Evidence of humans’ role in deluges starts to flow

Flooding in York, UK, in Autumn 2000. Credit: Gordon Hatton

Human greenhouse gas emissions made heavy rain and snowfall and local flooding more likely, scientists have shown this week. On Thursday top scientific journal Nature published two separate papers that are among the first to show that humans are contributing to the deluges that affect us.

Although global warming has previously been associated with more heavy rainfall, a lack of daily observations mean that prior studies have mainly relied on comparing different climate models. By contrast, Francis Zwiers of government ministry Environment Canada and University of Victoria, and colleagues instead compare observed and simulated changes. They looked at extreme rain and other wet weather brought together under the general grouping “precipitation”. Using data from 6,000 weather stations collected by the Hadley Centre of the UK’s Met Office, they looked at the highest daily and five-day consecutive precipitation amounts in any given year from 1951-1999.

“The work that we did was to examine how annual precipitation extremes change over Northern hemispheric land areas,” Zwiers told a press conference on Wednesday. “We see, as many others have done, that these events intensified over the latter half of the 20^th century.” Zwiers and co-workers from Canada and Scotland also found that there is a pattern of change in the observed data that matches those seen in models simulating greenhouse-gas driven global warming.

“That suggests that humans influence the intensity of precipitation extremes,” Zwiers said. The team also found that observed changes could not be explained by natural internal climate fluctuations alone. “Prior to this time there had not been a study that had formally identified this human effect. This paper provides specific scientific evidence that this is indeed the case.” Additionally, the scientists saw that the observed changes were larger than those predicted by climate models. “That raises some concerns about projections of the future change in these extremes,” Zwiers said.

The method Pardeep Pall and his colleagues used to determine the likelihood that flooding in the UK in the year 2000 was influenced by human greenhouse gas emissions could also be used to study other extreme weather events. Credit: Met Office

Separately, Pardeep Pall and colleagues compared climate model simulations to study the UK floods of October and November 2000, the wettest autumn in England and Wales since records began in 1766. “We looked at how greenhouse gas emissions due to human activity have affected the odds of floods occurring in England and Wales,” Pall, from ETH Zurich, the Swiss Federal Institute of Technology, told the press conference. “We found that these emissions substantially increased the odds of these floods occurring in 2000, with a likely increase of about a doubling or more.”

Pall and colleagues from the UK and Japan generated several thousand simulations of autumn 2000, using detailed climate models from the Met Office. “To do those thousands of repetitions would have been pretty tough to do all by ourselves, so we actually asked members of the public across the world to run our simulations for us, using their idle time,” Pall explained. “This was done using the climateprediction.net project. This is the first time that this kind of technology has been used to investigate a change in a specific type of damaging weather event.”

They looked at both realistic conditions and conditions where greenhouse gas emissions and resulting large-scale warming never occurred. In nine out of ten cases their model indicates that twentieth-century human-produced greenhouse gas emissions increased the risk of flooding in England and Wales by more than 20 per cent and in two out of three cases by more than 90 per cent. Pall’s team suggests that their approach could be useful for assessing how external factors may be affecting the likelihood of other types of extreme weather.

Co-author Myles Allen from the University of Oxford expressed his pleasure about the fact that Pall is now set to advise the Met Office to use this approach to determining whether extreme weather is affected by what we do. “It is a reasonable question, which you guys ask us all the time, ‘Is human influence on climate anything to do with this nasty bit of weather we’re having?’”, Allen said. “Answering it isn’t easy, but plenty of people seem to be perfectly happy to blame human influence on climate for bad weather, or equally confidently to say human influence on climate has nothing to do with bad weather. The point is we can address these things systematically.”