Lauded at its opening as the perfect marriage of flood defence and ecological regeneration, the Jubilee River was set to be the way all flood defence schemes were to be built.
Client the Environment Agency, main designer Lewin, Fryer & Partners and main contractors Nuttall and Balfour Beatty have all collected awards for their efforts.
But two and a half years on, the picture is far less rosy, with a damning report revealing a raft of design and construction faults that have left the scheme unable to fully perform its function.
The Jubilee River is a 11.6km man-made watercourse that looks and acts like a natural river. Flood flows are diverted from the Thames at Taplow Weir and routed along the channel to rejoin the Thames at Datchet.
Downstream from Taplow four other structures act to maintain minimum water levels in the channel for ecological reasons and soft engineering was used throughout to achieve the natural effect.
But it took a flood flow in January 2003 to leave the softengineered embankments dangerously close to collapse and in need of immediate repair.
Last summer $2.6M was spent on patching up the channel, this summer a further $3.5M will be spent and next summer more still will be needed.
This week consultant Atkins has exposed a horrifying series of failures.
'We were stunned, ' says Ian Thompson, technical advisor for Community Support Group (South), a residents group set up in the wake of the 2003 floods. 'It is far, far more serious than we ever knew.
'We're talking about a river that is structurally unsound from top to bottom. We have a $196M public scheme that is no good in the event of a flood. The whole report is filled with design, construction and specification errors, ' he said.
The report does indeed make grim reading for the Environment Agency officials responsible for the Jubilee River. Aside from the simple fact that the channel is not big enough (see news story), almost every structure is flawed.
At Taplow Weir it expresses amazement that 'a stilling basin, an important feature in the design of a hydraulic structure, is missing'.
Stilling basins are usually constructed in areas downstream of radial gates so that excessive energy from high water flows is dissipated.
Without this basin the damage caused when the gates were opened in January 2003 was significant, potentially undermining the weir's foundations. The problem could also have led to the eventual undermining of Mill Lane Bridge immediately downstream.
The report criticises the non standard layout employed for the works downstream of Taplow Weir. It argues that standard designs make it possible to reasonably predict flow conditions. Without these, 'it is likely that the design increased the potential for damage, ' says the report.
Other structures face similar criticisms. At Manor Farm Weir, Atkins reports that its convex shape simply acted to direct aggressive water flows into the earth embankments, immediately undermining them.
At Slough Road Weir anchors holding concrete block in place the earth embankment failed.
Load testing by Atkins revealed that all anchors failed to meet their design load.
Of the embankments, only 35% were found to have a high resistance to erosion. Up to 700m of embankment is 'A1' critical and in need of immediate replacement.
The Environment Agency stresses that despite its flaws the channel is effective. But it is considering ways of reclaiming cash from its contractors and consultants.
This will not be easy, says Environment Agency regional director Chris Birks. 'We have had around 30 consultants and contractors involved in this project so the interaction is quite complex. But we are looking for liability.'
Despite this, Birks has no doubt that he would build the river again:
'Jubilee River is different to anything we've tried before and there is some learning for us.
'But if you look at the alternatives there is no question that it is the best type of option - you can't countenance going back to straight sided trapezoidal channels.' MH