New paper: Assessing ExxonMobil’s global warming projections
Posted on 16 January 2023 by BaerbelW
This is a quick summary about the newly published paper “Assessing ExxonMobil’s global warming projections” by Professor of Environmental Science and Policy Geoffrey Supran, Professor of Physics of the Oceans Stefan Rahmstorf and Professor of the History of Science Naomi Oreskes. It leverages a thread tweeted by Geoffrey Supran shortly after publication on January 12, 2023. Please note that the full paper will only be available open access for two weeks after publication.
Climate projections by the fossil fuel industry have never been assessed. On the basis of company records, we quantitatively evaluated all available global warming projections documented by—and in many cases modeled by—Exxon and ExxonMobil Corp scientists between 1977 and 2003. We find that most of their projections accurately forecast warming that is consistent with subsequent observations. Their projections were also consistent with, and at least as skillful as, those of independent academic and government models. Exxon and ExxonMobil Corp also correctly rejected the prospect of a coming ice age, accurately predicted when human-caused global warming would first be detected, and reasonably estimated the “carbon budget” for holding warming below 2°C. On each of these points, however, the company’s public statements about climate science contradicted its own scientific data.
The peer-reviewed paper Supran et al. (2023) published in Science on January 12, 2023 shows Exxon scientists predicted global warming with shocking skill & accuracy between 1977 and 2003, contradicting the company’s decades of climate denial. It is the first ever systematic assessment of an oil and gas company’s climate projections.
The authors began by collecting every available global warming prediction reported by Exxon scientists in internal memos and peer-reviewed papers. This yielded 16 temperature projections from 1977-2003 in 12 graphs, shown here with subsequent *observed* temp changes overlaid in red.
For example, below is a graph from a “proprietary” 1982 Exxon memo showing “an estimate of the average global temp increase” from 1960-2080 due to future fossil fuel burning. Overlaid are:
Blue = Observed CO2
Red = Observed temp change
Immediately, the overlap is striking.
Next, the authors digitized all 16 global warming projections reported by Exxon scientists, and plotted them on one graph:
Gray = Exxon’s predictions
Red = Observed temperature change
This, in a single image, is what #ExxonKnew:
The analysis allowed the authors, for the first time, to put a number on what #ExxonKnew: that burning coal, oil, and gas was going to heat the planet by 0.20° ± 0.04°C every decade.
And they were right on the money…
Using established IPCC statistical techniques to test the performance of Exxon’s temperature predictions in comparison to historical observations, the authors find that most of Exxon’s global warming forecasts were accurate (63-83%, depending on the metric used).
Exxon’s global warming projections were also CONSISTENT with, and AT LEAST AS SKILLFUL as, those of independent academic and government models. Here are the stats…
CONSISTENT: Exxon’s average projected warming (0.20° ± 0.04°C per decade) was, within uncertainty, the SAME as that of independent models (0.19° ± 0.03°C per decade).
SKILLFUL: Global warming projections modeled by Exxon scientists had an average ‘skill score’ of 72 ± 6%, with the highest scoring 99%.
For comparison, NASA scientist Dr. James Hansen’s predictions presented to US Congress in 1988 had skill scores of 38 to 66%.
Accounting for differences between forecast and observed atmospheric CO2 levels, the ‘skill score’ of projections modeled by Exxon scientists was 75 ± 5%, with seven projections scoring 85%+. Hansen’s 1988 projections had corresponding skill scores of 28 to 81%.).
The authors also show that Exxon scientists:
- Correctly rejected the prospect of a coming ice age.
- Accurately predicted when human-caused global warming would first be detectable.
- Reasonably estimated the ‘carbon budget’ for holding warming below 2°C.
In public, however, ExxonMobil contradicted each of these conclusions.
Here’s ExxonMobil repeatedly exaggerating the uncertainties of climate science and modelling:
ExxonMobil has also often denigrated climate models. In 1999, CEO Lee Raymond said future climate “projections are based on completely unproven climate models, or, more often, sheer speculation.”
In 2013, CEO Rex Tillerson called climate models “not competent.”
And here’s Mobil and ExxonMobil cultivating the myth that in the 1970s, there was a scientific consensus that the planet was going to cool down rather than heat up:
In sum, this study is #ExxonKnew2point0. It puts a number on, and paints a picture of, what #ExxonKnew, which is that burning fossil fuels would heat the planet by 0.20° ± 0.04°C per decade.
We now have airtight, statistically rigorous evidence that Exxon accurately predicted global warming years before it turned around and publicly attacked climate science and scientists.
In that sense, this graph (Figure 3 from above) doesn’t just communicate a crisis, it confirms complicity.
Supran, G., Rahmstorf, S. & Oreskes, N. (2023). Assessing ExxonMobil’s global warming projections. Science. 379(6628). Article
Also see the 2019 Skeptical Science post: In 1982, Exxon accurately predicted global warming
Upon publication, the paper triggered a flurry of articles, some of which are listed below:
New Scientist Podcast – Episode #158 Exxon’s 1970s predictions for climate change were super accurate (17min)
Bill McKibben – Godalmighty, Exxon Knew Absolutely Everything