Can we still predict the future from the past? Implementing non-stationary flood frequency analysis in the UK

Last week our colleagues published a new research paper in The Journal of Flood Risk Management (Chartered Institution of Water and Environmental Management) in collaboration with Peter Spencer from the Environment Agency.

Paper Title: Can we still predict the future from the past? Implementing non-stationary flood frequency analysis in the UK

Their paper explores the underpinning assumptions made when calculating flood frequency analysis in the UK. Those assumptions have been until recently, that flood frequency analysis is stationary and the probability of a flood event happening does not change over time.

Flood frequency analysis

Over the last 20 years in the UK flood frequency analysis has used techniques which assume that annual maximum flow is independent (when considered as a data series), and that it has the same probability distribution as all other values. However if the flood frequency of a catchment is not constant over time, in other words it is non-stationary and peak flows are not identically distributed, this assumption cannot be applied.

There is a lively debate in the research literature about the merits and drawbacks of stationary and non-stationary analysis.

“In this paper we first examine the scale of the problem by presenting results of trend tests across Great Britain and discussing potential reasons for the trends. We then give an introduction to techniques for non-stationary flood frequency analysis and present findings from a case study in north-west England.”

Why is this important?

The Environment Agency in England is half-way through investing £2.5 billion over a 6 year period with the aim of reducing flooding from rivers, the sea, groundwater and surface water to at least 300,000 homes by 2020/21. Flood data analysis underpins the decisions on whether to invest and how high to build, and these are usually made on the basis of stationary flood frequency analysis that assumes the probability of flood flows is unchanging over time.

Following a series of damaging flood events in parts of the United Kingdom in recent years, many people are starting to raise questions. North-west England was badly affected by flooding in 2005 (Carlisle), 2009 (much of Cumbria) and 2015 (much of Cumbria and Lancashire).

“People who are affected by such events can understandably be sceptical that two or even three extreme floods have occurred purely by chance so close together. This raises the question of whether something has changed so that the probability of such events is higher than it was previously. A related question, more difficult to answer, is how the probability might change in the future”

The full research paper can be read here.

Want to know more?

For more information about this paper and about non-stationary flood frequency analysis email Duncan Faulkner, Technical Director and Head of Hydrology. You can also find out more on our flood modelling webpages.



Leave a Reply