Earlier this year China admitted that its massive Three Gorges dam project was being dogged by widespread corruption. Local government officials, it was revealed, were taking bribes and pocketing re-settlement funds intended for the relocation of over one million people currently living in the 630km2 of the Yangtze valley which will be flooded by the new reservoir.

So now the cost of the US$9.8B project looks certain to rise. Moreover, quality problems on the dam construction itself — such as under-strength concrete — could knock the project off programme and delay its completion, currently scheduled for 2009.

The problems afflicting the Three Gorges project may not be typical of every dam construction scheme, but delays and cost over-runs are by no means uncommon on major civil engineering projects — especially dams, and especially those in developing countries.

Hydro power projects are in fact notoriously prone to such problems, although the reasons for this are often understandable, even forgivable. After all, big dams take years to build, and the client is often a government. All it takes is a general election, a coup d’etat or just a change in policy, and the project can be thrown into disarray. But the problems are just as likely to be technical or logistical, such as unforeseen ground conditions or bad weather.

Meanwhile, other types of power project are planned, built and commissioned with relative ease. Fixed price contracts and tight schedules allow them to be brought into commercial operation at an early stage. Major civils projects like bridges, roads or water treatment works, are built on the build-operate-transfer (BOT) model and similar schemes, where the contractor makes its money operating the project for a period after completion, before finally handing the project over to the client. So why can’t hydro power projects be built the same way?

‘The main reason is risk,’ says Stig Holmqvist, vice president of Swedish construction giant Skanska Civil Engineering International. ‘There are usually big risks involved but seldom any mechanism for sharing the risk between the parties concerned.’

Holmqvist should know. He has worked on dam construction projects for nearly all of the past 22 years, starting with Tunisia’s Barrage Bourgiba in 1978-81, followed by Sri Lanka’s Kotmale dam in 1982-86. His longest single scheme was India’s Uri project, which kept him occupied from 1989-97. He says that with BOT schemes and other novel procurement methods like guaranteed fixed price, nearly all the risk is borne by the contractor.

‘These kinds of project have lots of uncertainties,’ says Holmqvist. ‘Firstly, if you are to build a big reservoir, you have the problem of land acquisition to solve — and it’s a problem that gets worse every year.’

As the Chinese government has discovered on the Three Gorges project, negotiating the relocation of thousands (or in the Three Gorges’ case, millions) of people is a task fraught with logistical, social and political problems. ‘Even on the Uri project, which is a river scheme and has no reservoir, we couldn’t start on time because of difficulties with the (relatively small) land acquisition,’ Homqvist says. And he points out that this is a problem unique to dam projects: ‘With thermal power plants, the land required is much less.’

Another problem facing every hydro power project is that you are building in a river. ‘There is always the risk of floods,’ he added, ‘and if you’re unlucky you can be hit by the big flood that only happens once every thousand years.’

Similarly, dam construction involves a great deal of underground work. You have to build tunnels and underground caverns and it is almost impossible exactly to know what you are going to encounter. According to Holmqvist there are always surprises.

So building a big dam means weighing up the risks and deciding how much you can afford to leave to chance. Temporary diversion works can reduce the risk of floods, but only up to a point, beyond which they are no longer economical. Likewise, thorough geotechnical surveys will improve your knowledge of the ground conditions, but they take time. You might finish the job quicker by starting early and tackling any unforeseen conditions as they appear.

It is a matter of some pride to Holmqvist that all his projects have been completed on time and within budget (even if that budget has often been revised by the client during the construction process). But he readily agrees that few dam construction jobs can share that boast, and that being late and over budget is common:

‘In India, it’s a rule — these projects are always late. On the Uri project, Skanska was actually brought in by the Indian government to show the industry that there was an alternative to their traditional method,’ Holmqvist commented.

The traditional Indian method is to carve a major project up into lots of small contracts. Skanska’s preferred method is to offer a turnkey solution. Holmqvist says it is much simpler if one contractor is in charge of everything. If a project comprises a myriad of small contracts all running in parallel, then not only is co-ordination a nightmare, but also a problem which delays one contractor, can have a domino effect and, before you know it, everybody’s schedule is being revised and chaos reigns.

Turnkey contracts put the contractor in control of a project and therefore in a better position to respond quickly to nasty surprises. Skanska, like most international construction firms, is wary about tackling big hydro power projects on a BOT basis because of the exposure to risk. For instance, the vital question of land acquisition (plus various other legal preliminaries) is always the responsibility of the client and hence beyond the control of the contractor.

For Andy Hughes, director and water business leader of consulting engineer RKL (a subsidiary of Ove Arup), successful dam construction is largely a matter of good management. ‘I’d like to think we can usually do it with some degree of success — it’s all about good planning.’ If your risk calculations are accurate, then progress should be fairly smooth. ‘It’s only the political aspect, which you have no control over, that can really disrupt these projects,’ he says.

Someone with recent first-hand experience of this is Chris Cotterell, deputy site agent on the massive Katse dam project in Lesotho. He was one of a team of British engineers who fled the site last September when South African forces were sent in at the request of Lesotho prime minister Pakialitha Mosisili to attack rebel soldiers garrisoned at a military base 800m from the dam site.

‘The troops were sent in supposedly to protect the dam,’ Cotterell explained. ‘There was a perceived risk of the rebels sabotaging it.’ Cotterell and his colleagues were given just five minutes’ warning before the South African helicopter gunships launched their attack, killing 22 rebel soldiers.

Apart from the obvious danger to himself, Cotterell was keenly aware of the risk to the project. Luckily for the Lesotho Highlands Development Authority (and the South African government, which payrolled the project) the dam and its appurtenant works were largely completed, and it was in the commissioning phase. ‘It could have been a lot more disruptive if it had happened a year beforehand,’ Cotterell says. And if the rebels had sabotaged the dam, who knows what the cost would have been.

This sort of problem can seldom be planned for. Neither can materials price fluctuations over a 7-8 year project duration, as Stig Holmqvist of Skanska points out. However, Holmqvist has three pieces of advice for dam-builders:

•Firstly, make sure the required clearances and licences are in place before you start. That is the owner’s responsibility, not the contractor’s.

•Secondly, consider the turnkey model. Interface problems are eased, planning and programme can be more co-ordinated and hence there is a smoother passage through the work sequence. It is better that the contractor does that.

•Lastly, make sure the schedule and budget is reasonable. You should also look at the total cost of the project, and at the effects of early commissioning.

All of this is undoubtedly sound advice for anybody planning to build a big dam – although it comes a little too late for the Three Gorges project.