Weather@home 2014: the causes of the UK winter floods

Thank you for your help to answer the question of whether the recent UK flooding can be linked to climate change (they can,  here are our summary findings).

Results are published in Nature Climate Change .  Find an overview of the results here.


Nathalie Schaller explains the science behind this new project, and why we need your help to answer these important questions about the link between extreme weather and climate change:

The accumulated rainfall of December 2013 and January and February 2014 has been the largest ever recorded at the Radcliffe Observatory in Oxford. Other parts of South England and Wales also experienced extreme rainfall amounts during these three months, and as a consequence, large areas were flooded, some more than once during that period.

The floods affected thousands of people. The question that many are asking is whether this level of extreme rainfall and the resulting floods are linked to climate change. Answers to this question have been expressed from various sides but there has, as yet, been no specific scientific analysis (since the “event” has only just ended). While basic physics tells us that, on average, warmer air can contain more water and thus the heaviest individual downpours might be expected to become more intense, it does not follow that accumulated rainfall over a month or season should increase in any particular region.

It will never be possible to say that any specific flood was caused by human-induced climate change alone, in the sense that, without climate change, that flood would not have happened. Occasional floods have been a feature of British weather since time immemorial. We can, however, ask and answer the question how the odds of getting an extremely wet winter has changed due to man-made climate change: have past greenhouse gas emissions and other forms of pollution “loaded the weather dice” towards (or perhaps even away from) and event of this nature?

The challenge is that, on any measure, it appears this was a relatively unusual event, however the dice have been loaded, and all our experiments to date suggest that the impact of any loading will be subtle. So, in effect, we need to find out if we are getting “too many double-sixes” with the British weather dice. The chance of a double six with a good pair of dice is one in thirty six, so to check if we are getting slightly too many of them, you need to roll the dice hundreds of times (unfortunately, for the weather dice, we can’t work out how the odds of a rare event have changed simply by looking at more frequent events, like a single six). So, in effect, to pin down the role of human influence on climate in the winter that has just ended, we need to roll the weather dice again thousands of times.

So, to do this, we asked for the help of the general public. We ran two very large “ensembles” of weather simulations, one representing conditions and “possible weather” in the winter we have just had, and one representing the weather in a “world that might have been” if we had not changed the composition of the atmosphere through greenhouse gas emissions. By comparing the numbers of extreme rainfall events in the two ensembles, we can work out if the risk of a wet winter has increased, decreased or been unaffected by human influence on climate. The ensembles need to be as big as possible to obtain robust estimates of the probability of rare events.

Read more about the Experiment Setup.

Read more about the Expected Results.

Project References.

See the results of this experiment.