The Science That Has Oil Companies Running Scared

For decades, the fossil fuel industry had a reliable shield against legal and political accountability: uncertainty. Climate change, the argument went, was a diffuse, global phenomenon. No single storm, flood or heat wave could be pinned on any particular company’s product. Harm was everywhere and nowhere, and that ambiguity was worth more to oil producers than any public relations campaign.

That shield is cracking. A fast-maturing field of research — often grouped under the labels attribution science and source attribution — is doing something the industry long insisted was impossible: drawing lines between emissions, warming and specific damage on the ground.

From “consistent with” to “caused by”

Climate scientists used to speak carefully. A heat wave was “consistent with” a warming world. A flood was “the kind of event we expect to see more often.” Those formulations were scientifically honest, but they were also legally useless and politically easy to dismiss.

Attribution studies changed the grammar. By running large ensembles of climate model simulations — one set representing the world as it is, another representing a counterfactual world without industrial-era greenhouse gas emissions — researchers can estimate how much more likely, or more intense, a given event became because of human activity. The results are expressed in probabilities rather than certainties, but probabilities are precisely the currency courts, insurers and regulators already deal in every day.

A parallel line of work traces emissions back to their origins, estimating how much of the carbon dioxide and methane in the atmosphere can be linked to the products of a relatively small number of large producers. Combine the two strands and you get something genuinely new: a chain of reasoning that runs from a company’s balance sheet to a particular neighborhood’s flooded basement.

Why that matters beyond the lab

Causation is the hinge on which liability turns. Plaintiffs suing over tobacco, asbestos or lead paint had to show not merely that a product was dangerous in the abstract, but that it contributed to the harm in question. Climate litigation has historically stumbled at exactly that step. Attribution research offers a way over the hurdle — or at least a way to get past the early motions to dismiss that have buried so many cases.

The implications ripple outward. Insurers and reinsurers, who price risk for a living, have obvious reasons to care about which losses are attributable to warming and who might ultimately be made to pay for them. Investors face questions about contingent liabilities that do not appear on any current financial statement. Governments weighing adaptation budgets — seawalls, cooling centers, water infrastructure — may increasingly ask why taxpayers alone should foot the bill.

The counterattack

Expect the science itself to become a battlefield. Modeling choices, the selection of counterfactual baselines, the handling of natural variability — all are legitimate subjects of scientific debate, and all are ripe for weaponization by well-funded litigants seeking to manufacture doubt. The playbook is familiar: fund competing analyses, attack individual researchers, insist that anything short of certainty is nothing at all.

But the trajectory of the field is not favorable to that strategy. Computing power keeps improving, observational records keep lengthening, and methods keep being tested against one another. Uncertainty is shrinking, not growing.

The oil industry spent a generation betting that no one would ever be able to connect the dots. That bet is looking worse every year — and the people doing the connecting are not activists with placards. They are scientists with supercomputers. Read More