Daniel Negreiros

This study evaluated the reliability of functional data imputed in the classification of ecological strategies according to the CSR model (competitive, stress-tolerant, ruderal) in tree species of a riparian Atlantic Forest in Brazil. Empirical data collected in situ from 68 species were compared with data imputed using the Bayesian Hierarchical Probabilistic Matrix Factorization (BHPMF) technique, considering key functional attributes such as leaf area and specific leaf area. The results showed significant divergences between the classifications based on measured and imputed data, especially for the traits directly involved in the formulation of the CSR model. While imputation overestimated selection for stress tolerance, field data indicated a greater diversity of strategies, with a greater presence of competitive species. The discrepancies point to limitations in the exclusive use of imputed data in local assessments of functional diversity, highlighting the need to integrate empirical collections and improvements in imputation methods for applications in environmental management and restoration.

Mining has left lasting scars on campo rupestre , one of the world's most diverse and threatened ecosystems. Here, we provide the first empirical evidence, on a secular scale, of the limits of passive recovery in these environments. We compared two abandoned mining post‐mining sites in Mariana, Brazil [one abandoned 50 years ago (Short‐term) and another more than 300 years ago (Long‐term)] with four well‐preserved reference sites to assess vegetation structure, floristic composition, and soil attributes. We estimated the structure and composition of woody vegetation, along with soil nutritional and textural properties, in replicated plots. We assessed multivariate differences in woody species composition among sites using Bray–Curtis‐based PCoA and PERMANOVA, and summarized soil multivariate patterns with PCA. Both sites exhibited reduced species richness, abundance, and structural heterogeneity, along with differences in soil nutritional and textural properties that were associated with plant community assembly patterns. Despite centuries of abandonment, the Long‐term site did not converge to reference conditions, while the Short‐term site showed only partial similarities (PCoA cumulative variation = 37.40%; PERMANOVA p < 0.001). Taken together, these findings suggest that passive recovery in campo rupestre is insufficient, even on centennial scales, compromising the return of biodiversity and ecosystem services essential to millions of people. They also underscore the urgent need for active science‐based restoration policies that are guided by the ecological attributes of reference ecosystems, going beyond mitigation rhetoric to ensure ecological functionality and resilience in post‐mining landscapes.

Background: Trait-based ecology has become central for understanding plant form, function and ecosystem processes, but progress has been hampered by biased representation in trait databases. As such, global trait syntheses remain strongly biased towards temperate forest biomes. Tropical savannas are the most extensive, biodiverse and disturbance-driven ecosystems worldwide, yet are poorly represented in functional trait databases, limiting ecological inference and applied decision-making. Scope: Here, we introduce the Cerrado Plant Traits (CPT), an open-access initiative compiling and standardising plant functional trait data for the Brazilian Cerrado, the world’s most biodiverse tropical savanna. CPT integrates trait information for all major plant organs (whole-plant, root, shoot, leaf, flower, fruit and seed) across vegetation types in the Cerrado, drawing on a collaborative and inclusive research network. The current version of CPT compiles data from 148 datasets, totalling 113,859 curated trait records for 2,134 taxonomically verified species across 150 families. Trait records span pristine, degraded and restored environments and capture both interspecific and intraspecific variation. Whole-plant and leaf traits dominate the current dataset, while belowground and reproductive traits remain comparatively underrepresented, highlighting key priorities for future research. Conclusions: By substantially increasing the representation of savanna species in global trait repositories, CPT enables tests of ecological hypotheses across multiple levels of organization, analyses of trait–environment relationships across fire, soil and climatic gradients, and robust comparisons across forest–savanna transitions. Beyond its scientific value, CPT provides a practical, standardised resource to support conservation planning, restoration programs and evidence-based policy in a biodiversity hotspot facing accelerating land-use and climate pressures.

Projected warming and drying raise concerns about the resilience of stress‐adapted ecosystems, including the Brazilian Campo Rupestre, an exceptionally biodiverse mountaintop grassland mosaic on ancient, nutrient‐poor substrates. Here, we combine field‐based trait data and long‐term remote sensing to assess the functional structure and temporal dynamics of these communities. Using foliar trait measurements from 247 vegetation plots across five contrasting habitats, we 1) quantify contemporary community‐level functional structure, 2) evaluate how edaphic and climatic filters shape spatial variation in community‐weighted foliar traits, and 3) reconstruct multi‐decadal trait trajectories by hindcasting from long‐term Landsat reflectance (1984–2022). Contemporary communities occupy a narrow and predominantly conservative region of the leaf‐economic trait spectrum, yet habitats differ in their functional positions within CSR strategy space, indicating non‐uniform trait coordination despite overall conservatism. Soil texture and acidity define the primary conservative–acquisitive axis of trait variation, while climatic water balance acts as a secondary modulator; together, these predictors explain 39% of the spatial variation in community‐weighted traits. Contrary to expectations of increasing conservatism under progressive climatic stress, Landsat‐based hindcasts reveal only modest temporal reorganisation. Specific leaf area and leaf area increase across habitats, while leaf dry matter content declines slightly, indicating a subtle relaxation of conservative trait expression. Temporal changes are small relative to the pronounced spatial differentiation, suggesting strong functional inertia in this OCBIL system. Overall, Campo Rupestre communities persist within a conservative functional domain while exhibiting fine‐scale, habitat‐dependent differentiation structured by enduring soil and water‐balance gradients.

Questions: How can a reference framework be developed for the ecological restoration of the campo rupestre, a biodiverse Brazilian ecosystem? Specifically, what are the challenges posed by habitat heterogeneity and species turnover in the mountain range, and how can they be addressed to create ‘Blue Lists’, defined as habitat‐ and ecosystem‐specific reference species lists that guide assisted ecological restoration? Location: The campo rupestre ecosystem of Brazil's Espinhaço Mountain Range spans a strong latitudinal and altitudinal gradient of nearly 10 degrees, extending from the Southern Espinhaço in Minas Gerais to the Chapada Diamantina in Bahia. Methods: Phytosociological data from 20 studies and 101 species lists were analysed. Species were classified by substrate (quartzite and ferruginous) and habitat types, further divided into herbaceous and woody strata. A cluster analysis using Sørensen distance identified floristic clusters, while zeta diversity analyses evaluated species turnover at species, genus, and family levels. Phylogenetic and spatial signals were assessed to examine species co‐occurrence patterns. Results: A total of 1,126 plant species were identified, exhibiting significant compositional variation between substrates and habitat types. Floristic clusters demonstrated high species turnover, with few species shared across habitats. Phylogenetic analysis revealed convergent evolutionary patterns, and spatial analysis showed significant geographic clustering of species composition. Conclusions: Restoration efforts must account for habitat heterogeneity and substrate‐specific species compositions to prevent biotic homogenization and enhance ecosystem biodiversity and resilience. Detailed Blue Lists, defined as habitat‐specific target species lists, are critical for effective ecological restoration. Future studies should prioritize closing knowledge gaps, especially in understudied regions of the campo rupestre.

Background and Aims: Understanding how environmental filtering shapes plant functional strategies is essential for explaining community assembly and, ultimately, for guiding ecological restoration under global change. In highly stressful ecosystems such as the Brazilian campo rupestre, strong edaphic gradients and low ecosystem resilience make trait-based approaches particularly valuable for identifying reference conditions and restoration targets. Here, we investigate how plant functional traits and competitor–stress-tolerant–ruderal (CSR) ecological strategies are filtered by soil properties across distinct campo rupestre habitats and explore the implications for restoration planning. Methods: We quantified soil physicochemical properties, plant functional traits and CSR strategies for 111 species across 120 plots representing four distinct campo rupestre habitats (rocky outcrops, stony grasslands, sandy grasslands and peatbogs) in the Espinhaço Mountain Range, Brazil. Trait–environment relationships were evaluated using fourth-corner and RLQ analyses to assess both bivariate and multivariate associations between species traits, ecological strategies and edaphic gradients. Key Results: Plant community assembly was strongly structured along a primary gradient of resource and water availability. Ruderal strategies were associated with higher nutrient concentrations and greater soil water-holding capacity, whereas stress-tolerant strategies were favoured in coarse-textured, nutrient-poor soils. Individual traits mirrored these patterns: higher leaf mass per area and wood density were associated with resource-poor environments, whereas greater plant height was associated with increased nutrient availability. These results indicate strong coupling between plant functional strategies and soil conditions, with limited functional redundancy under extreme environments. Conclusions: Our findings demonstrate that edaphic filtering is a key driver of functional differentiation in campo rupestre plant communities. By linking soil conditions to functional strategies and traits, this study provides a mechanistic framework for defining habitat-specific functional targets, offering a robust basis for conservation and ecological restoration in this highly vulnerable ecosystem.

Although holding high levels of biodiversity and providing uncountable ecosystem services, wetlands remain poorly understood and protected, particularly in South America, where these ecosystems are being destroyed quickly by infrastructure facilities, land-use change, water pollution, and fire, mostly related to climatic triggers. Nevertheless, few studies have assessed and synthesized fire patterns among South American wetlands. The lack of an overarching view on this issue has constrained our knowledge of how fire patterns affect these ecosystem, and thus hindered integrated fire management at a large scale. Overlapping wetland maps to terrestrial ecoregions and satellite-derived fire activity indicators, we ranked ecoregions mostly affected by wetland fires. Of 109 South America’s ecoregions, the Pantanal, Cerrado, Llanos, Beni savanna, and Humid and Dry Chaco accounted for the most of fire foci on wetlands and the largest burned wetland area. There is a pressing need to create a cross national system based on warnings, helping decision-makers, local communities, scientists, and fire brigades to properly manage fires in South America’s wetlands at a large scale.

Question: Understanding how nutrient availability and soil conditions influence ecosystem functionality is crucial for predicting changes in the plant community and informing ecological restoration programs. Here, we aimed to characterize the ecological strategies of the most abundant species in an ironstone ecosystem, to understand how edaphic parameters drive these strategies, and to investigate how species taxonomic, functional, and phylogenetic diversities differ among habitats within the ecosystem. Location: The study was conducted in three municipalities in the southern portion of the Espinhaço mountain range in Minas Gerais, Brazil. We sampled plots in three habitat types with distinct soil characteristics. Methods: We employed the CSR framework to characterize the ecological strategies of the most common species across 87,100‐m ² plots. We also measured 19 edaphic parameters in these plots to evaluate their influence on ecological strategies. Results: Most of the 46 sampled species displayed a high proportion of stress‐tolerant strategies due to their small leaves with high dry matter content. This pattern reflects the preponderance of S‐selection at the plot level. Conversely, plots with finer sand and relatively fertile conditions had more competitive strategies, and the species within showed larger leaves with less biomass investment. Conclusions: This study highlights the role of abiotic filters in shaping CSR strategies in ironstone campo rupestre , emphasizing the trade‐off between growth and resource conservation. The low growth rates of species reveal the ecosystem's vulnerability to increased soil removal and disturbances. Additionally, small‐scale edaphic variation relates mainly to physical properties, selected species composition, and functionality.