The Brazil nut tree Bertholletia excelsa is an icon of Amazon conservation through sustainable use. Moderate disturbance, such as that caused by swidden agriculture, favours this heliophilic species. Our systematic literature review of Bertholletia studies and historical records addresses the following questions: do slash-and-burn farming systems increase Bertholletia density and growth? What do historical records reveal about the links between Bertholletia life history and human occupation? And what policies and regulations shape the current context for harnessing this synergistic potential for sustainable use? Compared to mature forests, slash-and-burn fallow seedling/sapling densities (11–82 individuals ha–1, with a mean of 29 individuals ha–1) are greater and faster-growing. Extant Bertholletia trees that were cut and burned during swidden preparation resprout as forked individuals and supplement new seeds buried by Dasyprocta spp. The presence of large forked Bertholletia trees and the occurrence of anthropogenic soils, particularly brown soils associated with Brazil nut tree groves, provide evidence that extant Bertholletia groves may be islands of active and passive agroecological management by ancestral Indigenous populations and local communities. This supports the notion that conservation through sustainable use can maintain Amazonian megadiversity. Furthermore, fire has been used in the Amazon since the onset of crop cultivation (including Bertholletia) c. 4500 years ago, suggesting that a more effective approach than banning fires would be to implement a systematic and methodical fire and fuel management strategy, given the ineffectiveness of command-and-control policies in this regard. The 124 conservation units and Indigenous lands in the Amazon containing Brazil nut trees reinforce the importance of policies to create protected areas. Evidence suggests that the presence of an Amazonian biocultural forest – a phenomenon resulting from the interaction between human activities and natural processes – can be sustainably used to promote what might be termed ‘sociobiodiversity conservation’.

The Amazon comprises the most biodiverse region in the world, but, despite being highly threatened by human-induced environmental changes, little is known about how those changes influence the remaining forest’s extent and configuration in Brazil’s arc of deforestation. We analysed the spatial and temporal dynamics and the configuration of forest cover in Brazil’s state of Rondônia over 34 years. We calculated seven landscape metrics based on freely available satellite imagery to understand the habitat transformations. Overall, natural vegetation cover declined from 90.9% to 62.7% between 1986 and 2020, and fragmentation greatly increased, generating 78 000 forest fragments and 100 000 fragments of ‘natural vegetation’, which also includes forest. We found that c. 50% of the vegetation is within c. 1 km of the nearest forest edge, and the mean isolation between fragments is c. 2.5 km. Most natural vegetation and forest vegetation layers outside protected areas (PAs; Brazil’s ‘conservation units’) and Indigenous territories (ITs) are >10 km from the nearest PA or IT. This reduction of natural vegetation in Rondônia is posing major threats to the survival of species and is undermining the dynamics of ecosystems. Measures to control deforestation and avoid the reduction of large remnants are urgently needed.

Given the ongoing global extinction crisis, preserving genetic diversity is critical for long-term ecosystem resilience. A large, openly available DNA barcoding database could support this goal by allowing the identification of ‘genetic hotspots’ for conservation planning. We studied 77 woody species in the Monte Desert, South America’s largest dryland, using ITS2 and rbcLa markers to identify haplotypes for each species. We modified a previously published genetic diversity criterion, which prioritizes uniqueness, to also consider species ubiquity. We then mapped this genetic diversity metric, calculated Faith’s phylogenetic diversity (PD) index and overlaid our map with protected areas and permanent plantations. We identified five robust genetic hotspots, three of which coalesced into a central ‘mega-hotspot’. Alarmingly, most hotspots lay outside existing protected areas, and two overlapped with permanent plantations. As expected, high PD did not consistently align with high genetic diversity or species richness, suggesting that in the Monte Desert current protected areas overlook key genetic and PD. Our study highlights the importance of integrating DNA barcoding from understudied geographic regions into conservation plans.

Biodiversity knowledge gaps and biases persist across low-income tropical regions. Genetic data are essential for addressing these issues, supporting biodiversity research and conservation planning. To assess progress in wildlife genetic sampling within the Philippines, I evaluated the scope, representativeness, and growth of publicly available genetic data and research on endemic vertebrates from the 1990s through 2024. Results showed that 82.3% of the Philippines’ 769 endemic vertebrates have genetic data, although major disparities remain. Reptiles had the least complete coverage but exhibited the highest growth, with birds, mammals, and amphibians following in that order. Species confined to smaller biogeographic subregions, with narrow geographic ranges, or classified as threatened or lacking threat assessments were disproportionately underrepresented. Research output on reptiles increased markedly, while amphibian research lagged behind. Although the number of non-unique authors in wildlife genetics studies involving Philippine specimens has grown steeply, Filipino involvement remains low. These results highlight the uneven and non-random distribution of wildlife genetic knowledge within this global biodiversity hotspot. Moreover, the limited participation of Global South researchers underscores broader inequities in wildlife genomics. Closing these gaps and addressing biases creates a more equitable and representative genetic knowledge base and supports its integration into national conservation efforts aligned with global biodiversity commitments.

Megacities around the world are increasingly confronted with conservation and restoration bottlenecks due to the competing demands of urban expansion and environmental conservation. This study investigates conservation prioritization strategies for balancing biodiversity protection, ecosystem service (ES) supply and landscape connectivity in rapidly urbanizing Beijing. By employing spatially explicit modelling and prioritization scenario techniques, we identify spatially heterogeneous priority zones. We demonstrate that high-value areas for ES supply, particularly carbon storage and water regulation, concentrate primarily in Beijing’s north-western mountainous regions, covering c. 62% of the city’s area. Conversely, critical habitats for threatened species and key connectivity corridors are dispersed, with 22.89% of critical habitats located within urban built-up areas. Gap analysis reveals limited alignment between Beijing’s current ecological security patterns, with only 9.6% coverage of the identified top 10% conservation priority zones, especially within the metropolitan core. The study underscores significant trade-offs among different ecological objectives and multi-criteria conservation strategies. We propose an optimized conservation framework based on zonation analysis to guide targeted landscape planning decisions. This approach provides actionable insights for urban policymakers to achieve comprehensive sustainability, emphasizing the importance of protecting critical ecological areas in both urban and rural landscapes amid ongoing urban expansion.

Monitoring wildlife populations in vast, remote landscapes poses significant challenges for conservation and management, particularly when studying elusive species that range across inaccessible terrain. Traditional survey methods often prove impractical or insufficient in such environments, necessitating innovative technological solutions. This study evaluates the effectiveness of deep learning for automated Bactrian camel detection in drone imagery across the complex desert terrain of the Gobi Desert of Mongolia. Using YOLOv8 and a dataset of 1479 high-resolution drone-captured images of Bactrian camels, we developed and validated an automated detection system. Our model demonstrated strong detection performance with high precision and recall values across different environmental conditions. Scale-aware analysis revealed distinct performance patterns between medium- and small-scale detections, informing optimal drone flight parameters. The system maintained consistent processing efficiency across various batch sizes while preserving detection quality. These findings advance conservation monitoring capabilities for Bactrian camels and other wildlife in remote ecosystems, providing wildlife managers with an efficient tool to track population dynamics and inform conservation strategies in expansive, difficult-to-access habitats.

Every private rural property in Brazil must maintain a percentage of its area with conserved native vegetation, establishing a legal reserve area (LRA). This percentage is defined by the Brazilian Federal Law on Protection of Native Vegetation and is 20–80% of the rural property, depending on the political boundary and vegetation formation. However, the Environmental Code of Mato Grosso stipulates that vegetation type determines the LRA percentage in the state of Mato Grosso, considering both vegetation formation and floristic composition. In addition, the state adopts a coarse-scale vegetation map as a reference, despite the existence of a more accurate official map. In this study, we calculated the impacts of these provisions by combining legal interpretation, spatial analysis and ecological reasoning, in accordance with the scientific concepts of ecology, and intersecting official maps from the State Environmental Agency and the Brazilian Institute of Geography and Statistics, at the scales of 1:1 000 000 and 1:250 000, respectively. A total of 9 045 065 ha could have their LRA requirement reduced from 80% to 35% under Mato Grosso’s legislation, potentially authorizing deforestation in the Amazon, Cerrado and Pantanal, in apparent contradiction to federal law. These findings highlight that issues related to the concepts, classifications and mapping scales for defining native vegetation have significant implications for biodiversity conservation.