Massive citizen-powered climate simulation links winter floods to global warming

Going under?

Massive citizen-powered climate simulation links winter floods to global warming
Simon Redfern, 1 May 2014 theconversation.com

Britain’s warm, wet winter brought floods and misery to many living across southern England, with large parts of Somerset lying underwater for months. When in January rainfall was double the expected average over wide areas, many people made cautious links between such extreme weather and global climate change. There were nay-sayers at the time but it now seems that there is evidence for those links.

Speaking at the European Geosciences Union annual meeting here in Vienna, Myles Allen, a professor of geosystem science at the University of Oxford, presented his take on the issue. At the gathering of more than 12,000 geoscientists, Allen reported an ambitious computer experiment that his team has undertaken over the last two months to test whether the winter floods could be attributed to climate change. And it seems that they can.

The floods of January 2014 certainly were extreme. According to Oxford’s records of daily rainfall, they were unprecedented in 250 years. The records at the UK Met Office from the 20th century also show that this winter was, historically, uniquely bad.

Flooded rivers wash huge amounts of sediment into estuaries around southern Britain in February.

The IPCC report does suggest that extreme weather events should be expected as the world warms but the prediction is couched in cautious terms and the risk is assessed as “medium” confidence.

At the Environmental Change Institute in Oxford, researchers Nathalie Schaller and Friederike Otto analysed results from almost 40,000 climate model calculations to test the impact of climate change on Britain’s winter rains. Their calculations modelled the weather across the country on a 50km grid. They compared the results of 12,842 simulations based on the current global sea surface temperatures, with 25,893 results computed on the assumption that global warming had never occurred – that fossil fuel burning had not raised CO2 to today’s levels and ocean surfaces were cooler.

Such a huge number of calculations was needed to tease out the statistical differences between the two scenarios. It was only possible through the participation of thousands of members of the public in the work’s biggest ever climate modelling exercise: they offered up spare processing capacity on their home computers to run the calculations via the Climate Prediction citizen science climate modelling programme.The difference between observed winter (blue) and climate change-free simulated winter (green) shows increased seasonal rainfall and greater likelihood of extreme rainfall.

The results showed a subtle bias towards more extreme weather in today’s warming world. Events that would have been expected once in 100 years before global warming can now be anticipated to occur once in 80 years. In essence, the probability of extreme winter floods appears to have increased by 25% on pre-industrial levels.

Allen pointed out that this is the first quantitative evaluation of the influence of global warming on Britain’s 2014 floods. Thomas Stocker, a professor of climate and environmental physics at the University of Bern and chairman of the IPCC working group charged with assessing the physical origins of climate change, said that the Oxford group’s results had “shown movement in one direction only – toward greater risk”.

Although the results from the models cannot yet give definite measures of the probability of a flood, they do provide an insight into how those risks have changed and continue to change – information that is of great interest to insurance underwriters, among others.

Otto said: “Past greenhouse gas emission and other forms of pollution have loaded the weather dice", adding that she and others were still working on investigating the implications of the results, for river flows, flooding and ultimately the threat to property and lives.

Some will, no doubt, question the result on the basis that it is “simply” a statistical test. The results from the two modelling scenarios are, at first sight, very similar. But the fact remains that they are distinct, showing that rising global ocean surface temperatures directly influence UK winter rainfall.

The results affirm the strong and growing scientific consensus developing from the understanding of the physical origins and consequences of climate change, as outlined in the IPCC’s Fifth Assessment Working Group 1 report last September. Those that choose to ignore them, or contradict them, will (I predict) still be directly affected by them. And we will be hit where it hurts most – in our wallets. How likely is it that the insurance industry will ignore such results?