“When I first saw this vegetation, I thought I was on another planet. It was all so different. … Not just the architecture of the plants, but the entire place. It seemed like a nanoforest.” Geraldo Fernandes was hooked by the spectacular landscape of rupestrian grasslands in 1980 during his studies to become a biologist. Since then, he has been working within the ecosystem, which is largely unknown to Brazilians. A pioneer of rupestrian grassland studies, Fernandes is a professor of ecology at Minas Gerais Federal University and a member of the Brazilian Academy of Sciences.

His interest in insects led him to investigate the interactions between the flora and fauna here. His first field study was on cancer in plants. Just like humans and animals, plants develop cancerous cells due to causes like high temperatures, lack of water or nutrients, pollution and also a variety of organisms — mostly insects like flies and wasps — that genetically manipulate the plant so it will produce the cancer upon which they feed.

These tumors, called galls, don’t all kill the plant. Some are even beautiful, looking like ornaments or flowers, but “inevitably, 90% of the plants found in rupestrian grasslands have some sort of cancer or other,” Fernandes says. This is due to the fact that most of them grow at the tops of mountains; the strong UV exposure and lack of resources create stress for these plants and, with their defense mechanisms debilitated, they are more susceptible to developing tumors. Over thousands of years, however, this vegetation has evolved to adapt to stress factors, managing to live on despite the adversities.

Covering just 0.8% of Brazil’s territory, rupestrian grasslands occur in 490 places, most of which are located in the states of Minas Gerais and Pará. Others can be found in isolated mountain ranges in Goiás, Mato Grosso and Roraima, among others. They total some 83,000 square kilometers (32,000 square miles, about the size of Austria) of landscape.

Characterized by a rich variety of highly endemic herbaceous species and unique compositions of these species, these grasslands are home to 15% of Brazil’s flora (5,700 plant species have been catalogued). Of these species, 40% are endemic, meaning they occur only inside this ecosystem. This is one of the highest endemic rates found on the planet. In the Espinhaço Mountains, home to most of Brazil’s rupestrian grasslands, some plant families are as much as 80% or 90% endemic. Even after years of study, new species are discovered all the time: Between 2005 and 2014, 118 new plant species and 26 vertebrate species were discovered, including 11 frogs, eight lizards, four birds, two snakes and one mammal.

The Espinhaço Range is Brazil’s largest and oldest. It is 1.8 billion years old and runs from the state of Minas Gerais to Bahia. Its soils are composed of quartzite and arenite, the remains of ancient ocean floors and deserts. The southernmost part of the range lies in the so-called Iron-Aquifer Quadrangle and is home to layers of iron-rich crust called cangas, which can also be found in some hills in the Carajás region of Pará state. The cangas were formed from old sedimentary rivers or lake beds, which is why rupestrian grasslands formed the aquifers that form springs and feed rivers.

The soil within this ecosystem is nutrient-poor and has high levels of aluminum and heavy metals. This fact, together with constant exposure to wind, rain and high temperature variations throughout the day, limits the vegetation’s growth and geographic distribution. Plants are also subject to frequent fires and become food for the herbivores who share their environment. In order to survive in this hostile environment, they have developed certain characteristics and behaviors and relationships with some animals, known as ecological resilience. As an example: There are species of herbaceous and woody-stemmed plants that have developed specialized underground organs that allow them to re-sprout numerous times after being damaged so the plant can live on. Some moss and fern species can survive almost completely dry conditions.

Fernandes and other researchers have discovered hundreds of plant and animal species in rupestrian grasslands over the last 45 years in their studies to learn about and protect this ecosystem. Still, less than 10% of its existing area is protected inside conservation units, meaning the species’ survival is at risk. According to these researchers, 82% of the grasslands could be gone by 2070 — an area equal to 68,000 square kilometers (26,000 square miles).

The list of threats is long: climate changes, urban growth, real estate speculation for luxury homes, illegal plant extraction for sale, hunting, pollution of rivers and spring drainage, the threat of invasive exotic species, the use of land for pasture and forestry, tourism and, especially, mining.

Restoration presents many challenges, especially because mining companies generally remove all the soil for mineral extraction. But research teams, communities and partnerships between universities and mining companies are geared up to protect one of the planet’s most biodiverse ecosystems.

Restoration options

“For a long time, it was believed that restoration of rupestrian grasslands was impossible for the simple reason that the plants grow in an inhospitable environment and, therefore, raising them in greenhouses wouldn’t work,” says Fernando Augusto de Oliveira e Silveira, associate professor in the Genetics, Ecology and Evolution Department at Minas Gerais Federal University’s Institute of Biological Sciences. He says this discourse, used within government organizations, by large corporations and in academia, was used as an excuse to not restore the ecosystem.

But in February, Oliveira and his colleagues debunked this myth in an article showing how rupestrian grassland species can be propagated, and with low-cost technology. Based on their tests with 117 endemic species, the researchers proved they can be cultivated in normal greenhouses with success rates in 80% of the plants. “There is no longer any plausible or technical justification for saying rupestrian grasslands can’t be restored,” he defends. Oliveira has focused his work on restoration of this ecosystem since 2000.

Geraldo Silveira has been gaining knowledge on the germination of rupestrian grassland species since he sold his car in 1995 to raise the money to build a greenhouse where he could test inducing cancer in plants. Plant propagation in laboratories at that time had been enough for researchers to begin questioning the use of exotic species like signal grass to cover degraded areas to meet environmental licensing requirements. “The real challenge is to produce seedlings on a large scale and help decision-makers understand that this is the best thing to do,” Silveira says.

Discoveries made over the last four decades by the few specialized researchers of this ecosystem led to the publication in 2022 of a free database of the seeds of 383 rupestrian grassland species. It includes information on the genetics of the population, seed creation and dispersal, dormancy and germination and other aspects. The researchers also created a list of 10 principles for restoration of the ecosystem.

More scientific work is being done in the Rio de Janeiro Botanical Garden’s seed bank, which houses four officially threatened species. In 2017, Fernando Silveira joined other researchers in studying species that needed to be reallocated from sites with veins of iron ore in Carajás where a mining company was set to start working. Because it is illegal for a company to drive any species to extinction, the plants had to be cultivated and planted in other places. “We monitored the flowering and fruiting phases of four plant species and did modeling studies for their niches to learn how, in a future scenario of climate changes, which would be adequate locations for their survival,” he explains.

Today, Silveira and his colleagues are implanting a restoration project in the Carajás region using a technique that is new to Brazil: topsoil transposition. Financed by the mining company as a means of meeting mandatory compensation measures set by its environmental license, the project moved 36 truckloads of topsoil from an area to be mined to another site that had been degraded. In this method — the most effective one, according to Silveira — the propagules, seeds, spores, fungi and microorganisms present in the transported soil will help to regenerate the degraded area.

This method is being used together with both seeding and planting revegetation strategies. Transplanting the underground organs of some species is also a possibility for generating resiliency withing the ecosystem, but this technique has not been used yet. Now that the topsoil has been relocated, the team will monitor the area over the next two years and compare the success rates of the different revegetation strategies to define the most effective ones. Ideally, the project would be monitored for at least a decade because processes in rupestrian grasslands unfold at a much slower pace than in forests.

“If we were to let the ecosystem restore itself alone, it would probably take between 800 and 1,400 years. But we don’t have that much time, so we need to do active restoration, which involves a much more integrative and functional approach than simply throwing out some seeds,” Silveira explains.

Other regions have been or are being restored by UFMG professors, at the calling of environmental agencies and public prosecutors’ offices. It is a victory for the researchers, who have for years now been fighting so-called “rehabilitation,” a process still being used by most mining companies. In rehabilitation, signal grass and eucalyptus trees are planted on degraded land, which does nothing to address the biology or ecology of rupestrian grasslands. In fact, it is little more effective than painting them green.

The impacts of mining

Aside from the need for human assistance, another limitation in restoring rupestrian grassland destroyed by mining companies is topography. Areas where employee living quarters were built are flat and easier to restore, but in the deep valleys carved out in search of ore, it is impossible to replace part of a mountain. “It is restoration in a way, because we manage to bring back a part of the biodiversity, but it’s not the full ecosystem,” Silveira says.

Mining causes a number of direct and indirect impacts. One of the main alterations the ecosystem undergoes is destruction of the water table in these mountains, which especially happens during iron ore extraction. When it is underground, iron ore acts as a sponge that holds rainwater. In the dry season, water is released from the aquifer to form bodies of water. Taking out the iron means removing this sponge, meaning that rainwater no longer can be stored underground and ends up running downhill as surface water.

And most of the mining done in Brazil is iron ore extraction. On average, it constituted 61% of all compounds removed during the 10 years between 2015 and 2024, according to data from the National Mining Agency. This is another factor threatening rupestrian grasslands: The two states where most iron ore is mined are Minas Gerais and Pará, also the two states where most of this ecosystem occurs.

“The big problem with mining is the way it alters the landscape,” says Bruno Milanez, associate professor of production engineering at Juiz de Fora Federal University’s Engineering School. Quoting author Eduardo Gudynas, he says that “mining is ‘ecological amputation’ because, just like when your amputate a limb, that landscape and its ecological functions are no longer there. There is no environmental management able to bring back what was cut away.”

Aside from altering the region’s water cycle, which impacts supply to rivers and consequently to nearby cities, mining also affects air quality and consumes large amounts of water, which can lead to supply problems and water contamination. Wildlife is also directly affected both during the construction of a mine and during its operation, as animals must either run away or live under the stress of the operation, Milanez says. The waste produced is equally worrisome: Soil with low levels of mineral content is piled up at the site, eventually becoming covered with grass. Tailings from processing plants are stored behind dams, which then run the risk of bursting.

Milanez also explains that mining spreads mostly in rural areas or small towns in the countryside, which drives speculation, expropriation and rising prices because of an influx of people. Some mines employ as many as 7,000 workers. “Usually, they are young, single men without families who come to live in containers and earn salaries a little higher than they would earn in the city. This means they have more money to spend in the region. Alcoholic consumption, sexual exploitation and violence all increase,” he explains. If there are traditional communities at the site to be mined, they will most likely be kicked out, which leads to social conflicts.

Milanez says the idea isn’t to do away with mining, but rather not to normalize mining that is carried out in any old fashion and at any cost. “If we look at all the impacts it has, we need to think about how much mining we can actually do and also how we can adapt our way of life to survive with the amount of ore it’s possible to extract without destroying the planet or affecting the people who have nothing to do with the process.”

Determining rate, scale and income distribution are essential factors in a more fair and less distressing mining model, says Milanez, who points out that such models could provide alternatives for the mining industry itself. Improving implementation of the law and monitoring is also needed, together with decreasing waste through effective recycling policy, redefining production and consumption standards and including the communities directly affected by the mines in decision-making processes, providing them with veto power. This has, in fact, proven to be an effective tool in protecting the rupestrian grasslands in Minas Gerais.

A popular front to protect the ecosystem

By law, activities that could harm the environment, like mining, cannot be carried out inside Brazilian conservation units. Legislation stipulates two types: Those under full protection have minimal human interference and those for sustainable use depend on their management plans to bar certain activities. Citizens may influence the creation of these areas, and in many cases, community groups have formed in Brazil to prevent mining companies from coming in.

According to the Territórios Livres de Mineração (Mining-Free Territories) movement, protests against mining companies have been successful in the cities of Muriaé, Caldas, Santa Bárbara, Serro and Serra do Gandarela in Minas Gerais, as well as Anitápolis (SC), Lago Grande Agroextractive Settlement Project (PA), Açailândia (MA), Roseli Nunes Resettlement Project (MT), Santa Quitéria (CE) and São José do Norte (RS).

The city of Piumhi lies in the Canastra Mountains, a region rich in rupestrian grasslands in Minas Gerais. Here, a grassroots movement resulted in the creation of Serras e Águas de Piumhi EPA (environmentally protected area), which comprises 12,000 hectares (30,000 acres) that are now protected from mining. The movement began in April 2023 because of drilling being done by three mining companies in the region. They were after iron, chromite and manganese in the mountains that are home to the water sources that supply 90% of the city’s population. After meetings, biking and hiking trips and public hearings, a petition with 70,000 names and technical studies on the biodiversity that served as the basis for creating the EPA, Law 2.767/2024 was approved in September 2024.

“It’s a concrete example of how grassroots movements together with scientific knowledge can have positive impact on territorial planning,” says João Luís Lobo, a botanist, lawyer and current director of Congonhas Municipality Protected Parks and Areas, located at the far south end of the Espinhaço Mountains. When he was doing research for his master’s degree, Lobo took part in a plant study in the region that was part of the bill that proved just how diverse the native species were — some of which were at risk of extinction — and discovered a new species of sunflower.

“Experiences like this help us to understand that development and conservation don’t have to be at war with one another,” says João Luís Lobo, “but the decision-making processes must be qualified by a social control filter so the different ways of occupying the territory will be sure to consider the ecological limits, social diversity and heritage.”

Lobo stresses that solutions to protect rupestrian grasslands from different threats include rigorous adherence to legislation together with increased priority for conservation and the establishment of a specific legal framework for rural landscapes, especially rupestrian grasslands. This would clearly guarantee legal recognition for this ecosystem.

People also need to learn about these grasslands, he says. “You don’t protect what you don’t know. Only when people understand that these ecosystems are part of Brazil’s natural and cultural heritage will they really be valued.” He also mentions how some activities that keep the rupestrian grasslands alive also generate income: eco-tourism, artisan cheese production (like the Canastra cheese produced in Piumhi), active conservation, payment programs for environmental services and the bioeconomy.

“Many species in the rupestrian grasslands have medicinal, aromatic or ornamental qualities, or they can be food. If managed in an ethical and sustainable way, they could generate products with high aggregate value. This would require investment in scientific research, in cooperative networks inside local communities and in public policy that would incentivize these productive chains.” When people’s lives are integrated with their environment, life can prosper.

Banner image: Researchers search for endemic and endangered amphibians in Pico do Itambé State Park, Minas Gerais. Image by Augusto Gomes.

This article was first published here in Portuguese on Jul. 7, 2025.

Citations:

Fernandes, G. W., Barbosa, N. P., Alberton, B., Barbieri, A., Dirzo, R., Goulart, F., … Solar, R. R. (2018). The deadly route to collapse and the uncertain fate of Brazilian rupestrian grasslands. Biodiversity and Conservation, 27(10), 2587-2603. doi:10.1007/s10531-018-1556-4

Fernandes, G. W., Arantes-Garcia, L., Barbosa, M., U. Barbosa, N. P., L. Batista, E. K., Beiroz, W., … O. Silveira, F. A. (2020). Biodiversity and ecosystem services in the Campo Rupestre: A road map for the sustainability of the hottest Brazilian biodiversity hotspot. Perspectives in Ecology and Conservation. Retrieved from https://www.sciencedirect.com/science/article/pii/S2530064420300638

Fernandes, G. W. (2016). Ecology and conservation of Mountaintop grasslands in Brazil. Springer. doi.org/10.1007/978-3-319-29808-5_1

Faria, F. S., Ordóñez‐Parra, C. A., Granata, L., Resende, L. V., & Silveira, F. A. (2025). Low‐cost technology supports propagation of endemic species from a global biodiversity hotspot. Restoration Ecology, 33(4). doi:10.1111/rec.70005

Ordóñez‐Parra, C. A., Dayrell, R. L., Negreiros, D., Andrade, A. C., Andrade, L. G., Antonini, Y., … Silveira, F. A. (2022). Rock n’ Seeds: A database of seed functional traits and germination experiments from Brazilian rock outcrop vegetation. Ecology, 104(1). doi:10.1002/ecy.3852

Arruda, A. J., Medeiros, N. F., Fiorini, C. F., Ordóñez‐Parra, C. A., Dayrell, R. L., Messeder, J. V., … Silveira, F. A. (2023). Ten principles for restoring campo rupestre, a threatened tropical, megadiverse, nutrient‐impoverished montane grassland. Restoration Ecology, 31(7). doi:10.1111/rec.13924

Credits

Xavier Bartaburu Editor

Maya Johnson Translator

Topics

BiodiversityCommunity-based ConservationConservationEcosystemsEnvironmentGrasslandsMiningPlantsWildlifeBrazilLatin AmericaSouth America

See Topics