Das Model

Did you know that the global economy would be $28 trillion richer were it not for the oil industry? Not because said industry could have produced even more oil than it did. It’s because of emissions. And extreme weather, because emissions cause extreme weather on the planet whose climate some clearly clueless individuals have called “a coupled non-linear chaotic system,” italics mine. And that $28-trillion bill is only from the heatwaves.

Okay, now you know that the global economy has missed out on $28 pretty trillions because of extreme weather that, according to none other than the International Intergovernmental Panel on Climate Change could not be predicted at all. But, according to two other individuals, one from Stanford and the other from Dartmouth, the world — or rather, certain parts of the world — can recoup some of these tremendous losses by suing the oil industry. Because they’ve calculated exactly how much each of 111 oil companies is responsible for in that $28 trillion. Please give a warm welcome to Christopher Callahan and Justin Mankin, and their Carbon majors and the scientific case for climate liability paper.

The paper claims the alleged damage caused by oil and gas production to the “coupled non-linear chaotic system” that our climate is can be quantified and the authors have taken the pains of quantifying it. In case anyone’s curious, Aramco is on the line for $2.05 trillion in economic losses from extreme weather and Gazprom is responsible for $2 trillion, and I’m sure both these companies’ owners are really worried about that. Big Oil majors are all below $2 trillion but not by much, at least for Chevron, BP, and Exxon.

And how did Callahan and Mankin come to these financially devastating calculations? By applying attribution science to oil production and consumption data. Attribution science is possibly the freakiest brainchild of the climate change movement — and the potentially most rewarding one, literally. Because Callahan, Mankin and many others like them hope to sue the pants off Big Oil for those $28 trillion.

The artistic, I mean scientific, process involves using actual historical data from a really independent database called Carbon Majors, which is “a platform operated by the global, nonprofit think tank InfluenceMap,” which in turn is “An independent think tank producing data-driven analysis on how business and finance are impacting the climate crisis.” Not even a hint of a sliver of a possibility for a bias anywhere here.

On the basis of that historical data — self-reported by Big Oil, at least— the authors of that monumental paper did some modelling and some simulation, meaning a lot of modelling and a lot of simulation. In their own words, “Source attribution often uses an emissions-driven climate model to simulate historical climates and counterfactual climates, in which the latter is the same as the former, save for the removal of one emitter’s time-varying emissions (that is, a ‘leave-one-out’ experiment). The difference between the two simulations represents the contribution of the removed emitter, providing a test of ‘but for’ causation: but for the emissions of this actor, the climate would have been thus.”

That’s straightforward enough although I would’ve appreciated an explanation of the “emissions-driven” bit. Yet, you see, that sort of a climate model is “opaque and computationally expensive. A computationally tractable approach is to use reduced-complexity climate models (RCMs) that accurately simulate the behaviour of the Earth system using a smaller number of equations.” One wonders why anyone would need the computationally expensive, opaque models when there are reduced-complexity models but they’re probably for researchers with more money than these two. Also, this cannot be stressed enough, we’re talking about a chaotic system.

The paper features two magic words in abundance: model and simulation. There’s been a lot of simulation in the attribution art, I mean science, field —peer-reviewed, no less. These are serious people who’d never abuse science in any way, especially when they quantify the “socioeconomic impacts of climate change”. That’s why, to quantify them, they drew on “Recent peer-reviewed work [that] has used econometrics to infer causal relationships between climate hazards and outcomes such as income loss, reduced agricultural yields, increased human mortality and depressed economic growth.”

Peer-reviewed is another magic word that gets generously sprinkled over such papers even though some irreverent members of the scientific community have been known to dismiss peer review as something similar to an author giving their first book to a group of friends who’d never hurt the author’s feelings by being critical of the work. In the climate research case, it’s not just delicacy at work, of course. In climate research, it seems to be rather a case of “We’re in this together and we don’t want to lose our funding so let’s forget about all the factual data pointing to reduced mortality and higher yields over the decades since the Industrial Revolution”.

This might be a good time to recall, once again, that climate definition produced by the IPCC and add the rest of the quote. Here it is in its entirety: “The climate system is a coupled non-linear chaotic system, and therefore the long-term prediction of future climate states is not possible. Rather the focus must be upon the prediction of the probability distribution of the system’s future possible states by the generation of ensembles of model solutions.”

Models again. Models are everywhere in climate science because see above. Yet while the IPCC is honest enough to state the fact of the climate’s unpredictability, it has been really active in burying this fact beneath tonnes of modelling and simulation that has left a significant number of people with the impression that not only can future changes in the planet’s climate be predicted but that the prediction is 100% accurate, flawless and inarguable.

All this is based on models. That use data fed into them by humans. Who, by their own quite open admission, are the opposite of impartial. So much opposite, in fact, that they have essentially reversed the scientific method, which, per Oxford Language is “a method of procedure that has characterized natural science since the 17th century, consisting in systematic observation, measurement, and experiment, and the formulation, testing, and modification of hypotheses.” Italics mine again.

These days, instead of systematic observation, measurement, and experiment, we get data manipulation, cherry-picking, and modelling and simulation, along with the rather useful concept of a global average temperature that unrelenting old-fashioned scientists have debunked repeatedly as nonsensical for the very simple reason that it’s about as relevant to anything as the average weight of all humans on Earth.

Yet the anti-scientists now want to use their non-scientific methods, models, and simulations — peer-reviewed, of course — to sue the oil and gas industry. Personally, I’d love to see anyone sue Aramco or Gazprom for emissions at, say, a California court. But here’s the even funnier part that lays bare the multiple… let’s call them shortcomings of the model-based approach to climate science, as expressed, perhaps unwittingly, by the Washington Post.

“Some legal scholars say introducing this sort of calculation into a courtroom could devolve into a “battle of the experts,” with each side hiring scientists to run different models with different assumptions to reach conflicting conclusions about a company’s share of responsibility for climate change and its harms.” I’m really into italics today.

In a sane world, that sentence would be enough to blow up the whole idea of attribution art as a form of science, whose conclusions are worth the electricity that was expended on their formulation, let alone anything more. If you can make different models with different assumptions about the same events and come out with conflicting conclusions, what is the worth of a model? Not great, that’s for sure.

But of course we live in a world where one government wants to cover its country with solar panels and dim the sun at the same time, in order to achieve the singular purpose of reducing carbon dioxide emissions, and another wants to spend 500 billion euro on tanks and transition but can’t because of spending limit rules that its predecessors pushed for and got approved. The new guys simply… forgot about this.

In such a world, modelling and simulation appear to be the best way to do science. I mean, who has time to actually observe stuff when there are such handy reduced-complexity climate models, right? Especially when observation — not to mention unadulterated measurement — threatens to disprove your hypothesis that you’ve spent so many sleepless, climate anxiety-fuelled nights to formulate.