Floods, forecasting and flawed policy

The recent flooding events in Cumbria demonstrate the problems of forecasting and policy development based on historical data.

Representatives from the Department for Environment, Food & Rural Affairs and Environment Agency and the Environment Agency have consistently referred to the rainfall experienced in the North West of England on 6th December as an ‘extreme event’. (For example Elizabeth Truss, the secretary of state for Environment, Food and Rural Affairs on the BBC Today programme on 8th December).

Another well worn phrase in this domain is ‘a hundred year event’. Or the once in 200 year rainfall experienced in 2005. Or the once in 800 year rainfall experienced in 2009.

The point being made is that the investments in flood protection since the last major floodings in 2009 could not be expected to deal with something as unprecedented as 300mm of rain within 24 hours because it is such a rare event (the monthly average rainfall for Cumbria in December is 146.1mm). But it is a rare event that has happened 3 times in the last 10 years.

Modelling and forecasting is central to the consideration of what is adequate protection. The Environment Agency will have modelled what impact different levels of rainfall will have on water levels and hence the likelihood of flooding given the protections in place.

It might be that those models were incorrect – however the implication from what is being said is that they were not. The flood protections were overtopped because the level of rain experienced was an extreme event. It was historically unprecedented and so either not something that was considered, or of such low probability that it is not worth defending against given the costs.

The demonstrates the challenge of forecasting purely based on historical data. Models used in many different domains rely largely on extrapolating trends from what has happened in the past. However in an area of change that may not be sufficient.

Modelling how well flood defences will stand up to the worst rainfall experienced over the last 100 years will not be enough. They need to be tested against the range of possible worlds that climate change may bring us in the future.

Those possible worlds are only apparent from exploring models that reflect the causal mechanisms at play in the real world.

The Met Office is responsible for building such models of the climate and interpreting them for the benefit of policy. Professor Dame Julia Slingo, Met Office Chief Scientist makes this point on their web site:

“…just as with the stormy winter of two years ago, all the evidence from fundamental physics, and our understanding of our weather systems, suggests there may be a link between climate change and record-breaking winter rainfall. Last month, we published a paper showing that for the same weather pattern, an extended period of extreme UK winter rainfall is now seven times more likely than in a world without human emissions of greenhouse gases.”

So in our most likely future possible worlds we might expect what we previously understood to be 100+ year storms a lot more often. These forecasts then should be fed into models of flooding in Cumbria and elsewhere – to be fair to Elizabeth Truss, she acknowledges the need for this.

That will reveal much about what is being avoided in the discussion about the floods. Specifically whether flood protection to deal with the likely future extremes of rainfall are affordable or whether other, potentially drastic, strategies might need to be considered by government, local authorities, businesses and individuals in affected areas.

Leave a comment

Please prove that you are human: