- Independent monitoring has found a high prevalence of deformities in fish in the Volta Grande do Xingu area of the Brazilian Amazon, following the construction of the massive Belo Monte dam.
- Potential factors could include changes in the river’s flood pulse, water pollution, higher water temperatures, and food scarcity, all linked to the reduced flow in this section of the Xingu since the dam began operating in 2016.
- Federal prosecutors are scrutinizing the dam’s impact, alongside independent researchers, and at the recent COP30 climate summit warned of “ecosystem collapse.”
- Both scientists and affected communities say the prescribed rate at which the dam operator is releasing water into the river is far too low to simulate its natural cycle, leaving the region’s flooded forests dry and exacerbating the effects of drought.
On Feb. 17, 2016, the gears ground into life at the massive Belo Monte hydroelectric plant, on the “Large Curve” of the Xingu River, or Volta Grande do Xingu, in the Brazilian Amazon. By April that year, the 11.2-gigawatt plant was already in commercial operation. That same year, researcher Jansen Zuanon visited Volta Grande do Xingu, the 130-kilometer (80-mile) stretch of this major Amazon tributary, whose course had been diverted and its flow reduced due to the operation of the hydropower plant.
“I was there as soon as the first turbines started operating. At that moment, the reservoir was still filling up,” recalls Jansen, who was accompanied by observers from the Independent Territorial Environmental Monitoring (MATI) team and prosecutors from the Federal Public Ministry (MPF). “We’d go on canoes and find the caparari fish [spotted sorubim, Pseudoplatystoma corruscans] on the banks, in the shallow waters. They were clearly malnourished, with sunken eyes, wounds, missing teeth, and full of parasites. They were like zombie fish, dying little by little.”
Today, almost 10 years after the start of operations, new adverse impacts from the Belo Monte dam continue to emerge. In 2025, the same monitoring group released a technical note describing visible physical changes in silver croakers (Plagioscion squamosissimus). The specimens they found had squat, oval and rounded bodies — very different from their normal elongated aspect, indicating spinal deformities.
Born and raised in the Xingu area, in the riverside village of Belo Monte, Sara Rodrigues Lima was one of the fisherwomen who found fish in these conditions and began documenting them in 2021. Since then, she has reported not only deformities and impaired growth but also the surprising prevalence of these problems. She estimates that four out of every 10 fish caught are affected, but the figures still need to be corroborated by further research. By comparison, laboratory experiments found approximately 15% of the larvae of tambaqui (Colossoma macropomum) showed deformities in a controlled environment considered ideal.
“They named this monster after our community,” Lima says of the dam, which now threatens fishing and food security in the two Indigenous territories of Paquiçamba and Arara da Volta Grande do Xingu, and dozens of riverside communities along the stretch of the Volta Grande.
What’s causing fish deformities?
Lima has become an active voice in the fight against the dam, operated by Norte Energia. Her photos and notes not only inform independent monitoring but have also led prosecutors from the MPF to file a lawsuit against the diversion of the Xingu’s flow to Belo Monte’s turbines. In 2025, during the COP30 climate summit in Belém, capital of the state of Pará, where the dam is also located, the prosecutors presented their view of ecosystem collapse to the public and the press, citing fish deformities as one of the indicators.
Jansen Zuanon, with the National Institute of Amazonian Research (INPA), points out how deterioration of aquatic habitats linked to the dam’s operation has triggered a series of changes that may be associated with such high occurrence of deformities.
The first hypothesis is nutritional deficiency due to the change in the river’s food chains. Another possible cause is the replacement of an environment characterized by rapids that are unique to the area with one full of sediment, with warm, stagnant waters and, therefore, less available oxygen, which would affect the development of fish in their larval stage. It’s also likely that the water retained in the dam’s reservoir is contaminated with toxic substances such as mercury, which is known to cause deformities.
Anna Paula Costa Scherer, a doctoral student in freshwater biology at Nilton Lins University, and affiliated with INPA, has been conducting laboratory experiments with tambaqui larvae in conditions that simulate extreme climate scenarios, including warmer waters and higher concentration of carbon dioxide in the atmosphere. She’s already observed spinal deformities related to these extreme conditions in these experiments.
Scherer says the kinds of deformities found in the Volta Grande do Xingu fish probably develops in younger forms of the animal, not in adults. “There are many factors that could trigger this,” she says. “From the speed of the water due to abrupt flow oscillations, to classic and well-known issues such as accumulation of decomposing organic material in the water. But the analyses conducted during environmental impact studies for the dam were not really considering these points.”
The fish collected with deformities were sent for laboratory analysis. As of the time this story was published, there was no information available that could rule out or confirm these hypotheses for the deformities. To date, there are also no studies quantifying the prevalence of deformities in the Volta Grande area.
Volta Grande do Xingu: A unique — and changed — place
The Belo Monte hydroelectric dam complex includes two main sites: the Pimental site, about 40 km (25 mi) downstream from the municipality of Altamira, where the main dam is located, and the Belo Monte site, where the main powerhouse is located. Between them, a 20-km (12-mi) diversion canal connects the main reservoir, which covers 359 square kilometers (139 square miles), to the intermediate one, covering 119 km2 (46 mi2), which, in turn, feeds the turbines.
The Xingu’s naturally large curve has given rise to ecosystems that are unique even for the Amazon. But since the construction of the dam, about 70% of the river water that used to flow into the Volta Grande has been diverted to this system of reservoirs and canals. As a result, the area has been given a technical, supposedly neutral name in official documents: the Reduced Flow Stretch (RFS).
Water in the Volta Grande flows over a rocky bed formed where the river crosses areas of ancient and hard rocks known as crystalline shields, which create rapids. Seasonal flooding, peaking between March and May, combines with the steep slope and numerous intertwined channels to create a mosaic of habitats that supports a unique range of fish species, including some that are threatened and highly dependent on these fast, intense-flowing waters.
The rhythm of the Amazon rivers — their cycle of rise, flood, ebb and drought, scientifically defined as flood pulses — works as “the heartbeats of the river that sustain life,” in the words of Josiel Juruna, head of the Independent Territorial Environmental Monitoring (MATI) team, created by Indigenous and riverine communities as a watchdog for the Belo Monte dam’s construction and operation.
To describe these rhythms, Renata Utsunomiya, a hydrologist with the University of São Paulo, cites the research she conducted with the Arara Indigenous people from the Xingu, describing how fish behavior is one of the ways to track this pulse.
“Silver croakers disappear from the deep parts of the river, signaling the low tide, just as the shrubs on rocky outcrops turn green before the river begins to rise,” she says, showing how life in the Volta Grande is attuned to the rhythm of the river.
These conditions also create habitats with adapted vegetation, in direct interaction with aquatic fauna. The sarobais are low-growing formations over rocks, where trees bloom in late November, when the wet season begins. Fruit falls in the into the river, providing a feast for the fish that begin to enter the shallow waters of the side channels to spawn and renew life, a period known as piracema.
“But that doesn’t happen anymore,” says Adriano Quaresma, a botanist with INPA who researches the Volta Grande’s floodplains. “With the change in flow, the water doesn’t come. It arrives two to three months late. These trees bloom, but their fruit falls on dry land.”
Quaresma is a co-author, along with Lima and Zuanon, among others, of a recent study demonstrating the loss of vegetation in local sarobais as well as the replacement of plant species in igapós, the “flooded forests of the Amazon, with others that only occur in dryland forests.
The so-called Consensus Hydrograph (CH) is the flow regime established for the Reduced Flow Stretch approved by Brazil’s National Water Agency (ANA) in 2011, with slight changes made in 2014. In practice, it works as a benchmark that establishes the minimum amount of water that should continue to flow through the Volta Grande each month. That’s the minimum average volume that the dam operator must release, ostensibly to simulate the seasonality of the river flow.
The Consensus Hydrograph establishes two regimes of minimum monthly flow for Volta Grande do Xingu that should alternate. Under regime A, the minimum prescribed flow rate in April, the peak of the flood season, is 4,000 cubic meters per second (141,300 cubic feet per second). Under regime B, this minimum rate doubles to 8,000 m3/s (282,500 ft3/s). So after a year with less water under regime A, the area would be compensated the following year by higher flow under regime B.
But data gathered by Utsunomiya and 42 other researchers from the independent monitoring team show that neither regime can actually simulate the natural flood pulse. Regime A has already been deemed unfeasible in a 2019 technical report by IBAMA, the federal environmental protection agency, and communities and researchers are now demanding the rejection of the entire hydrograph.
According to Utsunomiya, the very name of the hydrograph reveals a problem. “The first contradiction is that there was no consensus for establishing the Consensus Hydrograph since it was not built with participation, especially by those who are currently affected,” she says. “The criteria adopted focused on flood and drought peaks in the months of April and October, respectively. But this was arbitrary, as it did not consider the flooding of sarobais and igapós. Even under regime B, with the highest volume, the rates are still very low and the flood time is too short.”
In 2022, even under regime B, there was an 82% decrease in the flooded area of igapó forests, which, by definition, must be fully flooded.
In the dry season, the minimum flow rate was also below the historical average. A flow rate of 700 m3/s (24,700 ft3/s) was set for October, below the historical average of 1,129 m3/s (39,870 ft3/s) for the period before the dam was built. Even so, in 2023-2024, the flow fell below even this minimum due to periods of extreme drought. At the peak of the drought in October 2024, a flow rate of only 536 m3/s (18,929 ft3/s) was recorded.
In practice, this prolonged the period of extreme drought. Before the dam, there used to be about 44 days of drought per year. This more than doubled to 103.5 days after the construction. With such long periods of low and warm water, fish species become more vulnerable to heat and disease. In addition, the sarobais are also exposed for longer periods, leading to high vegetation mortality, as observed in satellite images.
The independent monitoring has effectively revealed that the water flow released to the Volta Grande by the dam is insufficient and cannot simulate the river’s natural flood pulse, leading to the environmental and social impacts observed. Utsunomiya recalls how, during her research, it was common to hear people say things like “now the river is under control” or “the river is crazy,” in response to the erratic flows. Alongside the reduced amount of water, there’s also an irregular flow with rapid rises and falls in the river level, described by communities as “flood-ebb.”
Much of the electricity generated by the dam isn’t even used in the Amazon region, but rather in Brazil’s southeastern population centers. The turbines are turned on or off in response to demand from the grid, and this cycling cycle creates sudden oscillations in water flows, such as those recorded between January and February 2025, when the water level rose by 1.67 m (5.48 ft) in just two days. The National Water Agency deemed this a violation on the part of the dam operator.
Monitoring created a rift at the University
In its 48th Socioenvironmental Monitoring Report, covering the first quarter of 2025, Norte Energia doesn’t mention fish deformities. However, a study published in 2023 and led by Luciano Fogaça de Assis Montag from the Federal University of Pará (UFPA) had already recorded deformities in silver croakers in Volta Grade do Xingu.
Montag also worked with other UFPA experts, in partnership with Norte Energia, to conduct studies on fishing and fish species in the dam-affected area. The data they generated was included in the reports that Norte Energia is required to submitted periodically to various authorities, such as IBAMA and, in the case of the socioenvironmental monitoring report, BNDES, the Brazilian Development Bank. The team was coordinated by Tommaso Giarrizzo and Victoria Judith Isaac Nahum, who are no longer with UFPA.
The fact that academic articles had recorded deformities in 2023 that were then omitted in the 2025 report indicates a discrepancy between what’s observed and recorded in the field by the official monitoring team and what’s actually reported to the authorities, observers say.
The MATI team, which has long criticized Norte Energia’s downplaying of the impacts, also includes UFPA researchers such as Juarez Pezzuti, a professor at the university’s Center for Advanced Amazonian Studies. A UFPA source who asked not to be named said cooperation with Norte Energia led to a virtually unbearable level of tension among the faculty, generating pressure on the university administration to discontinue the partnership. The contracts and agreements between the university and the operator related to monitoring of fishing and fish species ended in February 2025, and haven’t been renewed since.
A researcher who worked in the Norte Energia-UFPA partnership, and also asked not to be named, said the sampling designs established by IBAMA and executed by Norte Energia were unable to capture the kinds of impacts occurring in the Reduced Flow Stretch. Coordinators Giarrizzo and Nahum didn’t respond to Mongabay’s requests for clarification about these studies. Both researchers are bound by confidentiality agreements with Norte Energia for at least 20 years, which prosecutors have challenged.
Other researchers interviewed for this report criticized the results obtained by independent monitoring. In general, they say the technical notes lack experimental designs capable of establishing a causal relationship between the construction of the dams and phenomena such as fish deformities. According to them, experiments that isolate different factors — temperature, pollution, oxygenation, etc. — would have to be conducted to establish these relationships in a more conclusive way.
In response to these criticisms, MATI’s Pezzuti defined the methodologies used to monitor Norte Energia as “a set of efficient strategies that work well to cover up the [company’s] growing impacts on fishing and fish consumption.” For him, “the causal nexus fallacy is a common argument always put forward by businesses that drive environmental damage. This is a strategy used by companies causing great damage to seek the benefit of the doubt. But what is observed by MATI is largely predicted.”
The future of Volta Grande do Xingu
Josiel Juruna, the MATI head, says he knows there’s no prospect of Belo Monte being removed from the Xingu River. “But at least we have been fighting to replace the hydrograph with one that follows the life pulse of the rivers,” he says.
His team has proposed a new regime, called the Piracema Hydrograph. It prescribes a peak flow in April of 14,000 m3/s (494,400 ft3/s), which is the threshold needed to flood the igapós, thus reestablishing synchronicity with reproduction. Compared to the historical flow, from 1971-2014, the Piracema Hydrograph would result in a maximum flow decrease of 30% during the critical months of December to May. And while far below pre-dam levels, it’s much less drastic than the current situation.
For Utsunomiya, this change is also crucial in the face of climate change scenarios, where extreme events such as the 2023-2024 droughts look likely to recur more frequently and with greater intensity, making the hydropower plant impossible to operate under its current model.
“Therefore, that period demonstrated that the Consensus Hydrograph cannot guarantee an ecological flow, that is, it’s impossible to guarantee life in that stretch of the river,” she says.
Norte Energia’s response
In response to Mongabay, Norte Energia says “the project was designed to generate a large amount of energy during the Amazonian winter, from December to May, and follow and respect the seasonality of the river in the remaining months of the year.” The company denies that the dam has created drought in the Xingu River, and that the phenomenon “occurs naturally.”
Norte Energia also notes that, between 1931 and 2007, there were 17 occurrences of average monthly flow below 700 m3/s, with the lowest rate ever recorded in the Xingu River being 380 m3/s (13,400 ft3/s), in October 1969. “Studies show, therefore, that the river sees large natural variation in flow throughout the year,” it says.
In response to the reports of a sudden change in the flow of the Xingu, the company says it abides by “the operation authorization, which determines the flow change rates that were set to maintain the ecological processes of Volta Grande do Xingu.”
Norte Energia also denies any environmental changes linked to the dam that could have caused fish deformities, adding that it conducts “ongoing monitoring of environmental and fish quality in Belo Monte’s catchment area.”
As for the confidentiality agreements with the UFPA researchers, Norte Energia says they don’t prohibit interviews. (Read Norte Energia’s full response to Mongabay here.)
Banner image: The Belo Monte hydropower dam on the Xingu River in Brazil’s Pará state. Image courtesy of TV Brasil.
This story was first published here in Portuguese on Dec. 17, 2025.
Citations:
Lopes, I. G., Araújo-Dairiki, T. B., Kojima, J. T., Val, A. L., & Portella, M. C. (2018). Predicted 2100 climate scenarios affects growth and skeletal development of tambaqui (Colossoma macropomum) larvae. Ecology and Evolution, 8(20), 10039-10048. doi:10.1002/ece3.4429
Quaresma, A., Zuquim, G., Demarchi, L. O., Ribas, C. C., Wittmann, F., Assunção, A. M., … Cruz e Silva, R. (2025). Belo Monte Dam impacts: Protagonism of local people in research and monitoring reveals ecosystem service decay in Amazonian flooded vegetation. Perspectives in Ecology and Conservation, 23(1), 39-50. doi:10.1016/j.pecon.2025.02.001
De Assis Montag, L. F., Peixoto, L. A. W., Seabra, L. B., Barbosa Gonçalves, L. A., Cardoso Lobato, C. M., Barreira Mendonça, M., … da Silva Freitas, T. M. (2023). First record of spinal deformity in the South American silver croaker Plagioscion squamosissimus (Eupercaria: Sciaenidae) in the Xingu River, Brazil. Fishes, 8(7), 363. doi:10.3390/fishes8070363
Credits
