The counterfactual world of climate attribution

It can’t be said too often just how perplexing the global temperature of the last year or so has been. Not since the debate about the so-called global temperature ‘hiatus’ earlier this century has there been so much consternation amongst climate scientists.

Nobody knows why the past year or so has been so hot. Scientists display a wide range of opinion, including the possibility that we may have entered a new phase of global warming, or that the blame might fall on solar effects or a lifting of the aerosol burden. There is also the suggestion that it’s all a consequence of the El Niño, even though temperatures were high even before the El Niño was declared in June 2023. Others have suggested that it’s not a climate event, and shouldn’t be considered as one as a single year doesn’t make a climate. It is noteworthy that this view comes from the UK Met Office, which has not always been so clear on the distinction between weather and climate.

If climate models cannot explain the recent temperature, you might be surprised to discover that scientists are using them to attribute the warmth of the past year to human-induced climate change. This is a curious example of circular reasoning. How can you attribute it to human-induced climate change if you don’t know if it was caused by human climate change? Clearly to do this you have to ignore a lot of things.

Andrew Pershing, vice president for science at Climate Central has done just that, saying, ‘This 12-month record is exactly what we expect from a global climate fuelled by carbon pollution.’  We have seen that this is not a view held by everyone, but it is necessary to connect 2024 to human-induced climate change.

Welcome to the counterfactual world of climate attribution.

Did it happen, or didn’t it?

Some background. Once, the most we could say about a particular weather event was that it was ‘consistent’ with climate change, which to some is not much use at all. Things started to change in 2004 when the first attribution paper was published in Nature by Peter Stott of the UK Met Office. He argued that the odds were doubled for the European heatwave in 2003 because of human-induced climate change. Since then, Attribution Science has grown, with over 400 analyses published on specific events. Such is its power that you can sometimes make ‘advance attributions’ about events coming up in the next few days. You can even attribute a climate event that should have happened but didn’t!

Attribution is regarded as straightforward for heat waves, but anything else is very problematical. Wildfires are complex events and are not easy, as is attributing hurricanes, at least up until now.

Despite these caveats, such powerful science must be championed. One big reason for its profile outside scientific circles is the work of Climate Central, a US non-profit pressure group operating out of  Princeton University. They started the World Weather Attribution Program, which is frequently the basis for articles in the media.

An example. Take a map and show how warm it is above some average temperature. Then add another layer – the climate shift index (CSI) – suggesting how human climate change could have produced that anomaly. The CSI regularly finds its way into weather bulletins as part of a service Climate Central provides for broadcast meteorologists.

Recall that hurricanes were difficult to attribute because, even though they derive their energy from elevated ocean temperatures, they stubbonly show no sign of increasing in intensity or frequency. In a few months, Climate Central is going to be rolling out ocean attribution temperatures, and by the end of the summer they hope to be able to make attribution statements about individual hurricanes. That should be interesting because no single hurricane, or season of them can be attributed to human climate change, unless you want to.

Then there is Michael Mann’s dramatic hurricane prediction for the coming season. He predicts 33 named storms this hurricane season, which runs from June 1st to November 30th. This is quite an increase on the 20 named storms of last year, which was itself the 4th most active year, tying with that well-known human-influenced climate year of 1933!

Whatever happens to hurricanes this season, attribution science will have an explanation, and to prove it’s a science to be taken seriously it will also attribute an index to it.