Why did flood defences fail against Storm Desmond?

The recent floods in Cumbria are a dramatic illustration of the risks faced by communities who live under the protection of flood defences. That is, that such defences can only provide protection up to a particular capacity, and when that capacity is exceeded, flooding will occur.

A map of the British Isles showing rainfall for the period 30 November to 7 December 2015
A map of the British Isles showing rainfall for the period 30 November to 7 December 2015

So the first question we might ask about the recent floods is, how well have the various communities been served by the flood defences so recently installed? Was the considerable investment well spent? Could more have been done?

Why didn’t defences work?

Without doubt, the amount and intensity of rainfall was considerable, resulting in run-off from the catchment that was well in excess of the capacity of the defences. This was not a matter of the water simply lapping over flood defence walls and embankments; the amount of overtopping was considerable.

Flood defences are typically designed for a specified extreme flood event, for example, an event that would occur on average once every 100 years. So in 2015, there would have been a 1% chance that such an extreme event would occur.

Equally, and rather depressingly for the inhabitants of Cumbria, there will also be a 1% chance of such an event occurring again in 2016.

Thus it is certainly possible (though unlikely) that we can have a number of extreme events in a relatively short time period, as has been the case in Cumbria. But we should not simply explain this away with statistics without further thought.

Is it the result of climate change?

Our estimates of rainfall frequency are based on a relatively short period of historic rainfall record, compared with the frequency of events that we are trying to predict. This means there is considerable uncertainty in our quantification of the 1% annual probability event.

However, we do know that the climate is changing. This may not only be due to global warming, but may also be influenced by cyclical climate trends. We need to understand more about this.

There is also the catchment area itself to consider. There is an important difference between the probability of rainfall and the probability of a flood. What was the state of the catchment at the start of the recent events? How much was run-off influenced by preceding rain? Was it unusually wet and were any lakes unusually full? What was the state of the landscape – has this changed significantly? Are sediments an issue? What is the state of channel maintenance?

It is clear that in at least one instance the blockage of a bridge opening by debris was a contributing factor.

There is certainly a case for a detailed review to determine how all of these facets should influence the design parameters we use for flood defence works in the future.

David discusses the Cumbria flooding crisis on the BBC News

Explore more of our knowledge on flood defences and control

About the author

ICE President 2014/15, David Balmforth
Professor David Balmforth

Professor David Balmforth is a national flooding expert, currently an Executive Technical Director with international consultants MWH. His work covers all aspects of urban flood control, pollution management and climate change adaptation in the wastewater sector, and he is currently advising on major flood alleviation projects in London and Singapore.

An independent advisor to the UK Water Regulator (OFWAT) and the Department for Environment, Food and Rural Affairs, and to Thames Water and the Greater London Authority, David is a visiting professor in the Department of Civil and Environmental Engineering at Imperial College, London.