Written by Steffen Dalsgaard 

The majority of Brazil’s electricity is generated by hydropower. Taking a look at data sources such as Electricity Mapstells us on a given day in July 2024 that the electricity generated in North Brazil – the home to two of the world’s largest hydroelectric dams, the Tucurui and the Belo Monte – is 86% renewable and 86% from low-carbon sources. Hydropower thus seem to be a great technology for providing the electricity needed for a green transition and for the decoupling of economic growth from climate impacts. However, the devil is in the detail.

It is easy to assume that hydropower is carbon neutral once the dam is built, and this has indeed been demonstrated to be a common belief (Kaldellis et al 2013). It is, after all, only water flowing through a turbine on its own. In July 2024 a colleague and I visited Altamira, the city closest to the Belo Monte dam, for a short stint of ethnographic fieldwork. I must admit that until then I also harbored the same belief that hydropower – notwithstanding the forced relocation of human and non-human populations by the building of a dam and of course the climate costs of constructing the dam itself – at least had a fairly low climate impact once it was operational.

Yet this is often not the case. Some dams are ‘run of the river’ or ‘diversion’ dams, but many others – including the two Brazilian dams in question – rely upon having reservoirs – flooded areas – needed to store water which can be released and generate electricity when needed. In these reservoirs, emissions of greenhouse gasses – carbon dioxide, methane and nitrous oxide – come from the especially anaerobic decomposition of the plant material that was previously growing in the flooded area (see Song et al 2018). A 2016 study estimated that reservoirs were responsible for 1.3 per cent of all greenhouse gases produced by humans. A Norwegian study added that this can be so much as estimated emissions from hydropower being up to 10 percent higher per unit of electricity than emissions from gas-fired electricity plants. A study of a specific dam in French Guiana (Petit Saut) stated that this was 19 times higher than natural gas-based electricity (Fearnside 2015). The emissions are particularly bad for dams and reservoirs in the tropics, where plant material decomposes faster than in boreal or temperate climates and releases methane (e.g. Song et al 2018).

In Brazil it has been known for decades how hydropower dams are thus fairly large emitters of greenhouse gases (e.g. Fearnside 1995; Gagnon and van de Vate 1997). It nonetheless still often avoids public attention in the Global North and thus mainstream scientific circles. The 2022 IPCC report on climate mitigation is no exception. It completely fails to mention the climate impact of hydropower dams in their section on hydroelectric power. Luckily, the 2014 IPCC report – apparently still the most authoritative source despite being 10+ years old by now – lists the ‘indirect’ methane emissions related to operations as around 88g of CO2e/kwh generated. This comes from the acceleration of the water flow through turbines and barrages which in itself releases methane (but possible also carbon dioxide) already existing naturally in the water. These emissions may not seem to be a lot compared to many other sources of emissions. The 2014 IPCC report writes that the lifecycle emissions of dams on the other hand tell of a vast range stretching from 1g CO2e/kwh to 2200g CO2e/kwh with a median of 24g CO2e/kwh. The maximum is more than twice the maximum of coal-firing plants. This range likely depends on the location of the dam, how it was built, how efficiently the dam is using its capacity, and – here the IPCC report is not entirely clear – how much the submerged biomass in the dam’s reservoir decomposes. Here, a 2018 research paper states that:

“reservoir-based dams located in boreal and temperate regions have much lower reservoir emissions (3-70 g CO2 eq./kWh) compared with dams located in tropical regions (8-6,647 g CO2 eq./kWh). Our analysis shows that although most hydroelectric dams have comparable GHG emissions to other types of renewable energy (e.g., solar, wind energy), electricity produced from tropical reservoir-based dams could potentially have a higher emission rate than fossil-based electricity.” (Song et al 2018).

So where are the biggest Brazilian dams – the Tucurui and the Belo Monte – in these calculations? Most likely at the high end. Both are reservoir dams, and when it comes to Belo Monte, the company behind the construction of the dam could have logged the reservoir area beforehand to limit the methane emissions, but failed to do so.

Furthermore, the construction of the Belo Monte dam in particular turned out to be a social and environmental disaster according to both researchers, NGOs and local communities, riverine as well as indigenous. The process of construction itself was fraught among other things demonstrating a lack of consultation with affected people, transparency of decision-making processes (Jaichand and Sampaio 2013), and misguided and inadequate mitigation and compensation efforts. When critics pointed out that the building of the dam was in violation of the Brazilian Constitution which protects indigenous people like the ones affected, the protest was overruled with reference to ‘national interest’ including security (e.g. Atkins 2019). Evidence of bribes paid were later uncovered by Brazilian media. To the inhabitants of the area it was a complete change of lifestyle for an estimated 20 000 people, who were forcibly removed from their homes and land. It changed how they earned a living, and even what they could eat with an increase in the consumption of processed foods replacing fish from the river and vegetables from their gardens. Altamira in turn became the most violent city in Brazilwhen its population grew with 50% from around 100 000 to more than 150 000 in just three years.

The owner of the Belo Monte project prides itself in having generated lots of jobs (although temporary) and permanent infrastructure in the vicinity of the dam, and they claim to have set aside 3.7 billion Reais for mitigation and compensation activities (approximately 675 million USD). Those forcibly removed have been given new houses (far from the river, in areas where there is nothing for them to do). Others have received everyday amenities such as boats with outboard motors, TVs or even cars. A representative of a local NGO told us that some ‘social’ projects had given chickens to hunting and fishing people, who had never tried rearing animals in their lives. To compensate for some of the biomass destroyed in the process of building the dam, another project (‘Amigos da APP’) has set up small areas around the city of Altamira meant to be left undisturbed until the trees are bigger. Yet these areas are miniscule compared to the territory of the reservoir and to this ethnographer’s eyes resembled small parks rather than a serious attempt at carbon storage. To add insult to injury, the Belo Monte dam does not even deliver the promised electricity – it has produced less than 3% of the promised capacity at one stage, and with climate change generating more extreme weather patterns, the flow of water in the rivers will become more unpredictable. Longer dry spells will mean less water for electricity. In fact, there was a shortage announced on public television while we were in Altamira with the TV station consulting experts who gave advice to poor people on how they could turn of electric devices – including their aircon – in a period where degrees were soaring. The most vulnerable were thus nudged to make the biggest sacrifices, where – for example – rationing could have been a more socially equitable strategy.

Everything written above has been covered extensively by others – NGOs, researchers and documentarists. What is of concern for me here is the question of how Belo Monte and possibly dams more generally speak to ideas of ‘decoupling’, and how attempts at decoupling play out in a Brazilian context. During our fieldwork, we looked for ideals but also examples of technologically supported decoupling. That is, a detachment of growth rates from rates of climate impacts. Decoupling is normally deployed in macroeconomic discourse and assessed against large datasets of economic and climatic indicators. My usage of it is a bit different since I am interested in how decoupling could work as an anthropological term, where the relation being studied is that between on the one hand the forms of value that people work to increase or ‘grow’, and on the other hand the social or environmental costs that this increase in value brings along. The value forms here may not merely be economic, but they can of course be so. The Brazilian hydroelectric dams may appear to contribute to a successful decoupling in the general economic sense, because they create growth (albeit through destruction), and they can be configured in accounting terms to appear carbon neutral if one takes the IPCC reports as the generic benchmark rather than looking at the specific dams. Yet the dams in Brazil are hardly examples of a successful decoupling when one looks at their externalities – the damage they produce in order to appear successful. This damage includes their embedded emissions, and the heavy destruction of livelihoods that come with building them, at least in Brazil. The electricity generated by the Belo Monte Dam has been a disaster for local communites and for the environment in many ways. But what discounts a dam like Belo Monte as being an example of decoupling is that it has failed to generate the amount of electricity promised and that any claims to carbon neutrality of the electricity falls short when emissions from the dammed reservoir are taken into account.

 

References

Atkins, E. 2019. Disputing the ‘National Interest’: The Depoliticization and Repoliticization of the Belo Monte Dam, Brazil. Water 11, 103. doi:10.3390/w11010103

Fearnside, P.M. 1995. Hydroelectric Dams in the Brazilian Amazon as Sources of ‘Greenhouse’ Gases. Environmental Conservation 22, 1: 7-19.

Fearnside P.M. 2015. Emissions from Tropical Hydropower and the IPCC. Environmental Science &Policy 50: 225-39.

Gagnon, L. and J.F. van de Vate 1997. Greenhouse Gas Emissions from Hydropower: The State of Research in 1996. Energy Policy 25, 1: 7-13.

Jaichand, V. and A.A. Sampaio 2013. Dam and Be Damned: The Adverse Impacts of Belo Monte on Indigenous Peoples in Brazil. Human Rights Quarterly 35: 408-447.

Kaldellis, J.K., M. Kapsali, E. Kaldelli, E. Katsanou 2013. Comparing Recent Views of Public Attitude on Wind Energy, Photovoltaic and Small Hydro Applications. Renewable Energy 52: 197-208.

Song, C., K. Gardner, S. Klein, S.P. Souza and W. Mo 2018. Cradle-to-Grave Greenhouse Gas Emissions from Dams in the United States of America. Renewable and Sustainable Energy Reviews 90: 945-956.