Ethiopia: Concerns about Gibe 3 Dam

BACKGROUND

The River Omo (also known as Gibe in its upper and middle basins) rises in the well-watered western highlands of Ethiopia at about the same latitude as Addis Ababa. It flows for nearly 1000km. and drops 1,600m from its source to its end point in Lake Turkana, the world’s largest desert lake, which lies wholly within Kenya. The climatic and topographic features of the Omo-Gibe basin give it a hydropower potential second only to that of the Blue Nile Basin, which accounts for half the hydropower potential of the entire country (Kloos and Legesse, 2010, p. 77). So far, however, only one dam has been completed along the Omo-Gibe. Known as Gilgil (‘Lesser’) Gibe 1, this began operating in 2004 and is currently Ethiopia’s single largest supplier of electricity. In 2010 a power plant known as Gibe II was opened further downstream. This does not have its own dam but draws water through a 26km. tunnel from the Gibe I reservoir.

Gibe I and Gibe II will be dwarfed in every sense by Gibe III. At 240m high it will be the tallest dam in Africa. With an ‘installed capacity’ of 1,870 megawatts (meaning its turbines will produce electricity at this level continuously), it will double Ethiopia’s current generating capacity which will then greatly exceed domestic demand. It is therefore planned to export up to 50 percent of the electricity generated by Gibe III to neighbouring countries. Two more hydropower dams will eventually complete the Omo-Gibe ‘cascade’.

Gibe III represents a huge financial investment (1.7 billion US) and an impressive feat of civil engineering which it is hoped will bring great economic benefits to Ethiopia. But it will have a potentially devastating impact on the downstream population by regulating the highly seasonal flow of the Omo, thereby ending the annual flood. This will directly affect all residents of the Omo flood plain and delta – around 100,000 people – who depend on the flood for their agricultural and pastoral activities (SOGREAH, 2010: 37). The ending of the flood will also indirectly displace many of these people from their existing farmland and grazing areas, by making possible the development of large-scale commercial irrigation schemes which are planned to occupy over 200,000 hectares of the Lower Omo. Since the Omo supplies 90 per cent of the water entering Lake Turkana, irrigation on this scale will significantly reduce the level of the Lake and increase its salinity. This in turn will adversely affect the livelihoods of another 300,000 or so people who live in northern Kenya and who depend on the lake for pastoralism and fishing (Johnston, 2010).

Any dam-building project that displaces a large number of people and/or restricts their access to vital resources would normally be expected to include a comprehensive plan to improve or at least maintain their long-term economic and social wellbeing. The downstream ‘mitigation plan’ so far proposed by the Gibe III project falls very far short of this expectation. It seems to be assumed by the project managers that those who are displaced from their land, livelihoods and resources will automatically benefit from generalised economic development and ‘modern’ forms of agriculture. This assumption is flatly contradicted by empirical evidence from other cases, to be found in countless academic studies and ex-post assessment reports by development agencies. These show that projects which displace people or restrict their access to vital resources and do not include comprehensive and fully budgeted livelihood reconstruction and development plans will result – at best - in the increased impoverishment of the affected population.

Not much time is left to avoid such an outcome in the lower Omo. The filling of the dam reservoir is expected to begin in June 2012 and the first of its ten turbines to begin operating in September 2013. Work should therefore begin as soon as possible on a livelihood reconstruction and development plan for the downstream population, designed to ensure that those who will carry the main burden of this project, on behalf of the nation at large, will be amongst the first to benefit from it. I want to explain in this paper why such a plan is needed and suggest specific steps that could be taken now towards achieving it.

DIRECT IMPACT OF THE DAM: ELIMINATION OF THE ANNUAL FLOOD

All the groups living along the lower Omo (Bodi, Mursi, Kwegu, Muguji, Kara, Nyangatom and Daasanach) depend on ‘flood retreat’ or ‘recession’ agriculture. Some, such as the Daasanach and Kara, are able to produce all the grain they need from flood cultivation while others, such as the Bodi and Mursi, must supplement it with rain-fed cultivation. But none could survive without the contribution made by flood cultivation to the household economy. The Daasanach, who live in the more arid southern part of the lower basin, also depend on the flood for the annual rejuvenation of their dry-season pastures.

These facts were not recognised by the Gibe III project until two years after dam construction had begun, when a revised Environmental and Social Impact Assessment (ESIA) was completed (CESI, 2009), together with an Additional study of downstream impacts (Agriconsulting S.p.A., 2009). The solution proposed was an artificial or ‘controlled’ flood, to be released annually over a ten-day period and timed to coincide with the natural flood. This, it was claimed, would ‘mitigate.....all adverse effects’ on the livelihood systems of the downstream population. But it later emerged that the controlled flood would in fact be a temporary measure, to be withdrawn ‘when deemed appropriate’. This information was contained (one might say the cat was let out of the bag) in a press release issued by the dam builder, Salini Costruttori, in March 2010, and has not, to the best of my knowledge, been officially confirmed by the Gibe 3 project office.

The effectiveness of the proposed controlled flood was, in any case, seriously questioned in an independent review (SOGREAH, 2010) of the Gibe 3 documentation, commissioned by the European Investment Bank (EIB). This concluded that the controlled flood had been planned without an adequate study either of the problem it was intended to solve or of its likely effectiveness. The review concluded that information needed to design an effective downstream mitigation plan was ‘still dramatically missing’ (SOGREAH, 2010: 120). It recommended that a number of further studies be carried out to make good these deficiencies, and that a detailed livelihood development plan should be prepared. Soon after the review was completed, the Industrial and Commercial Bank of China offered a 450,000 USD loan to the project, which prompted the EIB to cease giving any further consideration to making a loan itself. The additional studies recommended in the SOGREAH review were therefore not proceeded with.

The lack of a convincing livelihoods reconstruction and development plan for the downstream population will ring alarm bells for anyone familiar with the extensive literature on the human and environmental costs of large dams. This literature provides overwhelming evidence that the poorest, most vulnerable and most marginalized members of a country’s population, particularly ethnic minorities, are disproportionately affected by large dams and that most of those affected become even poorer, more vulnerable and more marginalized as a result. On present evidence, it is impossible to think that the Gibe 3 project will have a different outcome for the people of the lower Omo.

INDIRECT IMPACT OF THE DAM: LARGE-SCALE IRRIGATED PLANTATIONS

On 25 January 2011 the Ethiopian Prime Minister, Meles Zenawi, made a speech in Jinka, the capital of South Omo Zone, in which he announced that the government was about to establish sugar cane plantations in the Lower Omo, covering an area of 150,000 ha. (See details here). Since at least another 70,000 ha had already been leased to private investors in the southern half of the lower basin, the total area now earmarked for large-scale commercial agriculture in the Lower Omo has reached well over 200,000 ha. The Prime Minister also announced that local agro-pastoralists, who will lose all their best agricultural land to the plantations, would be given an unspecified amount of irrigable land for their own use. It has since become clear that what is intended is the forced ‘villagization’ of this supposedly ‘backward’ population in permanent resettlement sites, to make way for the take over of their land, without compensation, by government and private investors. The number of prospective re-settlers was not mentioned in the speech but, according to the 2007 census, the total agro-pastoral population of the South Omo Zone (Bodi, Mursi, Nyangatom and Daasanach) amounts to just under 90,000 individuals (FDRE, 2008, Table 5: 98-99), which is almost certainly an under-estimate.

In making his announcement, the Prime Minister referred to ‘the good results’ achieved by large-scale irrigation in the Awash Valley of Eastern Ethiopia. But these results have been far from good for the local Karrayyu and Afar pastoralists. Studies of the impact of commercial plantations on local people and the environment in the Awash valley have shown beyond any doubt that the majority of the local population has become poorer and more vulnerable to food insecurity as a result. (Kloos, 1982; Ayelew and Getachew, 2009; Kloos et al. 2010). The same studies make it clear that this could have been avoided if two conditions had been met. First, detailed feasibility studies and socio-economic impact assessments should have been completed and made publicly available, before the plans were finalised. Second, local people should have been genuinely involved in the planning process, so that their needs and interests could have been fully and systematically addressed, from the start. These lessons have clearly not been learnt by those setting up sugar plantations in the lower Omo. Planning has been entirely top-down and implementation has been surrounded by a wall of secrecy. According to reports coming from the area, strenuous efforts have been made by local administrators to stifle criticism and neutralise opposition from local people. It is also reported that these efforts have included the heavy-handed use of direct physical intimidation by police and the military. It is difficult to see, in these circumstances, how irrigation development in the Omo Valley will have any less disastrous consequences for local people and the environment than it has had in the Awash Valley.

Large-scale irrigation in the Lower Omo will also affect the 300,000 or so people who live around Lake Turkana, in northern Kenya, and depend on its waters for pastoralism and fisheries. No attempt was made in the ESIA to calculate the impact of potential irrigation schemes on the lake level but a calculation was made in an independent report commissioned by the African Development Bank (Avery, 2010). Data presented in this report suggest that an irrigated area of 150,000 ha in the lower basin could lead to a drop in lake level of twenty metres (3-14), leading to more than a halving of the lake’s volume and more than a doubling of its salinity level (2-50). Such a drastic drop in lake level and volume could lead to violent trans-boundary conflict between neighbouring groups, such as Daasanach and Turkana, because of increased competition for vital subsistence resources (Johnston, 2010). Nor is it difficult to imagine an interminable dispute developing between Ethiopia and Kenya over the equitable use of the water resources of the Omo Basin, as the impact of irrigation schemes on the level of Lake Turkana becomes increasingly apparent.

WHAT CAN BE DONE?

It is far from unusual for the most well-intentioned scheme to improve the human condition to end in tragic consequences for those whom it was intended to benefit. As James C. Scott (1998) has shown, this is particularly likely to happen when an autocratic regime is intent on pushing through its chosen policy ‘for the good of the nation’, using all the power at its disposal and in the absence of a functioning civil society. Given that these are the conditions that prevail in Ethiopia today, the question ‘What can be done?’ may appear ‘academic’ in the extreme. It nevertheless seems worth suggesting what could be done, even at this late stage and given the necessary political will, to ensure that the dam and the irrigation schemes bring real and sustainable economic benefits to the local population.

First, the additional investigations recommended by the SOGREAH review should be completed as soon as possible. These studies would provide the data needed to design a sustainable livelihood reconstruction and development programme for the downstream population.

Second, the plan to establish commercial sugar plantations in the Lower Omo, which is already being implemented, should be put on hold while a livelihood reconstruction and development programme is being designed. This will allow detailed land capability studies and socio-economic impact assessments to be completed and made available for public discussion and consultation. Above all, it is vital that affected communities are genuinely consulted - not simply informed of decisions that have already been made by the planners.

Third, the wellknown public health risks associated with large-scale irrigation schemes should be fully investigated and, again, publicly discussed with the affected population. These include the increased transmission potential of vector-borne diseases (especially malaria which is already endemic in the Lower Omo), the spread of disease agents (such as HIV) through the influx of large numbers of migrant workers, and the contamination of ground and surface water by factory emissions, fertilizers and crop protection chemicals.

The argument for ensuring that river-basin development brings real development opportunities for local people is fundamentally an ethical one: these are the people who must carry the main burden of the project, on behalf of the nation at large. But there is also a pragmatic argument. Projects which ignore standard safeguards intended to protect the environment and the interests of vulnerable local people risk not only impoverishing these people still further, but also creating irreversible and unnecessary environmental damage and tarnishing the public image and international standing of the state in question. Unless something is done to avoid these outcomes in the Lower Omo it is likely that Gibe III will come to be seen as a textbook example of how not to do river-basin development. This will not make it any easier for Ethiopia to achieve its ambitious energy generation goals, which depend so heavily on hydropower.

BROUGHT TO YOU BY PAMBAZUKA NEWS

* David Turton is a senior research fellow at African Studies Centre, University of Oxford.
* Please send comments to editor[at]pambazuka[dot]org or comment online at Pambazuka News.