Hydroelectricity projects have the potential to transform the economies of less developed nations but many schemes currently being proposed are in seismically active areas
Energy use is growing around the world and the trend is unlikely to be reversed in the medium term with current demand expected to increase by at least 50% by 2030 as the economies in developing nations, such as China and India, grow. Clearly, being self-sufficient and having surplus energy for export is a factor in the rate of economic development of these nations and this is one of the drivers behind an increase in hydroelectric schemes.
Most of the projects currently being discussed fall into the 400MW bracket but there is one scheme in the Democratic Republic of the Congo – the Grand Inga Project – which plans to harness the power of the River Congo to produce 40,000MW of power.
The problem is that despite the huge potential these schemes hold to change the economic fortunes of the countries promoting them, many are located in seismically active zones so the development of these projects is not straightforward.
However, according to Mott MacDonald senior principal engineer Barnali Ghosh, although the locations would have been considered less than ideal a few decades ago, advances in seismology mean what was once too risky is now feasible.
“Traditionally, hydropower developments have avoided seismically active zones but all the potential sites in low seismicity areas have now been developed and, to meet demand for energy, sites in more challenging areas are now being considered,” says Ghosh.
The environmental credentials of using water power to create electricity is one of the main drivers in developments of this type, but it is the advances in seismic modelling through finite element analysis and a performance-based design approach that is allowing these schemes to become a reality.
Schemes such as the Inga project have huge potential to meet the growing demand for energy and to change the economic outlook of countries that develop them. “The Inga scheme alone could meet 70% of the energy needs for the whole of Africa,” says Ghosh.
“The hydropower schemes that are currently being discussed are just the tip of the iceberg of what could be developed so the number of schemes in seismically active areas is likely to increase too.
“Almost every day I am dealing with a new enquiry to look at the seismicity issues for a hydropower scheme – the demand is really growing.”
Two things have changed that are allowing this type of development to proceed despite the seismic risks, explains Ghosh. “We have a better understanding of the seismic forces and the type of movements that are induced, plus the probabilistic seismic hazard assessment allows us to quantify the risks.”
The ability to monitor sites and the availability of information on seismicity has also increased and the focus is very much on assessing the risks of seismicity and the impact of that seismic activity on the structure.
While site selection and monitoring is key, the site investigation is critical to the success of these major schemes in earthquake-prone areas.
Mott MacDonald is currently working on the site investigation and detailed design of a scheme in Georgia where seismicity is of concern and Ghosh says that it is typical of the type of projects that are now being considered.
“When you have a dam development in a seismic zone, you have to consider how the ground will behave during an earthquake and also how the dam structure and ground will interact,” says Ghosh.
“Most countries have design codes developed to cope with the level of seismicity in the region for buildings, but these do not cover dam design. For dams we not only need to assess the region’s seismicity but also the height, size and type of dam being planned.”
According to Ghosh, there is a general rule that the greater the mass of the dam, the greater capacity it will have to withstand a seismic event. However, the larger the dam, the greater the cost, so most clients want there to be a balance when it comes to design.
“For the Georgian project we are using the ICALD 2010 revision guidance which looks at the consequences of failure and promotes a performance-based design using the predicted earthquake design period,” says Ghosh.
“There are two types of event considered. The first is the Operating Basis Earthquake (OBE) which is a high frequency but low impact event, and the other is the Safety Evaluation Earthquake (SEE) which would not cause the dam to collapse but may result in the need for structural repairs.
“To carry out these assessments we have to review the earthquake history for a 300km area around the site.”
Source: Finlay Booth
At the site in Georgia the OBE peak acceleration is 0.01g, but to be sure that the figure is accurate the Mott MacDonald team needed to assess if there were any active faults in the area that could rupture and cause the OBE value to be higher.
“This is a real challenge with no surface fault outcrops so we worked with the Georgian Geophysical Society to set up a micro seismic network with six monitoring stations,” says Ghosh. “Local relationships on projects like this are very important. In Georgia, the Georgian Geophysical Society carried out the monitoring work and have now taken ownership of the network.”
The aim of the network was to identify if there was a larger magnitude event locally over a six month period, which would enable the team to identify active fault movement. “No faults have been identified,” says Ghosh.
“The SEE level at the site is 0.3g based on a 6.5 magnitude event.”
The dam will be a 30m concrete structure but the exact design will depend on the final site investigation work. “The ground conditions are mostly gravels with soft clays, which present potential for liquefaction issues,” says Ghosh.
The other issue facing projects such as the Georgian hydropower scheme is the scale, technical detail and complex insitu testing needed during the site investigation. “Local contractors with the skills are difficult to find so often we have a large supervision role to ensure the necessary information is collected,” says Ghosh. “The geophysics for the Georgian project was undertaken by an international company because there were no local contractors with the right capabilities.”
Assessment of the seismically induced ground movements at the site in Georgia were also complicated by the shape of the valley. “Numerical analysis is essential in situations like this to ensure we fully understand the site response to seismic movement,” says Ghosh. “Without numerical modelling a performance-based approach would be almost impossible to deliver.”
The seismic risks to the dam and infrastructure are not completely resolved through careful design of the structure itself though. Once the dam is built and the hydropower plant is generating electricity there is still a risk from seismicity even if the dam structure has been built to deal with OBE and SEE events. The risk comes in the form of reservoir triggered seismicity. “The impact of this phenomenon was seen at the Koyna dam in India in the 1960s,” says Ghosh. “At Koyna the seismicity of the area was seen to increase after the reservoir level was raised and probably results from the increased loading of the water on the local bedrock.
“Analysis of the stress history of the area is needed to understand if there is a risk of this at a new hydropower site but it is still difficult to definitively assess.”
It is clear that Ghosh believes that meeting the increase in demand for hydroelectric schemes in earthquake-prone areas of the world is possible, what she is concerned about is the rigorousness of testing and analysis at the feasibility stages to mitigate the risks. She points to the failure of California’s San Fernando dam in 1971 as an example of the risks. “The 1971 San Fernando dam was badly damaged by a 6.6 magnitude earthquake and the collapse had a significant impact downstream,” she says. “There were no seismic guidelines for dams at the time of its construction but the replacement structure was built to a design that considered the seismic setting and used ground improvement to mitigate against liquefaction and that structure survived the 6.7 magnitude Northridge earthquake in 1994 with no damage.”