An old problem in a new form1 January 2003
Although flood defence is a familiar problem in the UK, climate change will make the issue far more important. New and expanded solutions w ill be needed, as Karl Hall reports
FLOODING is not a new problem for the UK, and being a natural occurrence it will always be with us. But it is currently the frequency and severity of significant flood events that are focusing attention on the issue, together with the real concern that future climate patterns will worsen the outlook.
Inland flooding is generally the result of high rates of run-off from land, occasioned by intense local rainfall or by longer-term heavy rain. Most people in the UK remember the 2000 floods as being notably bad: it was the wettest autumn since records began and resulted in wide-scale inundation as defences were overtopped or breached, and drainage systems overwhelmed. Around 10,000 properties were flooded. These floods were in some areas 1 in 200 year events (flood levels that would normally be considered as having a 0.5% chance of occurring), but the frequency of 100 or 200 year events is now increasing. Unfortunately, the historic nature of urban development has biased it towards rivers so many cities are now at considerable risk of regular and damaging flooding.
Hard-engineered (and therefore expensive) solutions appear to be the only options for alleviation, at least in the short term. This philosophy follows on from most previous thinking, which endeavoured to channel high-velocity flood water and discharge it to the sea in the shortest time. Although the thinking is understandable, this method of flood management has actually made some areas more vulnerable to severe flooding at little warning.
Research suggests that by the 2080s, winter rainfall may increase by up to 30%, with potential for greater incidence of flooding. At the same time, summer rainfall could decrease by up to 50%, particularly in the south. As the world climate becomes warmer, greater levels of evaporation in summer months may be translated into increased rainfall and perhaps also at times later in the year than was previously expected. There are additional factors involved, in a sometimes complex combination of events that include large-scale physical degradation of arable land, with increased run off from dried-out encrusted areas or saturated land. There may also be indeterminable natural factors at work which are influencing climate change and climatic 'surprises' such as changes in ocean currents, which could promote further unpredictability.
Assumptions about future climate patterns are, however, a complex matter, and the UK has undertaken much research in this area. The result is a range of predictive climatological models developed by the Meteorological Office's Hadley Centre under the auspices of the UKCIP. These assume a range of effects, based on various levels of greenhouse gas emissions, but whichever yardsticks are taken (low, medium low, medium-high or high, where medium-high, for instance, assumes a 1% per annum increase in CO2 emissions), the outlook is not favourable, and wetter winters and drier summers are predicted. There is additionally the problem of predictive confidence, and whilst global mean temperatures and CO2 levels are considered relatively predictable, factors such as climatic and regional variability are less so.
The difference between low and high predictions is very wide - it could be as much as 20% - so there is an imperative for more accurate research and modelling. The problem in accurate forecasting stems from the fact that a majority of existing data is based on historic records and it is widely accepted that this is well out of date. Interpolations derived from existing data are therefore inappropriate, as this would assume that the probability of a given rainfall event can be calculated and preventive actions can be taken. From this methodology, we would find that the only resolution is to design schemes to a probability of the worst case occurring approaching zero, which would result in substantially over-engineered designs.
The pathways for flooding are generally by river (fluvial) flooding, coastal, high groundwater levels and snowmelt. Sometimes there can be a combination of factors, for instance in coastal areas where a high tide combined with high fluvial flows and storm surge creates an unusually high water level, causing problems at both the tidal and inter-tidal zones.
What makes matters worse for the UK (although as we have seen from the summer floods in Europe, the problem is a global one) is the large amount of urban development carried out in flood plain areas. Current estimates place up to five million people at risk from river and coastal flooding, with 10,000km2 of land at risk from river flooding. Rising sea levels will only worsen the scenario, as many of the world's major cities lie in coastal areas. London is an interesting example, and although it is considered quite well defended at the moment, the future situation may be rather different.
Models have been formulated and estimates for future increased dimensions of the Thames assume there is a risk of inundation of parts of Hammersmith and the Victoria Embankment, Greenwich Peninsula and large areas of the lower Thames valley, in all some 125 km2. Other cities at risk from rising sea levels include Cardiff, Swansea, Bristol, Grimsby and Hull. At the moment there are no models that will definitively predict future effects from rising sea levels, but it has been estimated that average global sea levels could rise by between 10cm and 20cm by the year 2100.
For the UK, the Intergovernmental Panel for Climate Change has predicted a 30cm rise over the next 50 years. As a result, defence heights will have to be raised to the tune of up to 6mm per year. The effects will not be constant and some areas of the UK will see more dramatic effects than others. In Scotland, for instance, sea levels have risen at Aberdeen by 70mm since 1900 and many parts of the Scottish coast are now at risk from coastal flooding, particularly below the 5m contour. In Scotland alone 93,000 properties are at risk from coastal flooding, and 77,000 from inland flooding.
Some countries are now considering an alternative approach to flood management. Initiatives such as the Rhine Action Plan adopt, in principle, alternative strategies that include widening existing flood plains in conjunction with conveyance methodologies.
Factors affecting the UK include:
• Climate change and increasingly unpredictable rainfall patterns.
• Extensive coastline to all sides.
• Increased run-off from land due to agricultural practices and increasing urbanisation.
• Long tidal rivers (Humber/Trent, Severn etc).
• Downward land tilt to some areas.
• Inadequate or poorly maintained existing drainage.
• Under-investment in flood protection schemes.
• Large catchment areas into some rivers.
• Difficulty in analysing the probability of severe weather events.
Accepting that there is little that can be done to reduce the future risk of flooding (although the Rio Earth Summit and Kyoto Protocol were intended to mitigate the worsening climate situation by reductions in greenhouse gas emissions), then a sustainable flood management strategy can at least reduce its effects. Such a strategy may include a combination of factors: reducing building on flood plains; installing additional flood defence measures; reforesting upland areas; and allowing wider flood plains.
Implementing effective, integrated flood defence schemes requires a considered approach, taking into account the long term effectiveness of planned measures with regard to capital cost, disruption to amenities and townscapes, downstream effects and maintenance. A factor that is bound to compromise future flood management is the inexorable rise in demand for housing. Much of this demand is centred on land that lies in flood plains near to existing major centres of population and industry. Unfortunately, some of the 'brown' (industrial) land that the Government is insisting be re-used is also within flood plains, immediately creating a conflict in those areas. It is however still early days and new planning guidance (designated PPG25 and currently being developed) may help balance the needs of developers and land users.
An effective flood protection scheme must consider factors including: the morphology of local rivers; the likely effectiveness of engineered flood defences; the downstream effects from the implementation of engineered defences; and the socio-economic benefits that would be derived from such defences.
At present, significant amounts of public money are spent each year in simply maintaining existing defences, but many of these defences are either nearing the end of their effective lives or will be inadequate to cater for more severe floods in future. The UK's National Audit Office estimates that up to 40% of existing hard-engineered defences are in fair, poor or very poor condition (those classified as 'very poor' may be considered as derelict or failed, these representing 165 km of defences).
Although the government has already pledged additional capital resources for flood defence, under the Comprehensive Spending Review, additional investment is needed to stop long-term decline of the defences.
For coastal areas, a separate strategy may be appropriate, and shoreline management plans assess the balance of factors in terms of producing 'sustainable policies for the coastal defence of our shorelines taking into account natural coastal processes and issues relating to the environment and human needs'. Coastal effects can be difficult to model, because alluvial morphodynamics have no real 'equilibrium' state - what is put into effect today may be less effective in future years as natural processes take effect. There is no better illustration of this than the east coast, where loss of land is a continuous process and flooding a regular event.
Intervention can have unpredictable results. Providing hard-engineered defences to one part of the coast can lead to the denuding of sediment at adjacent coastal areas, requiring additional defences. Current government thinking tends towards a less-interventionist approach, allowing natural processes to take their course ('managed realignment'), although this is not likely to impress the population in these vulnerable areas unless the government introduces appropriate compensation.
There is understandably a good deal of consternation in the public domain that can only be allayed by appropriate action from government, but it is encouraging to observe that the issues are now seen as requiring a committed long-term strategy. Whatever the effects of weather and climate turn out to be in the next hundred years, the formulating and instigation of effective flood defence schemes will require major political will, planning and investment, based on an holistic approach.