Final Paper - Executive Summary
November 2000

The World Commission on Dams (WCD) selected Kariba Dam as 1 of 10 dams worldwide that would be studied to contribute to the current debate on the development effectiveness of large dams. The WCD developed common methodology that was followed in all case studies. This methodology comprised two phases, where Phase 1 was a scoping phase. The purpose of Phase 1 was to identify all the critical environmental, economic, social, and any other issues relating to the dam. The issues identified in Phase 1 were then examined in detail in the subsequent Phase 2 of the case study.

In the Kariba case study, the scoping phase was completed in August 1999, and was immediately followed by Phase 2 in the following September. This report addresses the issues raised in the scoping document and presents findings of Phase 2 for the Kariba case study.

Kariba Dam, constructed in 1955-59, with a storage capacity of 180km³, extending over a length of about 300km, and having a surface area of some 5500km² at full supply level, is one of the largest dams in the world. The dam was constructed on the Zambezi river, at grid reference 28.74778^o E and 16.51222^o S, along the border between the countries of Zimbabwe and Zambia and is jointly owned by the two countries. The main purpose for the Kariba Dam at the time of its construction, and to this day, is to provide hydropower to the two countries that own the dam. Two power stations with a combined generation capacity of 1320MW have been installed at the dam.

The scoping report for the Kariba case study identified a large number of issues for detailed study in Phase 2. These can be classified into the subject areas summarised below:

• an analysis of the history of Kariba Dam and the decision processes that were followed in the planning implementation and operation of the Kariba Dam project;
• an analysis of the impact of Kariba Dam on the supply of hydropower in the participating countries;
• an examination of the impacts of Kariba Dam on agriculture and irrigation in the area around the lake;
• an examination of the ecological impacts of Kariba Dam on surrounding areas, as well as on downstream areas;
• an examination of the impacts that Kariba Dam has had on the people, especially on those displaced by the lake;
• the impacts of Kariba Dam on wildlife;
• the impacts of the dam on the hydrology of the Zambezi river, especially downstream of the dam;
• the impacts of the dam on the economies of the participating countries;
• the impacts of the dam on fish and fisheries;
• the impacts of lake Kariba on the development of tourism around the dam;
• an examination of the impacts of the dam on public health and the spread of water borne diseases; and
• an analysis of the seismic impact of the dam. This issue arose from the observation that the frequency of earthquakes appeared to have increased in the Kariba area since the dam was built.

All the issues specified in the scoping report as critical to the Kariba case study were examined in Phase 2 and the results of these studies are presented in this report. By way of approach, this study relied mainly on the review of existing reports and publications. Fieldwork and original research was only carried out with respect to sociological impacts in which field visits were made to the communities that were displaced by Kariba Dam to record their experiences and views on the resettlement wrought by the dam.

The Project

Kariba Dam, constructed in the period 1955-59, is one of the largest dams in the world. The dam was constructed on the Zambezi River, at grid reference 28.74778^o E and 16.51222^o S, along the border between the countries of Zambia and Zimbabwe (called respectively Northern and Southern Rhodesia during the colonial times) and is jointly owned by the two countries. Kariba was designed as a single-purpose hydropower project, but as it turned out both fishery and tourism became important benefits. The main technical characteristics are as follows:

Type of Dam Double Curvature Concrete Arch dam
Height 617m
Spillway Gates 6 gates, 8.8m wide and 9m high
Discharge Capacity of Spillway 9500m³/s
Length of Reservoir 280km
Minimum Retention Level 488.50m
Minimum Operating Level 475.50m
Total Storage 180.6km³
Live Storage 64.8km³
Maximum Surface Area 5577m²
Depth of Stilling Pool 78m
Volume of Stilling Pool 410 x 10⁶ m³
Kariba South Bank Power Cavern 6 x 117.5MW = 705MW max capacity
Kariba North Bank Power Cavern 4 x 153.5MW = 615MW max capacity
Total Generation Capacity 1320MW

Predicted versus Actual Development

2.1 Project Cost and Construction Schedule

The project was built in two stages, first the dam wall and the Kariba South power cavern, and then in the second stage, the Kariba North Power cavern.

Stage 1 was originally assumed to cost £72.2 million ($1230 million: all costs converted to constant 1998 US dollars). This was revised upward to £79.38 million ($1350 million) during the tender stage. The cost increase was largely because of raising the dam by 6m, and the addition of one 100MW turbine unit. The final expenditure was £77.61 million ($1320 million), or 97% of projected cost, in spite of the addition of two more spillway gates and extra work on the south abutment. Stage 1 was completed in 1960, on schedule, in spite of two record floods during the construction period.

In the 1955 planning document Stage 2 of the project would immediately follow Stage 1 and was then estimated to cost $57.1 million. However Zambia decided to implement the Kafue project first, which delayed Stage 2 by several years. In 1970 a World Bank loan for Stage 2 was negotiated. The World Bank appraisal report mentions that with Kafue in operation, most of the power of Kariba Stage 2 was to be exported to Zimbabwe. The new target date for commissioning was 1974, with costs estimated at $195 million.

Due to unforeseen geological difficulties (there was only one borehole drilled at the selected powerhouse location) and the lack of experience of the first civil works contractor, who went into receivership, the project was delayed. The second civil works contractor was directly appointed (it was the company that had just finished Kafue) and progress of the civil works portion of the project was from then on satisfactory. However there were political difficulties between Zambia and Zimbabwe, which led to closure of the border, which meant that the hydroelectric equipment had to be re-routed through Angola or Tanzania. The first two generator units came through Angola, but then war broke out in Angola. The last two generators came through Beira in Mozambique via Southern Rhodesia, for which a special permit had to be negotiated. This all led to further delays and costs. Stage 2 was finally commissioned in 1976/77 and the final cost amounted to about $480 million.

The construction of Kariba caused a large number of fatal accidents, probably more than 100. Men were caught in machines, electrocuted, crushed by falling rock, killed on the road, plunged to death after scaffolding collapsed, and so forth.

2.2 Hydrology

Mean Flow

During the planning process flows measured at Kariba were correlated with those of Livingstone, at Victoria Falls for which a much longer period of data was available. The 1907-1923 data were discarded because they were of dubious quality (but indicating the occurrence of a major multi-year drought), and therefore only the 1924-1955 data series were used. On average the mean flow of the 1960-1998 period of operation are 10% higher.

Design Flood

Kariba was designed for the safe passage of a 1 in 10000-year flood. The original design was for a 3-month flood with a volume of 68km³, based on 1924-1955 data series. During construction, in 1957, a flood with a peak of 8200m³/s occurred, the highest on record, which led to revision of the design flood value from 68 to 74km³. In the following year a flood with a peak of 16 000m³/s and a 3-month volume of 61km³ occurred, and again the spillway capacity was increased, this time to 92km³. Had the project been built a number of years earlier with the original design capacity of 68km³, the spillway would have been completely under-designed.

2.3 Hydropower Generation

In the original planning document of 1955, the full station output was expected to be 6720GWh per year. The average output for the period 1977-1996 was about 6400GWh per annum, in spite of a slightly higher than planned maximum capacity of the turbines (1320 instead of 1200MW) and a 10% higher than expected flow. This indicates that the scheme is generally operated at lower heads than expected, and therefore also with lower efficiency.

2.4 Power Economics

If Kariba had not been built, then - in an all-thermal scenario - coal-fired steam plant would have been built in Harare (then called Salisbury), Bulawayo, Umtali and at several copper mines. The 1955 Kariba project report gives details. The data were used to construct a cash flow of power benefits (investment, operation and maintenance, fuel), which constitute the predicted power benefits attributable to Kariba.

In this study a cash flow was constructed for actual conditions: the thermal plant offsetting Kariba Stage 2 were delayed to 1976, the actual energy production of Kariba rather than the planned value was taken into account. In a few of increased open pit mining, the high prices of coal were gradually reduced from $30 to $20 per ton over a period of 30 years, which was the adopted life time of the thermal alternative. Then, when reinvestments were due, it was assumed that the stations would be built in Hwange, with a further reduction in coal price to $10 per ton, and that station efficiencies would be greatly improved. Coal was generally assumed to have a lower calorific value of 28GJ per ton.

On the cost side, the predicted development takes into account the project cash flow (investment, operating and maintenance cost) as published in the 1955 Kariba project report, with implementation of Kariba Stage 2 completed by the end of the '60s. In the actual case the real project cash flows are considered, with the delay of Kariba 2 to 1976.

The cash flow analysis resulted in an economic rate of return of 16.5% for the project as planned, and 14.5% for the project as built.

As a result of the relatively low construction cost of Kariba and Kafue the average electricity cost in the region dropped by about 30% in the period 1961-1977, while the average price for other commodities and services rose by more than 75%.

2.5 Resettlement

The pre-project planning document (1951) estimated the number of people to be resettled at 29 000. In the Kariba 1955 project document, there is little detail on the resettlement programme, except for a budget allocation of £4 million that was to be spent on this programme. A decision was made that each of the governments in Zambia and Zimbabwe would have responsibility for managing resettlement in their country. This decision meant that the resettlement programme was removed from the main project. The actual number of people to be resettled increased from 29 000 to 57 000. The budget for resettlement remained unchanged.

Unexpected Benefits, Costs and Impacts

3.1 Technical Aspects

Geological Problems

The extra works at the southern abutment and part of the problems encountered during the construction of the Northern Kariba power cavern, which both led to cost increases, were unexpected, but most likely the result of the drilling of an insufficient number of boreholes during the planning phase.

Problems with the Civil Works Contractor during Stage 2

There are diverging opinions on the construction delays and cost overruns regarding the north bank powerhouse, but the most plausible view is that the contractor submitted a low bid to secure the works. The second and third ranked contractor was about 80% more expensive. When unexpected geological problems occurred, which were not surprising as only a single borehole was drilled in the power cavern area, and the methods of excavation and rock securing had to be changed, the contractor could not fulfil his obligations and went into receivership. The time lost and the extra cost in recruiting a new contractor were substantial.

Earthquakes caused by Kariba Reservoir

The water storage volume of Kariba reservoir of 180km³ translates into a mass of 180 billion metric tons. The reservoir is located in a tectonically active area, at the southern end of the African Rift Valley. Since its construction and filling in the early 1960s, Kariba has caused numerous earthquakes in the area, 20 of them in excess of magnitude 5 on the Richter scale. This is of significance for dam safety.

The project documents for Kariba did not discuss the possibility of reservoir-induced seismicity and the need to take this into account in the design of the dam.

3.2 Environmental Impacts

As already stated, there was no Environmental Impact Assessment (EIA) done for Kariba and therefore baseline studies on any of the environmental parameters are not available. Here only a few of the impacts can be dealt with. For more information see the main report.

Drowned Vegetation

Almost 1000km² of the reservoir area were cleared as preparation to gillnet fishery. This constitutes about half the area of the top 20m of the reservoir, the depth occupied by most indigenous types of fish. The remaining 4500km² were not cleared. With respect to biodiversity it is unlikely that Kariba caused a serious extinction of species since most of the vegetation types in the flooded area are abundant elsewhere within the Zambezi valley.

The rotting vegetation caused the formation of H₂S, which when released caused severe corrosion of copper works, both in the powerhouse and downstream.

The eutrophication of the reservoir led to a boom in fish in the first years of its existence. Also various waterweeds, including the water hyacinth, were prolific. The situation normalised after a few years.

Operation Noah and the Creation of Wildlife Sanctuaries

"Operation Noah" began in response to concerns about the fate of animals that would get drowned when the reservoir would start to fill. A wide variety of 4 000-6 000 large animals and numerous small ones were rescued. Mammals were initially released to the nearest shore, but after 1961 most animals were translocated to other parts of the country. Of particular note was the translocation of 40 black rhinoceros to Hwange National Park.

How many of the rescued animals survived and found a new habitat is unknown.

This project led to the creation of wildlife sanctuaries along the southern lake embankment, which was one of the factors that stimulated tourism.

Increased Habitat for Some Species

The Kariba lake environment has been beneficial to some species such as crocodiles and aquatic birds. The number of crocodiles on the southern shore was around 10 000 in 1985. In the early stages of the lake the number of hippos and buffaloes may have declined, but it has since risen considerably as a result of the development of Panicum grassland on the shore.

Change from Riverine to Lacustrine Environment and Effect on Fish

Pre-impoundment studies on the mid-Zambezi identified 28 species in the area upstream of Kariba. The number of species now is 42, including those in the reservoir.

In the reservoir, cyprinids, which need flowing water, have almost completely disappeared, whereas Cichlids have become the main fish in the littoral zones of the lake. Pelagic zone fish were absent before impoundment, but Limnotrissa miodon was introduced in 1976/77. Fish yield in Kariba reservoir rose from 5.6 kg/ha per annum to about 33kg/ha per annum in 1986.

Fishery and fish farming have become one of the most important secondary benefits of the Kariba project.

Downstream Effects on Fish

Eels were found in the mid-Zambezi. Adult eels live in fresh water for 20 years before migrating to the sea to spawn. The Kariba Dam has created a barrier for eels and their numbers have declined in the mid-Zambezi.

The African lungfish, which lives in tropical vleis, pans and swamps has perished, and this may be the direct result of building Kariba. It inundated some swamps in the area now covered by the reservoir, and it regulated releases to the downstream area so that Mana Pools downstream of Lake Kariba is no longer swampy.

High DDT Concentrations

Among the persistent pollutants in Lake Kariba, DDT has been found to be the most dominant. DDT was extensively used during and after construction of the dam to eradicate the tsetse fly in the region, but also upstream in Namibia's Caprivi Strip and from agricultural areas upstream of the reservoir.

Resettlement of People into Wildlife Habitats

Some of the resettlement areas had prolific wildlife before resettlement took place. The resettlement of an essentially farming community caused a conflict with the wildlife in the area. In the Binga district for example, 64 elephants were killed in 1956, and 31 in 1957. In 1964 more land was made available for resettlement and 120 elephants were killed. Altogether 17 rhinos were captured and translocated to Hwange National Park.

Downstream Effects

As Kariba regulates most of the incoming floods, the number of times that the downstream Mana Pools are inundated is less than in the pre-dam period. This reduces the deposit of fresh alluvial material, which has led to a decline in grazer density. At the same time the occurrence of the dominant canopy tree Faidherbia albida declined and the occurrence of termites Macrotermes increased.

The downstream impacts of Kariba Dam extend all the way to the Indian Ocean. Because of the dam the natural rhythm of flow of the Zambezi has been changed and evened out. The seasonal high and low floods do not occur as much as they did before Kariba. As a result, the Zambezi River does not break its banks as it used to, and the delta floodplain ecology has been negatively affected. Shrimp catches have decreased, floodplains have been invaded by upland vegetation because of the absence of annual flooding, mangrove are dying off because of poor flooding of coastal areas, productivity of artesianal fisheries in the delta area has decreased. And wildlife populations in the delta areas have been negatively impacted upon.

3.3 Social Impact

Resettlement

Population affected by the Dam

Before resettlement the Tonga inhabited both banks of the middle Zambezi Valley from Kariba Dam to Mlibizi. Villages were situated along the Zambezi and around the deltas of the tributaries. The main occupation of the people consisted of riverine and upland farming, livestock rearing, hunting, fishing and manufacturing. Riverine farming was based on recession agriculture, which depended on the flood regime of the river. Away from the flood plains, on the poorer soils of the upland valley area, shifting cultivation was practised. Typical crops were maize, bulrush millet, sorghum, sweet potatoes, groundnuts, a variety of cucurbits, tobacco and cotton. Of these, tobacco was primarily a commercial crop.

Due to the annual flooding of the Zambezi that fertilised the riverine gardens and deposited alluvial soils the Valley Tonga could harvest twice a year and were seldom victims of hunger and famine.

Livestock rearing was important although the tsetse fly invasion of the 1830s and the rinderpest epidemic of the 1890s reduced the cattle population. Goats, sheep and fowls became important instead. The restrictive game laws introduced in Zimbabwe in 1906 and Zambia in 1925, led to an increase in game and tsetse fly. Prior to the enactment of these laws, hunting was one of the Tonga's main economic activities and a major source of animal protein. Fishing was regarded as secondary economic undertaking, in spite of their proximity to the river.

That said, the pre-dam livelihood conditions should not be overly idealised. Health care was virtually absent and the Tonga suffered from many diseases, including leprosy, hookworm, malaria, bilharzia, dysentery, and sleeping sickness. Communication and transport were virtually non-existent, although there were relatively limited needs for such facilities. There were no roads, only footpaths, so no vehicles could reach the valley. The presence of wide animals made travel by foot rather dangerous and men and women often went in groups whenever they had to make long journeys.

Resettlement Process

The 1955 Kariba project report, the 1956 World Bank appraisal report and the annual reports of the Central African Power Corporation (CAPCO) include a one-line cost estimate for resettlement, and offer no further discussion. The racist attitude of the time did not consider the resettlement of Africans as a problem.

The Central African Council on the Kariba/Kafue Hydroelectric Power Committee in its 1951 annual report estimated that in Northern Rhodesia 14 300 and in Southern Rhodesia 15 000 Africans "would be affected by inundation" and that "no provision has been made in the estimates for the establishment of these persons, but it is understood that suitable land is available in local territories for this purpose". As it turned out, the actual number of people to be resettled was 57 000 and suitable land was not sufficiently available in the direct vicinity.

The cost of resettlement had to be borne by the developer, the Federal Power Board, but the responsibility to actually undertake resettlement was left to the territorial governments. In Southern Rhodesia 23 000 people had to be moved; food would be provided during the resettlement period; adult males would be exempted from the annual poll tax of £2 for two years. In Northern Rhodesia compensation would cover: 34 000 people to be moved; resettlement costs; compensation to the individuals moved; tribal compensation for hardship, inconvenience and loss of tribal lands and customary rights; compensation in respect of loss of earnings while clearing new lands at the rate of £5 per acre, allowing one acre per person; compensation for the lost earning while building new huts at the rate of £10 per hut; and compensation for the loss of crops. In both countries, the Tonga were neither well informed nor consulted. When they heard that they had to abandon their ancestral land and move to new areas that were barren and hilly, and not situated at the river, they at first thought that the white men tricked them to take over the fertile lands at the river.

The resettlement plan aroused the Tongas' anger and caused strong anti-government feelings. Some were prepared to fight and even die for their land. In June 1958 for example, a group of people in the Chiefdom of Chipepo stoned a district commissioner. Three months later, in the same area, anti-resettlement protests culminated in violence when a group of men armed with spears, pangas, knobkerries and shields attacked a police party. In return, the police opened fire, killing 8 people and injuring 32.

Actual resettlement took place in 1957 and 1958. It was reported that the people to be resettled "were treated like animals or things rounded up and packed in lorries" to be moved to their new destination.

Most of the new land was of poor quality and easily erodable. Also, as no recession agriculture was possible due to the far distance to the river, only one crop per year could be produced. Resettling too many people to areas too small aggravated the problem. It is therefore not surprising that food production decreased and famine occurred in the first years after resettlement.

In later years many more problems occurred, caused by lack of water, breakdown of wells and other basic infrastructure provided as part of the resettlement programme, as well as influx of commercial fishermen.

There are a few things, which are certainly better than in pre-Kariba times, such as the access roads to the area, schools and medical facilities. But not all promises made during the resettlement campaign have been met. Most villages are still without electricity.

During the first years after resettlement, the Tongas were allowed to cross the lake and meet relatives and friends at the other side. However, the frequency of visits was no longer the same given the long distances involved, as some were located more than 100km from the lake. Furthermore, when the Federation of Rhodesia and Nyasaland broke up in 1963 and Zambia became independent in 1964, free movement ceased to exist and border posts were established at Chirundu and Kariba. From that time, the Tonga could no longer afford the long journey through the newly established border posts. This meant that relatives and friends on the opposite side of the border lost contact with each other.

Reparation and Rehabilitation Programmes

Currently several programmes are underway to make up for the mistakes of the past and to uplift the living conditions and employment opportunities of the population affected. The Zambezi River Authority (ZRA), the Zambia Electricity Supply Company (ZESCO), the Southern Africa Development Bank (SADB) and the World Bank play an active role in this effort.

Health Aspects

Health issues were not included in the original Kariba Project Report. However the Kariba Lake Coordinating Committee, formed after the decision to build Kariba, surveyed the incidence of diseases prior to impoundment and instituted programmes for disease control after the establishment of the dam.

Schistosomiasis

Prior to construction intermediate host snails were not found at the dam site, but they were present north of the gorge where Kariba Dam site is located. The possibility for an outbreak of schistosomiasis once the reservoir filled up was forecast.

Construction workers, who came from all over Zimbabwe and Zambia were treated. About 45% of them were infected.

Intermediate host snails were discovered in the lake, most frequently at places used by humans for domestic, recreational or occupational purposes. Dispersion of the snails seemed to be facilitated by drifting Salvinia auriculata.

The control programme consisted of spraying infected areas with nuclosamid and keeping shorelines free of the Salvinia auriculata. This programme lasted until the 1980s when the incidence of schistosomiasis was on the decline. From then on the number of people infected seems to have gradually grown until about 1990. Then there is a slight decline, but numbers are on the increase again due to a general drug shortage, coupled with low staff levels at the health centres, particularly in Zimbabwe.

In Kariba town the incidence of Schistosomiasis in the area is about 20%, with about equal shares of Schistosoma haematobium and S. mansoni.

The treatment of an infected patient costs about $0.10.

Malaria

Malaria was prevalent in the area before dam construction. The construction workers were required to take prophylactic drugs. The contribution of Kariba to the disease was restricted to the construction phase due to the creation of borrow pits, poor living conditions, lack of understanding the importance of prophylaxis and a high susceptibility among the immigrants.

The reservoir does not in any way worsen the malaria situation, as the vector mosquito does not breed in large water bodies.

Other Diseases

During the construction period, a high number of sexually transmitted infections were observed. Of late, there is a steep increase in the number of HIV/AIDS cases in Kariba, as a result of tourism and the fishery industry (transport).

3.4. Other unpredicted impacts

Tourism

A thriving tourism industry has developed around lake Kariba since the dam was built, based on water sport, wildlife resources of the area and infrastructure (eg, airport and paved roads) provided by the Kariba project. 17 hotels have been constructed, 9 of them in Zimbabwe. They have a total of 933 beds, 706 of them in Zimbabwe. The positive and negative impacts of this tourism development should have been recognised in the project document.

Irrigation

The potential use of Kariba water for irrigation was not investigated as part of the project. A number of irrigation schemes have been established around the lake, and are drawing water from this lake. Fortunately, the amount of water being extracted by these irrigation schemes is too low to have any impact on power generation. In addition, analysis carried out in this study showed that more irrigation development than has been established to date, could have been carried out for the benefit of the people of both countries without significantly affecting the power generation capacity of the dam.

Fishing

One major impact of Kariba Dam that was not mentioned in the project document is the fisheries industry that has developed on the lake. A large kapenta fishing industry developed, mainly in Zimbabwe, following the introduction of the fish from Lake Tanganyika in the 1960s. Unfortunately, the capital intensive nature of the kapenta fishing business makes it difficult for most of the local displaced people to easily participate, and the industry has therefore not significantly benefited the local people who were displaced by the dam.

Distribution of Costs and Benefits

It is clear that the main losers of the project were the Tonga people who had to be resettled, although a few of them - who used the compensation money wisely - actually became better off. But for most of them the resettlement was a traumatic event, in which they lost access to their ancestral grounds, areas suitable for recession agriculture and easy access to their friends and relatives across the river. Losses were all encompassing: monetary, psychological, cultural and social.

Those who gained were millions of electricity consumers, the copper mines and other industries, who could enjoy low prices for electricity. Other beneficiaries include employees of the national parks created as a result of Kariba, fishermen and workers in the fish industry, and those who found employment in the tourist industry. People living in the area also benefit from improved access roads to the area. Most of these opportunities (fisheries, tourism, irrigation and wildlife development) benefited people from outside the dam basin. The local people, especially those displaced by the dam generally failed to compete for these economic benefits for a number of reasons which include lack of capital or professional skills or education. Thus, although the Tonga bore most of the social costs associated with Kariba Dam, people from outside the dam basin took up most of the benefits.

The people living in areas where, without Kariba, coal plants would have been built, enjoy cleaner air and less acid rain, and are indirect beneficiaries. The global population and environment gained from the project also, as Kariba offset coal-fired generation which has very high greenhouse gas (GHG) emissions.

Decision-making

Intense debate about the need to construct Kariba Dam began with the end of the Second World War in 1945 and the drive for industrial development that followed. At that time, both Southern and Northern Rhodesia were colonies of Britain, and the governments of the two countries were closely linked. In fact, plans were already afoot to form the Federation of Rhodesia and Nyasaland, consisting of the three territories of present day Malawi (then called Nyasaland), Northern Rhodesia and Southern Rhodesia.

The intensification of copper mining activities in Northern Rhodesia between 1920 and 1945 and the rapid expansion of the manufacturing sector in Southern Rhodesia brought about a need for a cheap and stable source of electric power in the two countries. The copper mines of northern Zambia were developing fast in response to a strong demand for the mineral worldwide and high copper prices on the international market. At that time, the copper mines were being supplied with electricity by a number of small coal-fired power stations on the Copperbelt of Zambia.

Investigations prior to 1946 had identified two suitable sites for the establishment of a hydroelectric power station to meet the demand for power being shown by the Copperbelt of Zambia. The first major step towards the establishment of a large hydropower plant was made in 1946 when the governments of Northern and Southern Rhodesia formed the Inter-territorial Hydroelectric Power Commission. The commission was composed of representatives from the two countries and responsible for investigating the most efficient way of solving the power shortages that were being experienced. In 1948 the commission appointed an advisory panel to decide upon the relative merits of the construction of the Kariba and Kafue dams as power generating schemes. The panel initially recommended Kariba, particularly in light of the inadequate hydrological information available for Kafue. The irrigation potential of the Zambezi downstream of Kariba and the possibilities of fishery and other developments related to the lake were pointed out, but also with the recognition that the larger and more costly Kariba scheme would take longer to construct. As the power needs on the Copperbelt were becoming critical, the Northern Rhodesian Government was keen to see development at Kafue and accordingly invited the panel to make a detailed assessment of that scheme.

When the Federation was formed in August 1953, the support of Southern Rhodesia for initial development at Kafue had apparently been gained, largely on grounds of reduced costs and speed with which it could be constructed. This was seen as essential for the well being of the Copperbelt as well as the Federation. However, in spite of the debate and application for a World Bank loan for Kafue, Kariba was still considered by some in the Federation as a more favourable site, whilst the potential at Kafue was seen to have been overestimated.

Following the formation of the Federation in 1953, the Inter-Territorial Power Commission was replaced by the Federal-Hydroelectric Board which was established in May 1954 to further the development of both schemes, the Kafue then being seen as a precursor to the larger project at Kariba. To resolve the argument for Kariba versus Kafue, a panel of experts was sought to examine the two projects and to advise on which the board should implement first. Once again the panel advised in favour of the larger Kariba scheme and the Federal Government accepted that advice in March 1955. Construction at Kariba started later that year. Predictably there was a major outcry in Northern Rhodesia, especially in Lusaka, but to no avail.

In 1956 the World Bank concluded its appraisal of the project and stepped in with finance. It is interesting to note that a substantial share of the funding came from the copper mines, which were in dire need of cheap power.

When it commenced operation in December 1959, Kariba fell under the Federal Power Board. In 1963 the Central African Power Corporation was established, and this organisation was replaced by the ZRA in 1987.

The ZRA was responsible for all matters concerning Kariba. Furthermore it was to collect hydrological and environmental data for the Zambezi River, and subject to the approval of the council of ministers operate and maintain any other dams on the Zambezi. In 1987 ZRA's responsibility for generation and transmission was handed over to the two utilities, the Zimbabwe Electricity Supply Authority (ZESA) and ZESCO.

Compliance

In the 1950s there were not many laws and regulations in place for the construction of a major dam project in North and South Rhodesia. When rockfalls occurred during the excavation of the Kariba North Bank Cavern, Zambia was seeking to invoke its Mines and Minerals Act, which contained a number of safety measures. According to the World Bank this "made it virtually impossible for the contractor to implement its work programme". Zambia eventually agreed to lift the act for the rest of the construction work.

There were no internationally binding design conventions, but by hiring reputed consultants it was assumed that all works would be state-of-the-art.

In terms of environment there were no regulations whatsoever. Voluntarily Northern and Southern Rhodesia approached the government of Portuguese East Africa (now called Mozambique), to negotiate a minimum release from the dam. It was agreed that this mandatory release would be 10000cusec or 283m³/s. This value is still being adhered to.

In 1957 the Kariba Lake Development Company (KLDC) was formed. KLDC was formed hurriedly, with a brief to cover a wide range of subjects, including transport, industries, fisheries, national parks, tourism, irrigation and forestry.

The formation of KLDC is significant in that the committee effectively took over the responsibilities for environmental protection and management on behalf of the project. KLDC commissioned a fish survey, carried out studies on water borne diseases, was involved in setting up wildlife sanctuaries, etc. KLDC thereby developed the beginnings of ad hoc environmental and social regulation.

Views on Development Effectiveness and Lessons Learnt

The ultimate objective of the WCD case study process is to make an assessment of the overall development effectiveness of large dam development and to derive lessons for other large dam projects worldwide.

Assessment of the Development Effectiveness of the Kariba Project

The concept of development effectiveness is a subjective one. It represents the overall opinion, all things considered, that a person has of a dam project. In the WCD case study methodology, the development effectiveness of a dam was to be evaluated by the stakeholders of the particular dam. The stakeholder meetings were used for this assessment, with questionnaires being administered on the stakeholders.

During the second stakeholder meeting for the Kariba case study, the stakeholders were asked for their view on the general effectiveness of the dam and to explain what they believed to be useful parameters on development effectiveness of dams. The stakeholders engaged in intense debate on this issue. In the end they agreed on a set of questions that must be asked in order to evaluate the development effectiveness of large dams. The following are some of the questions that the stakeholders believed should be used in assessing the development effectiveness of large dams:

• Was every possible development option considered and exploited in the in the planning and implementation of the dam project?
  According to the stakeholders, the decision to implement an effective large dam project should be preceded by an analysis of all possible development options to ensure that these are incorporated into the project. In the case of Kariba, one of the main weaknesses of the project according to the stakeholders, was the failure by the project planners to look beyond the provision of hydropower to the two participating countries. The stakeholders were of the view that development options in fisheries, tourism, irrigation and rural electrification should have been considered at the planning stage and implemented as part of the project.
• Did the project increase regional co-operation?
  In the view of the stakeholders, a dam is more effective if it promotes regional co-operation. The stakeholders felt that large dams are best implemented in a regional rather than national context. In this regard, Kariba was viewed as being effective by the stakeholders because it became the nucleus for the development of the Southern African Power Pool and the power from the dam led to the industrial development of Zimbabwe and Zambia.
• If people were displaced by the dam project, did these displaced people benefit from the dam project?
  For a dam to be considered effective, the benefits from the dam must be enjoyed by those displaced by the dam. This is one area in which the stakeholders found Kariba to have failed the test of development effectiveness, because the Tongas who were displaced by the dam benefited little from the project.
• Did the project meet the required rate of return?
  This is a purely financial and economic consideration, whereby an effective dam project should meet the predicted economic performance targets.
• Did the project exceed original objectives?
  According to the Kariba stakeholders, an effective dam project should exceed original objectives. The stakeholders noted that Kariba did in fact exceed the original stated project objectives and that it can be viewed as having been effective.
• Was there an equitable distribution of benefits from the project?
  This was another one of the issues that were most hotly debated at the stakeholder workshop. The stakeholders were reacting to the fact that most of the benefits from Kariba were not being enjoyed by local people who had suffered most as a result of the lake. In the view of the stakeholders, an effective dam project is one in which an effort is made to ensure that the benefits are distributed as equitably as possible. This was one area in which the stakeholders believed that the Kariba Dam project had not been effective.
• Did the living standards of the people improve as a result of the project?
  An effective dam project should be associated with an improvement of the living standards of the people, especially those displaced by the dam. Again, the Kariba Dam project was viewed by the most of the stakeholders as not having been entirely effective because the living standards of the displaced people were prejudiced.
• Did the affected people participate in the decisions that related to the project?
  In an effective dam project, the local affected people must be allowed to participate in decisions that relate to them. In the case of Kariba, the people were not afforded the opportunity to participate in the decisions that related to them and the dam, and stakeholders were mainly of the view that the Kariba project was not effective here.
• Was an EIA carried out and was a costing of impacts carried out?
  An effective dam project should have an EIA and an environmental management plan. Since no EIA was carried out at Kariba, this was one area where the project failed. However, the stakeholders excused Kariba on this issue because at the time, environmental issues were not yet in vogue.
• Is the dam project multi-purpose?
  This question is linked to the one where the stakeholders believed that all development options must be investigated in the planning of a large dam project. The stakeholders felt that because the planners for Kariba confined themselves to hydropower issues, some opportunities for multi-purpose use were lost.
• Were efforts made to minimise the displacement of people?
  The decisions that were made on the size of Kariba were aimed at maximising the amount of power from the project through the construction of the largest dam possible. No analysis was carried out to reconcile the need for a large dam and power station; and the need to minimise the number of people to be displaced. In fact the Federal Power Board went on to increase the height of the dam wall by 20 feet resulting in the increase in the area inundation and a concomitant increase in the number of people to be displaced.

The stakeholders for Kariba Dam had divergent views on the different aspects of the development effectiveness of Kariba Dam. Views were mainly divergent with respect to technical and economic issues, but there was general consensus on issues relating to social impacts of the dam. This consensus is not surprising when one considers the high awareness of the stakeholders with respect to these social issues.

When the stakeholders were asked to give their overall assessment of the development effectiveness of Kariba Dam, 15.5% rated Kariba very highly effective, 56% high, 13% were neutral, while the remaining 15.5% rated Kariba poor. None of the stakeholders rated Kariba in the lowest category of very poor. Therefore overall, the negative social impacts notwithstanding, the Kariba stakeholders viewed Kariba Dam as having been effective. These figures need however to be considered with care, given the fact the group who attended the stakeholder meeting is far from being a perfect reflection or a statistically representative sample of the broader population affected by or interested in the Kariba Dam.

Lessons Learned

The information and data collected by consultants with respect to the Kariba Dam and the views expressed at the stakeholders meetings were carefully examined to isolate lessons that may be useful or inform large dam development in the future. The following is a list of lessons that were identified in the Kariba case study for the large dam debate. The consultants drew the list of lessons first and presented them at the second stakeholders meeting. The latter made their contributions and editions to the consultants' list of lessons. The lessons are:

The design of spillways for large dams should make allowance for hydrological uncertainty. For safety, the dam spillway should be designed for the Probable Maximum Flood (PMF), which would be the highest flood that can physically occur.

Kariba was designed for the safe passage of a 1 in10 000-year flood. In the original design of the dam, the spillway had been designed for a 3-month flood of volume 68km³. During construction, in 1957, a peak flood of 8200m³/s occurred, the highest on record. As a result, the dam engineers revised their spillway design to a 3-month flood of 74km³. In the following season, 1958, a peak flood of 16000m³/s was recorded and a 3-month flood of 61km³ occurred. This led to a further revision of the spillway design to its present capacity of 92km³. Had the project been completed before these two floods, the spillway would have been seriously under-designed, with equally serious safety implications.

Hydroelectric schemes may have far more positive implications than just the production of electricity.

The Kariba project provided access roads to hitherto isolated area; led to the development of a thriving fishery industry; triggered the formation of important wildlife sanctuaries; became a major tourist attraction; and was the basis for an interconnected electricity network of two countries. These developments cannot all be predicted accurately, but in future studies the development potential of non-electricity benefits should to be accorded more attention.

Major hydropower projects can lead to continuously low tariffs, with major benefits for the residential, commercial and industrial consumers.

As a result of the relatively low construction cost of Kariba and Kafue the average electricity cost in the region dropped by about 30% in the period 1961-1977, while the average price for other commodities and services rose by more than 75%. It must be said however that in Zimbabwe ZESA almost went bankrupt in this time, as prevailing tariffs, dictated by the government, did not reach adequate levels.

The study found that there is good anecdotal and non-quantitative evidence of a good correlation between GDP of Zambia and Zimbabwe and electricity consumption. The completion of Kariba resulted in growth of manufacturing and mining sectors in Zambia and Zimbabwe respectively. For example, following Kariba, Zimbabwe established high power consuming industries such as fertiliser and ferrochrome plants.

Large man-made reservoirs can cause earthquakes, especially when they are constructed in a tectonically active area, as is the case of Kariba. The induced seismicity may affect dam safety and lead to other damages such as flooding of downstream areas and should therefore investigated in the planning stages of the dam to ensure dam safety.

The water storage volume of Lake Kariba of 180km³ translates into a mass of 180 billion metric tons. It was further noted that the lake is located in a tectonically active area, at the southern end of he African Rift Valley. Since its construction and filling in the early 1960s, Kariba has caused numerous earthquakes in the area, 20 of them in excess of magnitude 5. This has significance on dam safety.

The project documents for Kariba did not discuss the seismicity of the Kariba area, and the need to take this into account in the design of the dam. The planning of any future dams that are of similar size to Kariba, and which are located in tectonically active areas will need to take account of the potential dam-induced seismicity in the design.

With the numerous unexpected impacts that arose since its completion, the Kariba Dam illustrates the importance of systematic impact assessment in the planning of large dam projects. Many of the negative impacts of Kariba could certainly have been avoided if some impact assessment had been applied in a systematic way at the time. The impact assessment should as a minimum, address the following components: environmental impact assessment (EIA), social impact assessment (SIA) and health impact assessment (HIA)

In the Kariba project, the project document and the World Bank project appraisal report did not consider any environmental impacts of the dam. These documents confined themselves to the issues of the construction of the dam wall, the power stations, associated civil works, and their costing. As a result, the Kariba Dam had numerous unexpected impacts, especially in the environmental fields of wildlife, water pollution, tourism, water borne diseases and fisheries, and in social sectors (displacement and resettlement in particular).

A dam project is governed and guided by the prevailing laws of the country. In situations where the laws are unjust, it is difficult for the project to deliver benefits equitably, and to minimise social and economic costs. As a minimum, it is important for a dam to ensure that the land rights of the people (especially the tribal land rights) are not lost as a result of the project.

Because of the differences in the political settings in Zambia and Zimbabwe (Northern Rhodesia, an "indirect rule" colony, which became independent 5 years after the completion of the dam, and Southern Rhodesia, a white settlement colony until 1980), the resettlement procedures were different in the two countries. While still far from optimal, the resettlement of the Tonga in the North was less inhumane than what happened in the South. That said, in both countries, the laws were such that people who were displaced by the dam had little protection by the colonial authority of the time. According to the laws of both countries peasants did not have title to the land on which they were settled. They only had usufruct rights, and the land belonged to the state. Because of this legal situation, there was no need for the project to take the interests and land access rights of the local people into consideration. If the laws had been just, and the people had had some land access rights, the project developers would have been compelled to consult the people and to take their rights and interests into account.

In dam projects, cases of involuntary resettlement require detailed planning and the full participation of the affected people in the planning process. In addition, the planning process must be carried out well ahead of resettlement process and must ensure adequate infrastructure in the new areas of settlement to minimise the trauma of resettlement for those displaced by the dam. As far as possible, the displaced people must be equally or more comfortable in their new settlement areas than their areas of settlement in the dam basin.

The Tonga people who were displaced by Lake Kariba were never involved in the planning of their resettlement. All evidence available shows that the government officers of the day did all the planning of the resettlement exercise and the general selection of the new settlement areas for the Tonga. The Tongas were not provided with an opportunity to explore alternative ways of resettlement and to state their preferences. Another weakness of the resettlement programme is the fact that it was hurried. Not enough time was allowed for the Tonga to be prepared for the relocation, with the result that some had to be rescued on islands as the waters of the lake rose.

On the same issue of the need to involve the affected people in the planning of their resettlement, dam project developers should involve the people and negotiate with them on the most appropriate manner of resettlement. In the Kariba case study, it is reported that displaced people were bundled into lorries and left at their new homes. Cattle were driven over long distances on the hoof and it is reported that some died in transit. Any materials of cultural value, such as clay pots, which could not be transported on the lorries, were left behind. These losses or some of them could have been avoided if the people had been involved in the planning of the move and allowed to participate in decisions on the best way of relocation.

Where resettlement process leads to community fragmentation, one of its potential consequences can be a loss of cultural identity.

Some of the displaced Tonga people were settled in areas where they became minority groups. In these cases (examples of Tongas settled in Sambakarouma and Binga in Zimbabwe), they ended up speaking the languages of their host communities, and lost not only their native languages but also their cultural Tonga identity. This loss of identity has been accentuated by the fact that the frequency and feasibility of visits between groups settled in the South (Zimbabwe) and those settled in the north (Zambia) were practically eliminated after impoundment because of the long distances involved (some were located over 100km from the Lake), and because of the tightening of the conditions for crossing the border, especially after the Federation of Rhodesia and Nyasaland broke up in 1963. As time passed, relatives and friends on the opposite sides of the border lost contact with each other. It was during the stakeholder meeting organised to discuss the draft Kariba report that some of the representatives of these communities met for the first time since the displacement took place, 40 years ago. A 40-50 year old traditional Zimbabwean Tonga chief who attended the Kariba stakeholder meeting needed an interpreter in order to communicate in Tonga with other chiefs.

The effectiveness of a dam project must be evaluated on the basis of the extent to which it provides meaningful development opportunities for the people who are affected by the dam.

In the Kariba case, it would be fair to say that the displaced Tongas were never viewed as stakeholders in the dam or potential owners of the dam. As a result, there was no attempt on the part of the dam developers and the governments of the day to investigate ways in which the Tonga could maximise their benefits from the dam. In Zimbabwe, for example, the settlement areas were located far from the new water body where they could not easily access the dam for fishing and/ or irrigation. Some of the communities were settled more than 120km from the dam, with most of them being about 50km away. It is not feasible, from these distances, for the displaced people to take advantage of the dam.

The Kariba Dam was established for the production of electricity, and yet very few of the Tonga settlement areas are connected to the national electricity grid. In Zimbabwe, the first Tonga area to be connected to the national grid was Binga in 1985, 25 years after the first generator went into operation at Kariba. Since 1985, no other areas have been connected in the Tonga areas of Zimbabwe. It therefore follows that the Tongas of Zimbabwe were never provided with an opportunity to benefit from the main product of Kariba Dam.

In resettling people who are displaced by large dam projects, efforts must be made to ensure that the relocated people are familiar with the agro-ecology of the new areas of settlement, and that they have the necessary agricultural skills for their new settlement areas. If the new and old areas of settlement are agro-ecologically different, the project should make budgetary allocations and adequate plans for the provision of long term, appropriate agricultural training for the settlers, to ensure success. This recommendation applies to areas where agriculture is the main occupation for the displaced people. However, where the affected people are involved in non-agricultural activities, the project must ensure that the people receive skills training in their field of livelihood.

The people who were displaced by Kariba in Zimbabwe were settled in areas that were agro-ecologically different from their original home on the banks of the Zambezi River. The agricultural studies carried out showed that while the Tonga were able to grow two crops per year in some areas along the Zambezi River, their new areas of settlement are in dry areas of unreliable rainfall and generally infertile soils. As a result, the Tonga have not been able to become self-sufficient in food over the last 40 years.

One of the reasons why they failed to become self-sufficient in food is that they were never provided with the training to manage agriculture in the new areas. No investment was made into the training of these settlers to develop appropriate farming systems for their new area.

Resettlement is not completed when the people affected are relocated. In addition to compensation aimed at enhancing livelihood conditions in a sustainable way, continuous support is necessary over many years to help them to get adapted to a new style of life.

Compensation packages offered to displaced people varied between the countries of Zimbabwe and Zambia. The Zambians were provided with cash compensation paid to each individual as well as funds for some development programmes in the new areas of settlement. On the other hand, the Zimbabweans did not receive any cash payments to individuals and there was little provision for development programmes in the new areas of settlement. The present study found that the displaced people in Zambia were unhappy with the compensation that was offered, considering it inadequate.

40 years after commissioning the project, the people resettled in both countries because of Kariba still need help. Part of the project revenues should become available for this purpose.

The lesson that can be learnt from this observation is that designing an appropriate compensation package for people displaced by a dam is a complex and difficult exercise, and that dam developers need to be creative and to go beyond the provision of cash handouts. Compensation needs to be aimed at providing the displaced people with an opportunity to achieve a sustained improvement in their livelihoods. Compensation is best given in a form that provides opportunity to the displaced people to become economically self-reliant and must be consistent with the noble aspirations of the community.

In the Kariba Dam project, the compensation that was offered to the displaced people, in Zambia, was on a pari parsu basis (ie, a hut for a hut). This approach misses the opportunity to develop a compensation profile that enables the displaced people a chance to participate in economic benefits arising from the transformation of their land resources.

The United Nations principles on habitat, which state that "whatever the original conditions of the displaced community habitat amenities, the new homes must meet the basic needs of comfort, health and dignity". This United Nations dictate is exactly opposite to the "no worse no better" approach that was followed in the Kariba project by the Zambian territorial government in determining the situation of the displaced people after resettlement.

The Kariba study also shows that compensation alone does not suffice in many cases to guarantee improved livelihood conditions in a sustainable way.

This lesson was proposed by Kariba stakeholders at their second meeting, and it arose from the observation that all the complaints and expectations that the Tonga have with respect to the Kariba Dam would be addressed through legal means if there had been a legally binding agreement between the people and the project. At the time that Kariba was constructed, some verbal promises are reported to have been made by the governments of Zambia and Zimbabwe with respect to what the project was going to do for the displaced persons. The local people claim that many or some of these promises were not fulfilled by the project. For that reason, the local people now wish that they had recourse to a legally binding agreement for redress.

Effective institutional structures should be established to monitor and attend to all identified potential negative impacts after the implementation of the dam project. In projects such as Kariba, which are inter-country, it is important to ensure that these institutions can survive any changes in the political relations of the countries.

In the Kariba case study, the project did not carry out any impact assessment. However, if the project had done so, it would have been important to ensure that effective institutions were identified to address any negative impacts after the project has been implemented. This lesson arises from the observation that the Kariba Lake Coordination Committee (KLCC) ensured that some of the potential negative impacts of Kariba were addressed by institutions that they established. An example is the assignment of all issues pertaining to wildlife to the game department, and how the latter has managed the wildlife of the area effectively ever since the dam was constructed. Similarly, through the efforts that were initiated by the KLCC, the Lake Kariba Fisheries Research Institute was established to oversee all issues relating to fisheries on the lake.

The role played by Central African Power Company (CAPCO) in managing the dam and power stations during the 1965-1980 period when the relations between Zambia and Zimbabwe were at their lowest shows the importance of strong inter-country institutions. The recent (1987) establishment of he ZRA as an inter-country body with wide ranging responsibilities, which include studying and making recommendations on environmental and social impacts of the dam as lessons for future dam developers, is further testimony in favour of strong international institutions. It is interesting that the idea of the ZRA had been suggested as far back as 1948, by the Inter-Territorial Power Council, but it had to wait 30 years before it could be implemented.

The corollary of this lesson is that the implementation of large inter-country dam projects must be accompanied by the immediate establishment of strong, legally constituted institutions that have clearly defined responsibilities with respect to the addressing of negative impacts. Without effective institutions it is not possible to address negative impacts of dams.

As shown in the case of the Kariba Dam, a number of initiatives (trust funds, targeted development projects) can be considered for addressing some of the unsettled issues inherited from the past.

Most old dam developments, such as Kariba Dam, which were implemented prior to the 1980s, did not have the benefit of thorough impact assessments, especially environmental and social impact assessment. As a result, these dams are likely to be associated with some negative social impacts. The current ZRA Trust Fund and ZESCO's Tonga rehabilitation projects, initiated 40 years after the construction of the dam, are promising ex-post reparation initiatives. Some of the representatives of the Tonga people who attended the stakeholder meeting are doubtful about the effectiveness of these programmes, which they consider as too top-down if not mere cosmetic interventions.

Projects of the nature of Kariba not only require regional co-operation, but are also opportunities for fostering it. As a large international and inter-country hydroelectric project, Kariba facilitated the creation of regional power pools and ensured reliability of power supply to the participating countries. In turn, regional power pools can reduce the unit cost of power through the optimisation of use and economies of scale.

As an international dam, shared by the two countries of Zimbabwe and Zambia, the Kariba Dam project automatically involved the erection of power transmission lines to the two countries, thereby connecting the national electricity grids. It also meant that Kariba became connected to any other power stations within the two countries. When Kariba was built, the Zambian Copperbelt was already connected to power stations in the Congo, and therefore, Zimbabwe became connected to the Congo. Kariba enabled the formation of the Southern African Power Pool in which the electricity grids of the countries of Zimbabwe, Zambia, Congo, South Africa, Mozambique, Botswana, and Lesotho became connected.

Regional power pools have many advantages, which include the maintenance of reliable power supplies in the event of any one station breaking down, the reduction in cost of power through the optimisation of use of the different power stations in the pool, and more importantly the savings in investment that accrue in foregoing the expansion of individual country generation capacity.