How representative is flora rescue? A 15-year analysis of species composition in iron ore mining in the Brazilian Amazon and the gap for endemic and rare species conservation
A 15-year analysis of flora rescue as a mitigation measure in the Brazilian Amazon's Carajás mining region reveals that while programs have increased in yield, they capture only a small fraction of the local canga flora, hold less phylogenetic diversity than expected for their richness (SES-PD), and are dominated by a few abundant species, highlighting a critical gap in effectively conserving endemic and rare biodiversity.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
In the heart of the Brazilian Amazon, iron-rich rock outcrops known as canga form flat to undulating crusts across plateau tops. These landscapes are home to a unique collection of plants that have evolved over millions of years to survive in nutrient-poor soil and harsh conditions. Because these rocks are also rich in iron ore, they are prime targets for mining. When mining companies extract the ore, they must clear the vegetation, which threatens species found nowhere else on Earth. To mitigate this impact, environmental laws require companies to perform "flora rescue." This involves moving plants—including the removal of epiphytes from host trees—to nearby receptor areas. This mitigation measure aims to preserve biodiversity by relocating species from areas slated for clearing to nearby receptor areas. However, a critical question remains: does this process capture the true diversity of the ecosystem, or does it focus on a specific subset of species?
A team of researchers set out to answer this question by looking at fifteen years of data from one of the world's largest mining operations in the Carajás region. They examined the official records of a massive plant rescue and reintroduction program associated with iron ore and manganese mining. The database they analyzed contained more than 350,000 records of individual plants moved between 2010 and 2024. The researchers treated the corporate reports as a historical record to see what the program had actually accomplished. Their goal was to compare the list of rescued plants against a comprehensive scientific inventory of the plants that naturally grow on these ironstone outcrops. This comparison would reveal whether the rescue effort was capturing the full diversity of the canga ecosystem or just a small, unrepresentative slice of it.
The analysis began by cleaning up the data, a necessary step because the original records contained many uncertain names, such as "species similar to X" or plants identified only by their genus. After filtering out these uncertain entries, the researchers were left with a confirmed list of 277 distinct species. When they compared this list to the known canga species, they found a significant gap: the rescued plants represented about 10% of the known canga species. Even when the researchers narrowed their comparison to the floristic lists of the specific plateaus (N4 and N5) where the mining took place, the rescue program had only captured about 14% of the species. This means that a vast majority of the species characteristic of these plateaus were absent from the rescue records.
The study also revealed that the plants being moved were not a random sample of the ecosystem. The rescue effort was heavily skewed toward a few specific types of plants. Just 37 species accounted for nearly 90 percent of all the individual plants moved. These were mostly concentrated within a few specific clades, including many epiphytic groups. Furthermore, the researchers found that the phylogenetic diversity of the rescued assemblages was lower than expected for their richness (SES-PD), suggesting that the rescue process does not represent the broad evolutionary history of the canga flora.
Despite these shortcomings, the researchers observed certain patterns over time. The number of species rescued per year increased after 2018, coinciding with the publication of the regional floras; and rescue yield per suppressed hectare rose independently of suppressed area.
The researchers concluded that while the rescue program has generated a massive amount of data and moved hundreds of thousands of plants, it has not yet achieved true ecological representativeness. The current system measures success by counting how many individuals were moved, a metric that is easy to audit but does not guarantee that the unique biodiversity of the canga is being preserved. The study suggests that for these programs to truly protect the environment, regulators and companies need to shift their focus from counting individuals to measuring how well the rescued community matches the original ecosystem. By setting clear targets based on the known diversity of the area, the industry could move beyond simple paperwork and toward more effective conservation. Until then, the gap between the species absent from rescue records and the total known diversity remains a significant concern for the ecosystem.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.