Belo Monte reservoir has tripled local greenhouse gas emissions

Monitoring since 2011 shows that emission levels vary significantly from one part of the complex to another, making the standards used in environmental impact assessments and the carbon credit market unreliable.

By Maria Fernanda Ziegler  |  Agência FAPESP – Greenhouse gas emissions have tripled from the site of the Belo Monte hydropower development and dam on the Xingu River in Altamira (Pará state, Brazil), according to an article published in Science Advances by an international group of researchers.

Emissions were measured at different points of the site before, during and after construction. The study took about ten years to reach completion.

“The rationale for building Belo Monte was based on the assumption that hydro produces low emissions and is cheaper than generating electricity from other renewables. This argument has now been refuted, as our study shows,” said Dailson Bertassoli Jr., a researcher at the University of São Paulo’s Institute of Geosciences (IGc-USP) and first author of the article.

Some hydropower plants in the Amazon emit more than others, he added. For example, Balbina began operating in the 1980s and emits more than thermal plants producing an equivalent amount of power. “In the case of Belo Monte, the largest power plant in the Amazon, emissions range from 15 kg to 55 kg of CO₂ equivalent per megawatt-hour [varying due to seasonal factors]. That’s a fraction of the emissions from a coal- or oil-fired thermal plant, but even so, it’s far from negligible,” he said.

Belo Monte is the world’s largest run-of-river power plant in terms of installed capacity. Run-of-river plants harvest energy from flowing water instead of a massive dam and huge reservoir. Belo Monte has two reservoirs linked by a 20 km canal. The main reservoir is on the Xingu and has an area of 359 sq. km. The other has 119 sq. km. and is surrounded by 28 dykes and two diversion canals.

The researchers began measuring greenhouse gas emissions at the site in 2011, before the river was dammed or construction of the plant began. Over the years, three other surveys have been conducted at the same points. Scientists from Linköping University (Sweden), the University of Washington in the United States and the Federal University of Pará (UFPA) in Altamira took part. The research was supported by FAPESP via nine projects (2014/21564-2, 2015/09187-1, 2016/11141-2, 2016/02656-9, 2019/24977-0, 2018/15123-4, 2019/24349-9, 2011/14502-2, and 2018/18491-4).

Uneven emission levels

An important finding of the research is the unevenness of greenhouse gas emission levels across the Amazon, and even at different points of the same reservoir. Measurements taken in different parts of the Belo Monte complex, including areas that were flooded during construction, showed higher levels of emissions in certain areas than in others. As a result, standards such as those recommended by the Intergovernmental Panel on Climate Change (IPCC) and used in prior impact assessments as well as for sales of carbon credits are unreliable here, as they are based on hydropower systems in temperate zones.

“Understanding the mechanisms behind greenhouse gas emissions from the Belo Monte reservoirs is essential to efficient strategic planning of the expansion of Brazil’s electricity generating system. However, they’re so uneven that referring to ‘hydroelectric power plant emissions’ as a generalization is impossible. Each plant has its emissions, and these should be assessed on a case-by-case basis,” Bertassoli Jr. told Agência FAPESP, adding that one of the main sources of greenhouse gas emissions from hydropower complexes is methane (CH₄) from decomposing organic matter in sediment on the bottom of the reservoir.

The first set of measurements was performed by Henrique Sawakuchi, currently affiliated with Linköping University and second author of the article. According to him, emissions from different dams and hydro plants vary widely owing to factors such as vegetation, soil, temperature, climate, and microbial activity, as well as the type of reservoir.

“In contrast with temperate and boreal zones, tropical zones have high temperatures throughout the year and this results in intense microbial activity, with high levels of methane and carbon gas production,” Sawakuchi explained.

Standard deviation

The organic matter that emits methane comes from the upstream drainage basin and accumulates in the reservoir. The gas may also be produced by algae in the water column.

“These two factors vary greatly among rivers in the Amazon. The Xingu, for example, is a clearwater river, with plenty of sunlight and an abundance of algae, which eventually die and emit methane, whereas the Madeira is a whitewater river [actually brown, with a heavy sediment load], little sunlight penetrating the water column, and hence much less algae,” said André Sawakuchi, a professor at the Institute of Geosciences (IGc-USP) and last author of the article.

Reservoir depth also influences the variability of emissions. “In deeper zones, bottom water may be less oxygenated, favoring methane generation,” he explained.

The researchers also observed that depending on the type of soil and vegetation flooded during dam construction, more organic matter may be available for biodegradation and methane and carbon production. The study found that emissions from flooded pasture were surprisingly high. Pasture was not removed before the reservoir filled, whereas forest cover was.

“The topsoil also has a great deal of organic matter that gives rise to methane emissions. These kinds of variation make all the difference when it comes to adding up total emissions from a reservoir,” André Sawakuchi said. “Sometimes you find pasture located in soil with plenty of organic matter and forest areas with very poor soil. It varies a great deal, and all this has to be taken into account when the impact of a hydro development is assessed.”

According to the researchers, the results reinforce the importance of considering all variables that cause emission heterogeneity in impact assessments for hydro developments, including run-of-river plants like Belo Monte.

“It should be mandatory to monitor emissions before, during and for a long time after construction of dam reservoirs, especially in the case of those on the Madeira, which are selling carbon credits without properly estimating emissions. The carbon credits they’re selling are probably not mitigating,” Henrique Sawakuchi said.

The article “How green can Amazon hydropower be? Net carbon emission from the largest hydropower plant in Amazonia” (doi: 10.1126/sciadv.abe1470) by Dailson J. Bertassoli Jr., Henrique O. Sawakuchi, Kleiton R. de Araújo, Marcelo G. P. de Camargo, Victor A. T. Alem, Tatiana S. Pereira, Alex V. Krusche, David Bastviken, Jeffrey E. Richey and André O. Sawakuchi is at: https://advances.sciencemag.org/content/7/26/eabe1470.