The Vaal River System is complex: the Vaal Dam is augmented from upstream dams such as Sterkfontein (in turn fed from the Thukela through the Drakensberg Hydropower Project) and the Lesotho Highlands Water Project (LHWP) with its two large dams, Katse and Mohale. Currently the Sterkfontein Dam is full, but this is not the case with the Katse and Mohale dams, respectively 66% and only 36% full. This means the system is not flush – there is still space in the LHWP to store 1,000-million cubic meters of water, the equivalent of three Loskop dams.
Sophisticated computer models are used to operate complex water systems in order to maximise their effectiveness. It is, however, not necessary to do such an exercise to conclude that upper dams generally should remain as full as possible and for as long as possible. Withdrawal of water should be concentrated on lower dams, also for as long as possible. That way the opportunity to capture floods is optimised and evaporation losses are reduced as higher level dams generally are deeper, have smaller surface areas, and are located in cooler climates.
It is not clever to release water from the Katse and Mohale dams to the Vaal Dam when the latter is full, yet that is what is happening – at an average rate of 24.7 cubic meters per second, or about 2,100 Ml/d.
South Africa has been locked into this release by the old LHWP treaty with Lesotho, signed in 1986. A prescribed minimum flow was to be released, irrespective of the state of the dams downstream in South Africa. This suited Lesotho’s development of a hydroelectric station at Muela.
When negotiating the second phase of the LHWP, comprising the Polihali Dam and a tunnel to link to Katse Dam, the South African team pointed out the inefficiency of the operating rule embedded in the old treaty. It was agreed that an improved rule had to be developed, taking the full water resource system into account, counterbalanced by suitable arrangements to offset electricity supply losses and other costs that Lesotho might incur. This position was captured in the treaty of the second phase, signed in 2011, with the intent to have it settled before the start of construction. However, even though the start of the project was delayed by some five years, by 2018 very little progress has been made and there is currently no agreement for an improved operating rule. Therein lies the answer for the current storage imbalance in the system.
The Vaal Dam is currently spilling. At present the LHWP releases as much as 40 cubic meters per second, all of which is lost over the spillway of the Vaal Dam, and therefore not available to the users that pay for the project. As Gariep Dam has also started to spill, it may well be lost completely out of the Vaal/Orange system and run straight to the sea.
The only positives are the 80 MW being generated at Muela plus a few smaller plants along the way. These are minimal figures compared to Eskom’s total generating capacity; plus considering that its capacity is currently in surplus.
In the past, integrated systems analyses of the Vaal and Orange River systems have shown that the augmentation of the Vaal System was required by 2019, yet the completion of the second phase of the LHWP is already delayed to 2025. With such a strained system it is necessary that the releases from the LHWP should be ceased in order to fill the dams in Lesotho – saving the water for use when drought strikes again.
At the average release rate, this will mean closing the tap for at least one year. Banking the surplus NOW, rather than wasting it to the sea, should be done immediately. The governments of South Africa and Lesotho should urgently come to an agreement. Every day delayed means 2,100 Ml/d water wasted – more than four times Cape Town’s current target use! DM
Dr Peter van Niekerk previously headed water resource planning in the Department of Water Affairs and Forestry