Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more



Six months after New Orleans' flood protection system failed in the wake of Hurricane Katrina, a giant centrifuge is speeding the development of safer levees. John McKenna reports.

When Hurricane Katrina hit New Orleans on 29 August last year, US president George Bush initially claimed the city had 'dodged a bullet'. The storm's winds dropped from 282km/h, measured at sea, to 201km/h by the time it made landfall, and it appeared that, though it would sustain damage, the Big Easy had been spared the worst that nature could unleash.

But Bush's optimism proved unfounded as huge storm surges rushed into the city's canals from Lake Pontchartrain to the north and the Gulf of Mexico to the south.

Large sections of the floodwalls were overtopped and there were three major breaches, plus countless smaller instances where the levees buckled under the strain of the water. The levee breaches inundated large urban areas with up to 6.1m of water, destroying homes and businesses and leading to the death of 972 people.

In October, after the initial rescue efforts and plugging of the breaches by the US Army Corps of Engineers, US defence secretary Donald Rumsfeld ordered a comprehensive investigation into why the levees failed.

Initial analysis of the breaches suggested most were due to erosion.

Principal levees consisted of earth embankments crowned by precast concrete panel walls. As water sluiced over the walls, the leeward embankment slope was scoured away, exposing the embedded portion of the wall elements and removing their support.

These were then either undermined by water, which burst through beneath them, or were toppled by the pressing weight of water.

But it is important for engineers to understand the failure mechanism in detail if defences are to be improved.

Rumsfeld's investigation is being carried out by the Interagency Performance Evaluation Task Force (IPET), formed from more than 40 organisations and led by the Corps of Engineers. Using scientic and engineering methods, 10 teams are researching and recreating different aspects of the storm and local environment.

Scott Steedman, a British geotechnical engineer, is leading one of these teams. The only nonAmerican appointed to head an IPET research team, he was tasked with the physical modelling of the levee and floodwall performance.

Over the last 15 years Steedman has worked regularly with the Corps, helping it develop one of the world's most powerful geotechnical research centrifuges at the Engineer Research & Development Center (ERDC) in Vicksburg, Mississippi.

Steedman has recruited colleagues from the UK and from Dutch consultant Geodelft. 'We know from experience in the Netherlands that dyke failures keep happening by sliding mechanisms and by seepage, ' Steedman says.

'In [the Netherlands] there are various issues to do with constructing levees or dykes on soft clay or peat foundations. In New Orleans you have a mixture of foundations.

Some are soft clay and peat and there are also large sand deposits in certain parts of the city.'

New Orleans levees have generally been constructed without toeing into the underlying ground or with deeper cut-off walls.

Steedman and his team have set about constructing one-ftieth scale models of the levee sections that failed, for testing in the ERDC's centrifuge. Accelerating the centrifuge to create a gravitational force of 50G increases effective weight, forces and stresses in the model, making it behave like its full scale, real life counterpart.

'The point of the physical model is to have something independent of the numerical computer model [being generated by other parts of IPET], ' Steedman says. 'Otherwise people may say 'that's just a computer model. It looks pretty but how do I know it's telling me the truth-'

'The advantage of the physical centrifuge model is that we can have a parallel view and hopefully the two models will conform, and confirm the most likely mechanism of failure - explain the facts as we have them.'

Teams are sharing information.

'All 10 [projects/teams] depend on a huge amount of information which is interrelated, ' he says.

The leaders have regular progress meetings. Evidence gathered from witnesses, further soil samples and findings by the wave modelling project all contribute to Steedman's team constantly refining the physical models.

Most of the research has been completed and the ERDC centrifuge is now testing the physical models.

When this is finished, findings will be collated into IPET's report and presented to the US government on 1 June - just in time for the start of this year's hurricane season.

June is also the deadline for the Corps to complete repairs to New Orleans' flood protection system (GE March 06). Repairs are being carried out using original levee designs but as the IPET project advances, information that will improve the design is being fed through.

'We are trying to incorporate the IPET recommendations and so far it has led to some changes in our design parameters, ' says Brett Herr, project manager of Task Force Guardian - the name given to the race to restore the levees before the start of June.

'We are building to the proper design heights. In some cases the levees were below those heights when Katrina struck, ' he says.

'When you build a levee here it settles and a limitation in funding led to many of the levees being below the design grade height.'

Herr adds that this is not the normal season for levee construction.

Conditions on site are wet, making excavation and placement of clay a heavy, mucky business.

The Corps has also received funding to install fl ood control gates on the canals which drain water from the city into Lake Pontchartrain.

When the levees failed, water rushed from the lake and along the canals into the city. In future the gates will be closed when hurricanes and accompanying tidal surges threaten.

Water will be discharged from the canals by 12, 1.5m diameter pumps at each outfall.

Task Force Guardian's work is budgeted at £440M, split across 59 contracts, of which 55 have been awarded so far, but only 10 are finished. With just three months to go, the work is only 30%-40% complete.

'There are probably at least 10 contracts we are concerned about getting done by June, ' admits Herr.

'These are the bigger ones with the most work to do.'

He says two of the projects behind schedule are big levee repair jobs, as well as a couple of the closure gates.

He blames long design lead times for delaying construction.

Ultimately, as Steedman points out, the reconstruction effort may have been in vain. 'What happens if the IPET team comes up with some piece of evidence which means that the work they [Task Force Guardian] have done in the last six months is redundant?

'The answer is they will bite the bullet and redo it. If it comes to it they will have to immediately carry out additional works or replace what they have already done.'

Herr concedes that rebuilding New Orleans' levees twice is a real possibility. 'IPET is providing us with its findings and recommendations and hopefully they will be incorporated before June so we don't have that problem, ' he says. 'But if we do, we will have to go back and readdress the designs.'

Centrifuge testing a levee

Steedman's team is using clay, peat and sand taken from the sites of the levee failures; the models are built to original designs.

Water levels are taken from the high water marks on floodwalls and various possible modes of failure then tested.

The main advantage of this physical modelling is that it may reveal mechanisms the team had not anticipated.

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Please note comments made online may also be published in the print edition of New Civil Engineer. Links may be included in your comments but HTML is not permitted.