Overloaded

AT THE International Conference on Freshwater, held in December 2001, dams made headline news again. Klaus Toepfer, executive director of the United Nations Environment Programme (UNEP), addressed the world's press: 'The issue of dams can arouse strong passions on both sides,' he said. 'However what we are talking about here is the state and fate of the existing stock of dams and reservoirs, on whose waters billions of people depend for not only irrigation and drinking, but also industry and the production of hydroelectricity.'

Those who have spent much of their working lives studying the problem of reservoir sedimentation welcomed such an intrusive media glare.

'The loss of capacity of the world's dams should be of the highest concern for governments across the globe,' Rodney White, author of Evacuation of sediments from reservoirs, said. 'And at the moment I do not believe this issue is commanding the attention it deserves. The demand for water is rising and not falling. I am extremely concerned that water shortages in some of the poorer parts of the world will intensify unless we act to reduce reservoir sedimentation, and conserve storage in existing dams using sound management techniques.' Furthermore, White believes sediment removal should be a fundamental feature in the design of dams and their associated structures.

The much-debated work of the World Commission on Dams (WCD) also highlighted the problem of sedimentation. It stated that sedimentation and the consequent long term loss of storage is a 'serious concern globally'. The effects will be particularly felt by basins with high geological or human-induced erosion rates, dams in the lower reaches of rivers and dams with smaller storage volumes. Furthermore, WCD included sediment management under one of its guidelines for good practice.

Alessandro Palmieri, a dams specialist with the World Bank, believes that sedimentation is a hidden problem. 'It's hidden by water,' he says. 'When you take decision-makers to see a reservoir they just see the water and not the problem underneath. Consequently sedimentation has been a neglected and underestimated problem.' Although Palmieri admits it is easy to be critical with hindsight, he says that decision-makers have only realised the problem when it's too late. 'We now have to increase awareness of the situation we are facing.'

Facts and figures

As White explains, in many areas of the world the life span of reservoirs is determined by the rate of sedimentation which gradually reduces storage capacity. Eventually this process destroys the ability of the scheme to deliver the benefits for which it was built. The problem is that many reservoirs are fast approaching this stage in their life.

'Although there are reservoirs where sedimentation has proven to be a problem within ten years after construction,' Professor A Rooseboom, Chair of icold's Committee on Sedimentation of Reservoirs, says, 'sedimentation typically only becomes a significant problem 50 or more years after construction of a dam. With only about 7% of existing storage capacity in reservoirs worldwide having been created more than 50 years ago, it is expected that sedimentation will become a serious problem in many reservoirs during the next 50 years.'

Rooseboom also adds that due to the rapid growth in dam development during the 1960s and 1970s, about 40% of reservoir storage capacity will be seriously affected by sedimentation in only 20 years' time. Ultimately the age at which problems become serious depends primarily on the ratio between the annual incoming sediment load and the initial storage capacity of the reservoir.

Estimates suggest that sedimentation rates are now eight times higher than they were in the mid 1960s. Although annual losses vary from region to region, China is currently losing 2% of water storage capacity annually, followed by the Middle East which is losing 1.5% and Central Asia 1%.

This means that out of the 40,000 large reservoirs worldwide, between a half and one percent of the total storage volume is lost annually as a result of sedimentation. To simply maintain current storage levels 300-400 new dams, at a cost of approximately US$7.5M per dam, would need to be constructed each year.

'The problem is set to intensify,' White adds, 'as increasing populations and consumption per capita means that the demand for storage is rising inexorably, despite the harnessing of alternative sources and the efficient use of water.'

Many countries are now beginning to pay the price for development policies which encouraged dam designs with 50-year economic life spans. A lack of provision for the longer term makes it more difficult to replace reservoirs that have lost significant proportions of their storage capacities.

'It was a simplistic approach to storage,' Palmieri said. 'Dams are life long assets and should have been viewed as such.'

The extent of sedimentation and storage loss worldwide may well have been previously hidden in certain areas as new dams were specifically constructed to replace lost storage capacity. 'The importance of sediment management was never fully appreciated before,' says Palmieri. 'The dead storage approach was an easier alternative.'

Today however, such an approach is no longer feasible. 'There are fewer and fewer good dam sites available,' White said, 'and new dams can have serious environmental and social consequences.

'The objective of making reservoirs more sustainable using sediment management techniques is clearly laudable. However, many techniques are not applicable to all reservoirs and some dams will inevitably be either raised to regain storage or decommissioned.'

A closer look

Sediment deposition occurs as the river enters the reservoir and its sediment transport capacity decreases in the backwater created by the dam. Coarse sediment is typically deposited first while finer clay and silt fractions are transported much deeper into the reservoir.

Apart from the obvious fact that sediment build up within the reservoir leads to decreasing storage capacity, in specific areas it can lead to local problems. Sediment build up in a storage reservoir without facilities for preventing sedimentation typically only approaches equilibrium when the remaining storage capacity is a few percent of the original storage capacity. At that stage the volume of the sediments which have been deposited above the full supply level, upstream of the reservoir, can be an additional 10% or more of the deposit below the full supply level. The delta which is formed above the reservoir can have a serious impact on flood levels upstream, particularly when dense vegetation becomes established which further stimulates sediment deposition.

Reservoir sediment build up can have serious impacts on diversion and extraction facilities, including turbines, penstock inlets and pump station inlets, thus inhibiting the functions the facility was designed to fulfil. In addition there are safety issues.

'The most important effect of sedimentation is decreasing storage,' says Palmieri. 'This leads to many reservoirs simply being abandoned - something that is not just happening in developing countries. It no longer is purely an economic problem but becomes a safety issue as well.'

Agriculture, deforestation of natural forests, reforestation, overgrazing and other activities which lead to widespread detachment of sediment particles, which are eventually carried away by stormwater, are the main contributors to catchment erosion in many parts of the world. The important role that winds play in the transportation processes, where smaller sediment particles are being transported, has been underestimated in the past, says Rooseboom.

Speaking at the International Conference on Freshwater, Klaus Toepfer wants to help people realise that land use and sedimentation are linked. 'So whether you're using the land around a reservoir for farming or tourism, you'll have consequences,' he said.

Studies have shown that the felling and clearing of trees is aggravating the situation. The levels of erosion from hillsides planted with crops are 150 times higher than from the same land covered with trees. But it isn't just deforestation that's the problem. Palmieri also spoke about the misconception that reforestation of a catchment area can eliminate sedimentation.

'This is not true in all cases,' he said. 'Erosion can be linked to large catchment areas of more than 500km2. But reforestation on a catchment area with accumulated sediment will not solve the problem. It only takes a large hydrological event to occur and then the sediment will be remobilised and transported back into the body of water.'

Jeremy Bird from UNEP's Dams and Development Unit, which is continuing with the work of the WCD, agrees. 'Reforestation may be the answer but you have to careful,' he warns, 'or you can get problems. South Africa has found that if you simply plant pines and Eucalyptus trees on bare slopes they reduce the inflow to the reservoir by around 7%. So you need to think about whether to plant alien or indigenous species.'

Loss of storage

The rate of loss of storage for reservoirs is dependent on the sediment yield from the catchment which, in turn, is dependent on the rate of erosion and the transport, by water, of the sediment within the catchment.

Sediment rates vary greatly with the highest yields per square kilometre of catchment being found in geologically active regions where earthquakes occur. Sedimentation impacts on reservoirs are generally most severe in semi-arid regions where sediment yields are relatively high and dam catchments large, as is the case over large parts of Africa. On the other hand, there are reservoirs in England which have been in use for more than 100 years without noticeable sedimentation, even though intensive farming is practised within their catchments.

Sedimentation is caused by a multitude of factors. It's a natural phenomenon. Erosion cannot be stopped completely in its tracks but measures can be taken to mitigate the effects.

Rooseboom believes that the ideal solution is to minimise the accumulation of sediment within reservoirs. In practice this means that only relatively small diversion dams should be constructed on rivers that carry substantial sediment loads (small in terms of storage capacity to mean annual flow ratio of <0.03). These dams must be provided with low level gates to sluice incoming sediments during floods and to flush limited deposits out when necessary. While storage dams should be built in valleys with small catchments and should be sized with additional dead storage for future sedimentation of 50-100 years.

An alternative solution, which Rooseboom says has been implemented successfully, is to bypass incoming sediment-laden floodwaters around reservoirs via tunnels or canals. This solution however can only be applied economically under special topographical conditions.

Land use practices, agricultural methods and engineering measures to control erosion can all help minimise sediment yields. 'Whereas it would be wonderful if large scale catchment management policies could serve to limit reservoir sedimentation and to practise soil conservation at the same time, success in this regard has been very limited,' Rooseboom said. 'The decreases in sediment loads which have been observed in various Southern African rivers are due to the depletion of erodable top soils rather than successes with soil conservation measures. It is very difficult to get governments to apply strong soil conservation measures as these are generally expensive and often unpopular.'

Palmieri agrees, adding that it is also difficult to get reservoir owners to take full responsibility for erosion and sedimentation when it is part of a much bigger problem involving farmers and other land users within the catchment area. However, he gives the example of a World Bank project on the Loess Plateau in China, which is within the catchment of the Yellow river. By working with the local community, the Bank has encouraged farmers to change cultivation practices by building terraces and check dams, coupled with larger dams for protection. Sediment deposits in check dams have then been used successfully for cultivation purposes.

Removal problems

Once sediments have become deposited and consolidated it becomes very difficult to remove them cost-effectively by mechanical means, according to Rooseboom. White says that dredging or the use of other mechanical means is feasible but usually requires reservoir levels to be maintained at low levels for extended periods of time. 'It is expensive,' he says.

'Disposal of dredged deposits generally is a major problem since the sediment has to be contained in diked disposal areas,' Rooseboom adds. 'In the past it has been cheaper to create new storage space, for example by raising a dam, than to recover lost storage capacity by dredging. Sediment removal by dredging should be seen as a last resort as the removal and disposal of existing deposits often create new social and environmental problems.'

Estimates suggest that dredging is up to five times more expensive than flushing - the process of re-entraining deposited sediments and passing the sediment-laden flow through low level outlets in the dam. This requires reducing water levels in the reservoir, consumes significant quantities of water but is capable, under certain circumstances, of removing even coarser sediments (mainly of sand sizes). It is, after all, the sands and gravels that determine the ultimate sustainable volume of any reservoir.

Flushing can be applied to existing dams as well as to new dams. However, White cautions that it is not applicable universally.

Useful life

A variety of methods can be used to extend the useful life of reservoirs but their economical, technical and environmental feasibility depend on a variety of factors. White states that these include the:

• Availability of suitable engineering facilities at the dam to control water levels and outflows.

• Availability of surplus water and its value if used for other purposes.

• Predictability of river flows.

• Characteristics of the sediments entering and within the reservoir.

• Availability of disposal sites for dredged sediments.

• Effects on the downstream reach of evacuating sediments through the dam.

• Effects of sediment management on normal operation of the scheme.

• Effects of sediment management on other reservoirs within the region.

Sedimentation and loss of reservoir storage are problems that cannot be solved over night. There will always be natural levels of erosion which will contribute to a loss of water storage capability. But as time passes, and more dams come of age, even greater importance needs to be placed on sediment management. Other natural phenomena, such as climate change, are likely to have an effect on the situation. The severity of storms and rains is predicted to increase, accelerating natural erosion rates in and around rivers that feed reservoirs. It is also likely to exaggerate extremes in rainfall patterns, making it even more vital that the storage capacity of reservoirs is maintained.

'We need to create new ways of thinking,' Palmieri says. 'We need to increase awareness of the problem, and the signs are that this is beginning to happen.' He gives the example of North Africa where reservoirs are ageing, and countries experience intense floods and droughts. The World Bank has been talking with some of these countries, advising them that it would be feasible to flush sediments through the dams. However, the initial reaction to the suggestion was a fear of wasting water.

'We need to reclassify sediment management as a water user,' Palmieri adds. 'We have to make people understand that by doing this you can ultimately regain storage levels within a reservoir.'

Rooseboom says that the industry has learnt a great deal over the past 30 years. Tools do exist to provide designs to overcome sedimentation problems, and numerous dams and weirs have been built that cope adequately with sediments.

For example, dams which have been built high up in catchments as part of inter-basin water transfer schemes prove to be very beneficial not only from a sedimentation point of view, but also in terms of evaporation losses. While the concept of off-channel storage being served by the smallest possible diversion structures on rivers makes complete sense in terms of the present day concept of sustainable development. He cites Nagle dam in South Africa as a good example where a bypass canal has limited sedimentation of the main reservoir since 1950. The bypass canal has a high discharge capacity of 2000m3/sec and shortcuts a natural horseshoe bend in the river. However, mistakes have also been made.

From a water resources point of view, Welbedacht dam on Caledon river in South Africa has been a disaster but has provided ample opportunity to study sediment and deposition patterns in reservoirs, with the purpose of developing design tools for future reservoirs.

Welbedacht dam was completed in 1973 with a storage capacity of 114M m3. Three years later it lost one-third of the storage capacity and 13 years later had lost 65%. By 2000 the storage capacity had decreased to 8.5% of the original value, in spite of attempts to flush deposited sediments through its seemingly large gates, which are unfortunately located too high above the riverbed.

Within two years after completion in the 1980s, Mbashe reservoir in South Africa had lost most of its 9M m3 storage capacity. Although the 30m high dam was equipped with a bottom outlet and excess water is available for sediment flushing, the outlet discharge capacity is too small to allow free outflow conditions during flushing.

'While sedimentation problems in major reservoirs such as Tarbela in Pakistan have received a great deal of attention,' Rooseboom said, 'the total impact on numerous small reservoirs is probably more serious as many people are affected in societies which do not have the means to overcome these problems.'

With this in mind the World Bank started a research programme in 1997 which is scheduled for completion in October 2002. Called Research on an Economic and Engineering Model on Reservoir Sedimentation Management (RESCON), it aims to develop a framework and economic model for assessing the feasibility of sediment management strategies that would allow the life of dams to be substantially prolonged. It aims to reconsider the salvage value of dams to achieve sustainability; promote sustainable dam projects; and influence the way policy makers and engineers have been approaching the design of dams. Ultimately, Palmieri says any sediment management programme must have a built-in economic incentive for the project owners.

Piece of the puzzle

UNEP's executive director believes that sedimentation, and the resulting loss of storage capacity, is just one piece in the puzzle of delivering sufficient clean water supplies to the world's people. And the hydro industry has its part to play.

'We must act to reduce the loss of forests and to re-afforest cleared areas as part of a comprehensive strategy of watershed management of the world's river systems,' Toepfer said. 'We must also act to reduce the threat of global warming. However there will always be natural levels of erosion that will contribute to a loss of water storage capability. So I call upon engineers to provide technical and environmentally friendly ways of extending the lives of reservoirs worldwide.'