The majority of the world's major rivers have been dammed, leaving a legacy of environmental and social harms that has truly changed the planet. But managing dams in ways that mimic natural river flows can help offset the worst damages.
The Lesotho Highlands Water Project (LHWP), which transfers water from the mountain highlands of Lesotho to South Africa, is one of the world's largest water-resource developments. In addition to affecting tens of thousands of people living in the reservoir area, the project has, to one degree or another, impacted some 150,000 people living downstream of the dams.
It was only as the first dam, Katse, neared completion in 1997 that international pressure forced an assessment of the project's impacts to downstream ecosystems and communities. The result was a study that led to significant advancement of the use of "environmental flows" (EF) in water-resource decision-making.
The LHWP EF study analyzed how changes to the way water was released from the dams could reduce the impact on both downstream river ecosystems and on the livelihoods of people living alongside them. Two years of data collection was followed by a structured evaluation of the effects of different kinds of flow change. The team predicted social, health and economic outcomes that were linked with biophysical impacts, then devised scenarios that included consideration of an array of issues to make possible a decision on the future flow releases from the dams. Environmental issues included changes in river form and function, and losses or increases in the abundance of riverine animals and plants. Economic issues included loss of royalties from the sale of water to South Africa, and compensation to downstream communities for predicted losses of riverine resources.
The scenarios formed the basis for protracted negotiations between the Lesotho Highlands Development Authority, the World Bank, and the governments of Lesotho and South Africa, which led to agreements on the volume of water to be released from the dams, the timing of releases, and the compensation payments to be made to people living downstream. The resulting Instream Flow Requirement (IFR) Policy in 2003 also specified operating rules for the dams and a program to monitor compliance with the agreed releases. The operating rules provide for changes to releases depending on climatic conditions, so that some natural variation is maintained.
When implementation of the policy began in 2003, the focus switched to monitoring downstream rivers and administering compensation payments to local communities. Compensation included back payments to villagers in river reaches in close proximity to the dams - i.e., within 40 km - for losses from reduced flows. These payments totaled millions of US dollars.
The first three years of monitoring revealed some early problems with meeting prescribed flow regimes - chiefly that dam operators released enough water, but too regularly to mimic natural conditions. Most of the changes recorded in the downstream rivers were linked to this loss of regular flooding. Nonetheless, in 2006-07, nine years after the completion of Katse Dam and four years after the completion of Mohale Dam and Matsoku Weir, the rivers downstream of the structures were either in their target ecological condition, or better than their target condition.
Initial results have also opened an opportunity to assess the accuracy of predictions made by the EF team. Although some changes are expected to take upward of 20 years, the monitoring results indicate that some aspects have not responded as predicted. Downstream of Mohale Dam, for instance, 64% of the biophysical indicators changed as predicted, whereas 25% showed no change and 10% changed in the opposite direction from predicted. For example, nearly all of the predicted changes to the physical shape of the riverbed occurred, whereas water quality was not as badly affected as predicted. Fisheries, on the other hand, showed some of the predicted impacts (such as reduction in the abundance of indigenous species), but others were the opposite from predicted (such as an increase in exotic species in some areas).
Significantly, one of the aspects where change was opposite from predicted was "woody vegetation," which was predicted to decline in downstream rivers under the agreed flow regimes. Wood is rare in the Lesotho Highlands but crucial for cooking and heating houses, thus compensation for the expected loss of this vital resource makes up more than half of the total compensation payments. Yet the initial EF monitoring showed that woody vegetation had increased, which understandably led to serious questioning of the EF's recommendations. The situation is, however, not straightforward. The new trees have colonized newly formed in-channel islands. This process was predicted in the EF study, and floods were included in the recommendations partly to flush these mini-islands and the vegetation that colonizes them. Blocking the river channel with sediments and vegetation causes a number of problems for river functioning over the long term, and eventually leads to a severe decline in river condition and loss of valued species. Furthermore, in-channel trees are at risk of being uprooted during big floods. Indeed, in 2006 a large flood overtopped Katse Dam and removed almost all of the in-channel trees and much of the fine sediment from the reach downstream of the dam. The flood also eroded cultivated fields, and flood debris was linked to the destruction of a bridge downstream. Each subsequent large flood will remove progressively more sediment from the system, and progressively fewer seeds and cuttings will reach the site (because they will be blocked by the dam). It is thus likely that, with time, woody vegetation will decline, as predicted.
On the cutting edge
The LHWP environmental flows process represents one of the first sets of activities for developments of its kind. It resulted in the mitigation of downstream environmental and social losses associated with reduced flows being accepted as legitimate project costs, changes to the dam outlet valves and operating rules, and a 300-400% increase in the downstream releases specified before the process. A key aspect of the approach was the development of the IFR Policy. The first IFR audit, completed in 2007, found that implementation had been 60% compliant with the IFR Policy and identified issues likely to affect the sustainability of the process, which LHDA has committed to addressing. Notably, the World Bank showed that the costs of the EFs (including compensation) amounted to only about 0.5% of project costs and that EFs did not significantly affect the project's Economic Rate of Return.
The LHWP EF process has also provided some important lessons for future projects, only a few of which can be addressed here. First among these is that EF work should be undertaken very early in a project to inform engineering design and financial modeling, and allow for baseline data collection to support later monitoring and improve EF assessments.
Secondly, large dams will have impacts on the rivers and other ecosystems downstream of them, only some of which can be mitigated through the provision of environmental flows. Furthermore, there is no such thing as a "minimum flow" - the greater the divergence from the natural flow regime, the greater the impact on the downstream environment. In the case of Lesotho, the "target ecological condition" for the rivers immediately downstream of the dams is lower than their pre-dam condition. Also, despite a commitment to compensation, not all of the losses incurred could be costed or even compensated for.
Finally, incorporating the requirements of downstream reaches requires commitment to a complex management process for the life of the dam and beyond. Mitigating the impacts of dams on the living ecosystems downstream is not simple, and requires genuine commitment to environmental flows and adaptive management, sufficient funds and appropriately skilled personnel. Such a commitment is more likely if supported by legislation and guided by a structured decision-making process.
The author is an integrated flow specialist from South Africa, and has worked in Africa, Europe and Southeast Asia. She was the coordinator for the first Lesotho study, and an advisor to LHDA from 2002-07.