A new technique connecting individual weather events with the impact of greenhouse gas emissions could bring climate change into everyday weather reports
"Well, the record-breakingly hot summer is showing no sign of cooling down. No thanks to us: the heatwave was made 35 per cent more likely by human greenhouse gas emissions."
CLIMATE scientists tend to shy away from assigning blame for extreme weather events like the fictional heatwave described above. But that may be about to change, thanks to a new type of climate study that can connect individual weather events with the impact of human-made greenhouse gas emissions.
So far, these studies have been done retroactively, a year or more after specific extreme events. But the latest techniques are making it possible to examine the role that climate change played in shaping the season that has just past – whether it was a scorching summer or a particularly wet winter, for instance. The ultimate goal is for this to happen in real-time so that climate analysis can become part of the daily weather report.
"Explaining why we're getting the weather we're getting should be part of the job of meteorological offices, as well as predicting it," says Myles Allen at the University of Oxford. Allen was part of a team that carried out pioneering research that examined the impact of greenhouse gas emissions on weather. After a deadly heatwave in 2003 contributed to the deaths of some 70,000 people in Europe, the team used two simulations of the climate – one including human-made greenhouse emissions, and one without – to assess the likelihood of a heatwave of that scale striking where it did, when it did. They calculated that human emissions doubled the odds of it happening (Nature, doi.org/c7hxpt).
Since then, several studies have used similar methods (see "Blame warming?"), but they all have dealt with events long after they have left the public consciousness. Peter Stott of the UK Met Office, who also worked on the heatwave study, and Allen are part of a growing cadre of climate scientists who want that to change.
This year, Nathalie Schaller, also at Oxford, took a closer look at the unusual weather that the south-east of England was experiencing. Record rain fell at the Oxford weather observatory from late 2013 through to early 2014, and there was widespread flooding.
To explore whether climate change was making such precipitation more likely, Schaller and her team ran a similar experiment to Stott and Allen's. They used real-world data to simulate the season that had just passed, then stripped the data of the influence of greenhouse gas emissions and ran the simulation again. The scenario was simulated thousands of times in order to calculate the odds of getting a bout of extremely wet weather at that particular time of year.
They concluded that what was a 1-in-100-year event without global warming had become a 1-in-80-year event. In other words, human emissions made the extreme levels of rainfall experienced in south-east England 25 per cent more likely.
The team's results were published online on 30 April, just two months after the flooding abated. To speed things up even more, a project called European Climate and Weather Events: Interpretation and Attribution (EUCLEIA), led by Stott and funded by the European Union, will test a new system. Instead of waiting for an event to happen, the idea is to incorporate seasonal forecasts, which are done a month or more ahead of time, into the climate models.
"One of the designs EUCLEIA is looking at is to use forecast sea-surface temperatures," says Allen. Sea-surface temperature is an important driver of the weather, and because the oceans change temperature very slowly compared with the air and land, they form a key, predictable component of seasonal forecasts.
In the new set-up, a real-world seasonal forecast driven by data on current sea-surface temperatures will be run alongside a simulated "no global warming" seasonal forecast, in which greenhouse gas emissions have been stripped out.
If an extreme weather event occurs, researchers can look to see if the models predicted it. If it was predicted in the real-world seasonal forecast but not in the scenario which is stripped of emissions, then it was made more likely by climate change – a likelihood that can be calculated.
For now, assigning a share of blame for past temperature anomalies like heatwaves and cold snaps is perfectly feasible, says Stott. Rainfall is more complex to model, though Schaller's study shows it is possible. Hurricanes and tornadoes require specialised models to predict, but Allen says that in theory they could be used to run the same type of experiment.
A main obstacle to bringing this kind of powerful climate modelling into standard weather forecasts is computing power. Models must be run many thousands of times to obtain statistically significant results, which requires expensive supercomputers.
Web-based distributed computing, by which models are run remotely on borrowed downtime on volunteers' personal computers – is one way around the problem. Roberto Mera at the Union of Concerned Scientists in Washington DC is currently using this approach to look at the role of climate change in the record-setting drought now hitting California.
- Subscribe to New Scientist and you'll get:
- New Scientist magazine delivered every week
- Unlimited access to all New Scientist online content -
a benefit only available to subscribers - Great savings from the normal price
- Subscribe now!
If you would like to reuse any content from New Scientist, either in print or online, please contact the syndication department first for permission. New Scientist does not own rights to photos, but there are a variety of licensing options available for use of articles and graphics we own the copyright to.
