By Jeremy Williams

As every good weather reporter knows, it's not wise to blame any single event on climate change. Climate is not weather. It's often repeated, and it matters. If campaigners can point at every storm or flood as evidence of a changing climate, sceptics can point at every snowfall and cold snap to claim the opposite.

Nevertheless, we know that rising emissions are affecting the weather, and we need to get better at knowing what we can and can't say. And so a whole branch of climate science has emerged called 'attribution studies'. These studies analyse specific individual events, and establish how much they can be attributed to climate change. One of the pioneers of this method is Friederike Otto, and she describes how it works in her book Angry Weather: Heat waves, floods, storms, and the new science of climate change.

When a natural disaster occurs, everyone wants to know if it's climate related. Scientists want to give an answer, but there's no time to study it while it's still happening, or to run their findings through the traditional system of peer review. Rapid attribution analysis is a highly specialised response, calling on networks of scientists and their data sets, and modeling the event with and without climate change. By comparing the two, researchers can then say whether or not an event was influenced by climate change. Sometimes it might have made an event twice as likely. Sometimes it might be much more than that – such as the Siberian heat wave last year that was judged to be practically impossible without climate change.

There are times when the fingerprints of climate change cannot be detected, and that's important too. It's all part of the learning. Some events, such as heat waves, are easier to attribute than others. Certain parts of the world are trickier to model, such as East Africa. And non-events can also be notable. A November without frost might occur every 20 years or so, but that would be more like a one in 1,250 year event without climate change. What matters is to avoid jumping to conclusions according to what's politically convenient – in whatever direction that might be. "In an increasingly complex world, simple answers can be seductive."

Angry Weather describes why this science is necessary and how it is conducted. That could be a rather dry matter of scientific procedure (and there was a moment where it briefly lost me while "parameterising the variables") but the book brings it to life with the case study of the Houston floods. Each chapter tells a bit more of the story of the science behind the scenes, alongside events on the ground.

What I found most fascinating is how attribution studies can be used practically. They allow governments and local authorities to plan ahead and understand environmental change. We can use them to manage risk. They're also a tool for climate justice. Where a climate connection can be made, responsibility can be taken. That includes court cases against major polluters, and the book includes examples. Attribution studies "put climate science on the offensive, rather than the defensive. We can state whether and to what extent climate change is manifesting in our weather. We can stand up to the energy companies and mercenaries of doubt."

I'm interested to see how this is applied in holding fossil fuel companies to account in future, and securing compensation for natural disasters. "Attribution science is only just getting started" she writes, and Otto's book takes us behind the scenes of an emerging science that is going to make waves in the years to come.

First published in Earthbound Report.

 This post is ad-supported
​