WHY do we seem to have such trouble building flood defences to cope with nature鈥檚 extremes?
The chaotic behaviour of weather systems makes it impossible to accurately predict rainfall, river flows and the like more than a few weeks in advance. However, hydrologists tend to assume that these variables fluctuate randomly in the long run, which means that their average value, or the probability of exceeding a given threshold, can be estimated accurately from lots of observations. What鈥檚 more, these results do not change over time 鈥� a property known as 鈥渟tationarity鈥濃�� and so form the basis of flood defence plans.
Advertisement
But nobody who has carefully and doggedly monitored river basins would say that their behaviour is stationary in the above sense. After all, we interfere with practically all our rivers 鈥� drawing off water, building reservoirs, modifying channels and cutting down forests, for example 鈥� and so alter both their flows and the surrounding land.
Rainfall records also show patterns that don鈥檛 look random, on a range of timescales 鈥� not just seasonal variations, but also long-term natural oscillations, some associated with climate phenomena such as changes in ocean circulation.
Although hydrologists take all this into account, anthropogenic climate change throws a new spanner in the works. Some are saying that 鈥溾��.
The physics of the atmosphere, oceans and ice sheets tells us that significant changes in these systems will affect rainfall, soil moisture and so on, in complex ways. For example, my colleague has found that human-induced warming boosts the chances of heavy rain, such as that experienced the UK , 鈥� by anything from 5 to 80 per cent.
The conclusion is that the past cannot be a reliable guide to the future. Can climate models help? To a degree, yes, but they are not yet accurate enough on a local scale to be a reliable guide for much flood defence planning.
This doesn鈥檛 mean that deluges of the kind we have recently seen in the UK are bound to occur each winter. Climate change is loading the dice towards flooding, but we are still likely to have dry years and even droughts.
This all makes life very hard for designers of flood defences and reservoirs. They will have to model how their systems respond to a wider range of conditions. They will also have to devise ingenious schemes that do not fail catastrophically, even when surprising patterns of rainfall occur. And they need to think about how to build in capacity to adapt to unforeseen change.
More surprising stuff is going to occur 鈥� that鈥檚 what happens when you kill stationarity.
(Image: Andrzej Krauze)
This article appeared in print under the headline 鈥淒eluged with doubt鈥�
