Your article on New Orleans' flood defences (NCE 2 March) gives a glowing account of research sponsored by the US Army Corps of Engineers but it is not alone.
Following the UK's 2000 floods the Environment Agency, Department for Agriculture Food and Rural Affairs and the EPSRC have supplied major funds for research into the performance of flood defence assets, with Strathclyde University acting as a centre of expertise for geotechnical engineering.
Recent reviews of flood embankment failures in the UK and Europe have shown that the causes of failure are often due to poor detailing of construction joints between soft and hard defences or a lack of attention to the underlying geology which leads to slippage, uplift or piping (Dyer 2004 ICE Proc Water Management 157 WM4).
The theory for these failure mechanisms is generally well understood but often poorly implemented because geotechnics is not fully integrated into the design process. A recent report by the University of Berkeley indicates that the poor performance of levees in New Orleans was caused by a similar catalogue of failures.
In contrast to the centrifuge modelling described in the article, the main thrust of fundamental geotechnical research in the UK focuses on the long term deterioration of initially competent embankments - caused for example by desiccation fissuring that was recognised as one of the main causes of flood embankment failures in the 1953 North Sea Floods.
Professor Mark Dyer, University of Strathclyde mark. dyer@strath. ac. uk