Bruton flood retention dam.

Colin Clark diagram

Relationship between the percentage of woodland in a drainage basin and ‘bankfull discharge’ downstream. Graph from Dr Colin Clark, ‘Floods at Bruton’.


Information leaflet, August 2019

Serious flooding on the Somerset Levels in 2013-14 attracted the whole country’s attention.  The upper Brue valley, however – frequently flooded in the past – was completely free from  floods that winter. This can be explained by the flood retention dam that was built above Bruton in the 1980s, extended and upgraded in 2008-13.

This expensive piece of engineering addresses the issue of serious flooding in Bruton, which  occurred at irregular intervals from at least the eighteenth century up until 1982. The new dam prevented a further flood in 2005. As local hydrologist Dr Colin Clark has said, “The  residents of Bruton have been left well protected up to the design standard, but also left in  grave peril if that standard were to be significantly exceeded.” If it were, the catastrophic  result of the dam being over-topped would be rapid erosion of the structure and a very  sudden rise of water level in the town itself, probably by several feet per minute; and the way the climate is changing means that ever more extreme rainstorms are becoming increasingly likely.

The source of the River Brue is on the Mendip Hills above Bruton, with a relatively small  catchment  area (31 sq. km.) formed like a steep-sided bowl, down which several streams  flow and coalesce as they approach the town. The area is well known for torrential rainstorms – the greatest was in 1917 with 9.56 inches, then the highest that had been recorded in a  24-hour period anywhere in the British isles.

However, floods are not caused simply by rainfall. Selwood Forest used to extend to the edge of Bruton, with trees completely covering the slopes above the town. The tree cover has been steadily reduced since the seventeenth century when the forest was ‘disafforested’ by Charles I and sold as farmland. 90% of the trees have now disappeared. A tree canopy holds back one third of total rainfall from reaching the ground, and – with the ground broken up by tree roots – woodland soils are about seven times as permeable to rainfall as are the soils of pastureland.

River flow is therefore greater downstream from farmland than it is below forested land. The flow is measured as ‘bankfull discharge’, which represents the maximum flow that a river can hold before it bursts its banks. Dr Clark has carried out research showing that the effect on bankfull discharge downstream, between 0% tree cover and 100%, is at least a ten-fold reduction (see graph). This can be critical when the river meets a pinch point, such as in a town.

The need to recognise how management of upland areas can impact upon the lower floodplain – for example by planting trees to slow the rate of run-off into the floodplain – has been put forward as a response to climate change. This is to be welcomed, but it is an aspect of river management that needs more attention anyway. In 2013-14 there were loud calls for dredging – but this is expensive, disruptive to wildlife, and often inadequate. A more creative approach, and one that works with nature rather than in spite of it, is needed as soon as possible.