECS values best matched the LGM climate.

The result, described in “Climate Sensitivity Estimated from Temperature Reconstructions of the Last Glacial Maximum,” combines the output from 47 versions of the UVic Earth System Climate Model, each with different climate sensitivities ranging from ECS2xC = 0.3 to 8.3 °K. The LGM climate reconstruction data sets used provided more spatial coverage than had been available for earlier climate sensitivity estimates. In the figure below, zonally averaged surface temperature change between the LGM and modern times. The thick black line denotes the climate reconstructions, and the gray shading the ±1, 2, and 3 °K intervals around the observations.

As can be seen from the density of data presented above, the raw results require some interpretation. Though the report contains a wealth of detail the two most outstanding findings have to do with very low sensitivities and very high sensitivities. Here is a summary from the paper.

Models with ECS2xC < 1.3 K underestimate the cooling at the LGM almost everywhere, particularly at mid-latitudes and over Antarctica, whereas models with ECS2xC > 4.5 K overestimate the cooling almost everywhere, particularly at low latitudes. High-sensitivity models (ECS2xC > 6.3 K) show a runaway effect resulting in a completely ice-covered planet. Once snow and ice cover reach a critical latitude, the positive ice-albedo feedback is larger than the negative feedback because of reduced longwave radiation (Planck feedback), triggering an irreversible transition. During the LGM, Earth was covered by more ice and snow than it is today, but continental ice sheets did not extend equatorward of ~40°N/S, and the tropics and subtropics were ice free except at high altitudes. Our model thus suggests that large climate sensitivities (ECS2xC > 6 K) cannot be reconciled with paleoclimatic and geologic evidence and hence should be assigned near-zero probability.

It should be noted that this study, indeed, any study based on computer models comes with some significant caveats. First, the results were generated by one particular climate model of intermediate complexity and, as I have said innumerable times, no computer model accurately captures the complexity of Earth’s climate system. Different feedback strengths would undoubtedly yield different results using other climate models, particularly those that represent the physics more completely.

Second, the uncertainties in the forcing terms—for example, those associated with the distributions of ice sheets and vegetation—remain, well, uncertain. Change the estimated forcings and you change the climate sensitivity. As is always the case with such models and studies based on models, the uncertainties in the inputs are often glossed over when the results are presented.

A third caveat, reported by Hegerl and Russon, relates to Schmittner et al.’s finding that the LGM ocean temperature reconstructions exert a far stronger control on the ECS estimate than do the land reconstructions. “Even the most recent proxy LGM ocean data compilations remain biased toward the low-latitude ocean and the North Atlantic,” they maintain. But even with these potential flaws, the result of this study are instructive.

“The best-fitting model (ECS2xC = 2.4 K) reproduces well the reconstructed global mean cooling of 2.2 K (within two significant digits), as well as much of the meridional pattern of the zonally averaged temperature anomalies (correlation coefficient r = 0.8),” the authors state, while noting that: “Substantial discrepancies occur over Antarctica, where the model underestimates the observed cooling by almost 4 K, and between 45° to 50° in both hemispheres, where the model is also too warm.” To be fair, the researchers are quite frank in accessing the flaws in their model. Still, no climate alarmist would have believed that Earth was so insensitive to CO2.

The worst impacts of global warming are only predicted for high temperature rises, implying high sensitivities. In the figure shown below, taken from the IPCC’s AR4 report, the predicted ravages of global warming for different temperature increases are shown. Note that the temperature increase on the chart is measured from 1980-1999, whereas the temperature increase inferred by ECS2xC is for a doubling of CO2 levels, usually measured from “preindustrial times,” meaning the 1800s.

A usual value for preindustrial atmospheric CO2 is around 280ppm. We are now approaching 400ppm, an increase of 120ppm or a 42% increase. Recalling that the sensitivity increase is based on a doubling of CO2 we are almost half way to a predicted increase of ~2.4 °C. Given the widely quoted increase of 1.8 °C for the past 100 years, we should only see an further increase of about 0.6 °C if atmospheric concentrations reach 560ppm. The IPCC chart above predicts nothing for a 0.6 °C increase. Little wonder that the authors predict “a lower probability of imminent extreme climatic change than previously thought.”

Returning to the two interesting findings I mentioned above: If climate sensitivity is too low the model fails to reproduce observed changes since the start of the Holocene; If climate sensitivity is too high the model predicts that Earth would never recover from a glacial period, resulting in a frozen planet. But Earth has experienced a score of alternating glacial-interglacial cycles, so that prediction is also wrong.

Bottom line? The simulations fail when extreme values are used for climate sensitivity. The researchers conclude “we find that climate sensitivities larger than 6 K are implausible, and that both the most likely value and the uncertainty range are smaller than previously thought.” I draw some further conclusions.

What my career as a Computer Scientist and modeler tells me is that the climate model used for this study is deeply flawed. It only functions well over a small range of values and a limited time span. These are representative of the reasonably good data collected about climate change over the past several decades—the same data used in concocting the model in the first place. This is not unusual, modelers often take great pains to fit a model to the available data only to find they fail miserably when trying to extend the model’s predictions into the past or the future.

Basically, the number this paper found most probable for climate sensitivity is the one implied by recent observations, slanted by the assumption that CO2 is the primary driver of that change—combine hundreds of runs from a model so constructed and, surprise, you get back confirmation of your initial assumptions. “This implies that the effect of CO2 on climate is less than previously thought,” said lead author Andreas Schmittner. Indeed.

The choice comes down to dismissing the climate model as predicatively worthless or believe the model when it says all of the really scary, catastrophic affects of global warming predicted by the IPCC and other warmist fanatics are only possible at implausibly high climate sensitivities. Since all those dire predictions are predicated on models, the first choice says the IPCC claims are unsupported in the first place. And if the modeling approach is assumed valid, the reported results led the researchers to conclude: “Assuming that paleoclimatic constraints apply to the future, as predicted by our model, these results imply a lower probability of imminent extreme climatic change than previously thought.” In short, the carbon dioxide climate catastrophe has been called off. Of course, more rational, skeptical scientists would say there never was a looming crisis in the first place.

Be safe, enjoy the interglacial and stay skeptical.