Balbina Dam Was Supposed to Deliver Clean Energy to Manaus, But Its Flooded Forest Has Made It a Major Emitter

Commissioned in 1989 just northeast of Manaus, the Balbina dam remains a classic case study for researchers examining tropical hydroelectric power. Its shallow reservoir flooded a vast area of the Amazonian forest for a modest output. On-site measurements reveal how much submerged vegetation weighs on its climate accounting.

A 2 360 km² reservoir for an average production of about 112 megawatts, despite 250 MW installed

Built on the Uatumã River, the facility has an installed capacity of 250 MW. However, according to the work of Philip Fearnside, an ecologist at INPA, the Brazilian public institute dedicated to Amazonian research, its average output tops out around 112 MW, with roughly 109 MW actually reaching Manaus after transmission losses. The causes are a very flat relief and a relatively small watershed.

Thus, to achieve this low production, the normal operating level submerges nearly 2 360 km² of land. This ratio yields less than 0.05 watt per square meter flooded, a power density among the lowest of major dams. Moreover, a large portion of the forest had not been cleared before impoundment, leaving a substantial amount of methane potential beneath the surface.

Oxygen-deprived at the lake bottom, submerged vegetation decomposes and feeds methane emissions

At the bottom of the reservoir, the stagnant water becomes stratified and oxygen disappears. There, bacteria that break down the submerged trunks, leaves and soils produce methane, a greenhouse gas with a warming power well above CO₂ over a few decades. In parallel, part of the organic matter is transformed into carbon dioxide, which escapes to the surface by diffusion.

As early as 1995, Philip Fearnside estimated in the journal Environmental Conservation that Balbina’s balance exceeded that of a comparable fossil-fuel thermal production. Later, the team led by Alexandre Kemenes measured in 2007 a significant methane degassing downstream of the dam. These results undermined the idea of hydroelectricity as always carbon-neutral.

According to Kemenes and colleagues, Balbina emitted about 3 million tonnes of carbon per year in 2005

In 2011, Alexandre Kemenes, Bruce Forsberg and John Melack published in the Journal of Geophysical Research: Biogeosciences a detailed accounting of CO₂ emissions for 2005. The reservoir released around 2 450 gigagrams of carbon, against 81 downstream, for a total of about 2 530 gigagrams. Most notably, 51 % of the downstream emissions came from degassing at the turbine discharge.

Adding the CO₂ equivalent of the methane estimated previously, the authors arrive at roughly 3 teragrams of carbon per year. This volume corresponded to nearly half of the fossil emissions of the city of São Paulo. However, these figures describe a single year, because emissions from a tropical reservoir typically decline as the submerged biomass is exhausted.

Balbina does not summarize all hydroelectricity, but it illustrates the weight of shallow tropical reservoirs

Some publications link Balbina to studies by the University of São Paulo on Belo Monte, another dam on the Xingu and commissioned in 2016. Yet these are two distinct projects. Moreover, the reference studies on Balbina come mainly from INPA and its partners.

Consequently, Balbina’s case does not condemn hydroelectricity as a whole. Dams in mountainous regions or with narrow reservoirs emit considerably less per kilowatt-hour produced. Nevertheless, the example shows that the flooded surface and the amount of forest submerged should weigh heavily in evaluating future Amazonian projects.