Human influence on climate in the 2014 southern England winter floods and their impacts

Human-induced climate change increased the risk of severe storms like those that hit the south of England in the winter of 2013/14, causing devastating flooding and costing several people their lives.

The preliminary results of this study have been on our website since the time the flooding happened , but now we have looked not only at the rainfall, but also the influence of anthropogenic greenhouse gas emissions on the atmospheric circulation and how this propagates from rainfall, to river flow down to the direct impact of flooded houses in the river catchment zones.


We found that anthropogenic greenhouse gas emissions increased the risk of the once-a-century wet January in 2014 by 43% (uncertainty range: 0–160%). The increase in extreme rainfall that led to the flooding in 2013/14 was the result of two factors associated with global warming: an increase in the water-holding capacity of the atmosphere (a thermodynamic factor) and more January days with westerly air flow (a dynamic factor). The authors identified that approximately 2/3 of the increased risk could be attributed to thermodynamic changes in the atmosphere, and 1/3 to dynamic changes.

Among the worst-affected areas were Somerset, Devon, Dorset and Cornwall in the south west, and the Thames Valley in the south east. This first-of-its-kind, end-to-end study looked at the event from start to finish, taking in atmospheric circulation, rainfall, river flow, inundation, and properties at risk. The research is published in the journal Nature Climate Change.

To be able to do this world we made use of over 100,000 weather@home simulation of possible weather in January 204 in both the current climate and one in which there was no human influence on the atmosphere. Researchers analysed more than a hundred thousand simulations of possible rainfall in the UK run by citizens from all over the world.

For the hydrological modelling of the Thames river catchment done at CEH we showed that the changes in atmospheric circulation and precipitation caused higher peak 30-day river flow, while flood risk mapping revealed a small increase in flood risk for properties in the Thames catchment.

The heightened risk of rainfall found in the meteorological modelling led to an increase in the peak 30-day river flow of 21% (uncertainty range: -17–133%) and about 1,000 more properties at risk of flooding (uncertainty range: -4,000–8,000). These numbers show that we need to run the impact models as well as analysing the meteorology as the anthropogenic signal does not propagate linearly. Furthermore, each of these steps reflects different sources of uncertainty, but we find overall that there is a substantial chance of more properties having been placed at flood risk because of past greenhouse gas emissions, leading to potential damages that could be part of the losses incurred in 2013/14.

Some examples of the media coverage this work got:

« Back to News