The past few decades have seen numerous damaging floods across England. We can debate whether or not this is exceptional, and why, but many people are already wondering who will next be affected.
The Environment Agency provides probabilities of flooding based on present conditions – its maps show the seven per cent of the country at risk of flooding. Medium- and high-risk areas are those with a yearly flood probability of greater than one in 100 or one in 30, respectively.
These may not sound like short odds, but having your home flooded by what is, in effect, dilute sewage is a terrible experience. Furthermore, flood damage happens much more frequently somewhere in the country (the odds are specific to an individual site), and in a warming world, things are likely to happen more frequently in future. So can we say who will be next to be damaged?
The first thing to be clear about is that we’ve increasingly put ourselves in harm’s way. Since the late 19th century, floodplain development has taken off, and damage has increased across many English floodplains, regardless of whether over- bank flooding has become more frequent.
Second, although unusual precipitation may be the ultimate cause of rivers overflowing, how and where this happens varies. Floods may arise from snowmelt, especially if there’s a rapid thaw; from slow-moving low-pressure systems or intense precipitation ‘super cells’; and from summer thunderstorms covering relatively small areas. This year’s floods reminded us that the cumulative build-up of groundwater from a succession of individually unexceptional events can also cause inundation in permeable-rock catchments. Small-scale irregularities of floodplain shape are also important in conveying and channelling floodwaters, extending to create additional flow zones along shallow depressions that may hardly have been noticed beforehand.
Hence, inundation, damage and warning times have a strong geographical component. The steep, narrow valleys of the Pennines can experience both winter- snowmelt and summer-thunderstorm flash flooding with little warning. Valleys downstream have wider floodplains, and major rivers there are largely regulated and flood embanked, but are still liable to cause inundation in extreme events. The reclaimed artificial landscapes of the former wetlands of the Humber, the Fenlands and Somerset Levels have to be carefully managed if extensive season-long inundation is to be avoided.
The spread of urban housing and industry, and the extension of rural communities, has added yet further damage liability, such that vulnerability isn’t at all uniformly distributed across England’s floodplains – in terms of potential warning times, flood extent and duration, and, ultimately, damage costs.
These varied vulnerabilities can, in part, be identified on the Environment Agency’s maps and explored on the ground – something that everyone and every school should do, whether to determine what communal protection currently exists (being aware that this can always fail in extremis), or as a basis for personal decisions about property protection.
What about the future? Extreme events are seldom entirely ‘unprecedented’, even if the word makes for good media copy and provides some exoneration for not having learnt to act with sufficient foreknowledge. Our Victorian ancestors also had to cope with floods.
Today, we do have limited long-term quantitative flood data to analyse, and rapidly improving modelling capabilities for future forecasting. But translating ‘global warming’ into hydrological predictions for the flooding of very varied catchments through the behaviour of complex environmental systems is highly challenging. Flood frequencies are also likely to change – so is an educated guess possible for what we might at least expect in qualitative terms, and where?
Because precipitation events are local and regional rather than nationwide, it’s virtually impossible to precisely predict where flooding will take place. After the 2007 floods, the Pitt Report highlighted the problem of urban flash flooding arising from rapid runoff, and the Environment Agency usefully provides separate maps for ‘surface water’ flood liability. But it’s likely that future destructive floods will also occur along populated floodplains for other reasons. One can point to some of the floodplain ‘usual suspects’ – the Trent and Severn – but places that haven’t suffered so badly in recent memory (Manchester and Leicester, perhaps) are just as susceptible to rare events where their flooding probabilities are equivalent and protection is inadequate. Similarly, a summer thunderstorm may affect one river but not the adjacent one.
Hence, the answer to the initial question has to be that we can’t precisely say – but it’s better to be precautionary than complacent if living in a potential flood-scape. Sooner or later, physically possible rare events happen as the odds suggest, wherever.
This story was published in the June 2014 edition of Geographical Magazine