Numerous geometric earthworks have been found in southwestern Amazonia, evidence of a monument-building precolonial civilization1-3. This knowledge has contributed to the discussion about the deep history and the precolonial urbanism of Amazonia, its sociocultural contexts and its environmental legacy2,4-8. Estimates of the geographical extent and population size of the forest civilization have remained weak, because the detection of earthworks has been almost exclusively limited to deforested areas where vegetation does not completely hide the structures. Here we used canopy-penetrating LiDAR data over 4,430 km of flight to define the geographical distribution and quantity of earthworks in this region better. On the basis of both existing data and our LiDAR documentation, we show that this cultural area alone contains as many earthworks as previously estimated for the whole of Amazonia. We also propose that the total human population of the area in 100-300 AD was 1.25-3 million people. If similar results are found elsewhere in Amazonia, the total precolonial population density and its effect on the current Amazonian soil and forest structure, biodiversity and even some modelling of the global climate should be re-evaluated.

Humans have domesticated and used palms in tropical forests for millennia. The extent to which past human palm use affects modern tropical systems remains relatively unexplored due to lack of a good fossil record. However, palms produce an array of phytoliths, which are siliceous cell structures that preserve well in the soil even after the organic plant material has decayed. Phytoliths vary in morphology and can be identified to group level and sometimes even to species. We analyzed if palm phytolith assemblages sampled from the soil reflect palm species compositions and abundances in 63 western Amazonian transects. Principal coordinate analysis (PCoA) and Mantel tests showed that palm communities and phytolith assemblages had similar turnover patterns. Negative binomial generalized linear models indicated that the abundances of nine of the 13 common palm genera were significantly related to a combination of soil cation concentrations, grass phytolith percentages, and palm phytolith PCoA axis scores. These results suggest that phytoliths have potential as quantitative indicators of changes in palm abundances in paleoecological and archaeological reconstructions.

Amazon rainforests face intensifying water stress due to increases in vapour pressure deficit and changing hydrological regimes. Embolism resistance (Ψ50) is a critical metric of tree survival under drought conditions, it is defined as a plant's capacity to resist disruption of xylem water flow due to air bubble formation from water stress. However, measurements of Ψ50 are only available for a limited number of Amazon locations and species. Conversely, data on forest taxonomic composition are abundant across Amazonia, and if Ψ50 is conserved phylogenetically, these data could provide a way to scale-up drought resistance patterns. Here we evaluate Ψ50 measurements across non-flooded Amazonian tree taxa and reveal a moderate phylogenetic signal, with phylogenetic conservatism evident at the family-level. Notably, Fabaceae is amongst the most embolism-resistant tree families in Amazonia. Leveraging the phylogenetic signal we use species composition and tree size data from 448 forest plots across Amazonia to produce a macroecological assessment of Amazonian vulnerability to embolism. The resulting estimate spatial pattern reveals that forests in the Brazilian and Guiana Shield regions, where Fabaceae abundance is high, show strong resistance to embolism. In contrast, tree communities in Western Amazonia appear more vulnerable to embolism, suggesting a reduced capacity to withstand future drought conditions.

Deforestation of the Amazon rainforest biome has been of considerable international concern, as extensive forest loss has negative impacts on global carbon storage, hydrological cycles and biodiversity. Ongoing climatic change is predicted to make these effects worse, as climatic models suggest that increasing temperatures will often be accompanied by decreasing precipitation. Such change would put the continued existence of the rainforest biome at risk, as moisture-demanding species would be replaced by more drought-resistant ones. Field observations already indicate that even those Amazonian forests that are not directly affected by deforestation have started to change, and that they grow faster, store less carbon and contain more lianas than before. However, field measurements can only be carried out in a limited number of sites, and these only cover a minute part of the entire Amazon biome. To get a more complete understanding of the changes within the remaining Amazonian forests, we are carrying out broad-scale analyses using Landsat imagery. We first did pixel-based compositing using all image acquisitions within a 10-year time window in order to obtain a clean cloud-free reflectance surface. Two such composites were produced for different time periods (2000-2009 and 2013-2022) in order to identify potential changes across the forested landscape. The results are expected to identify areas where significant but non-obvious changes in the structure and/or function of the forest could be happening, and thereby to facilitate in assessing the degree of threat to the ecosystem

Wood density is a critical control on tree biomass, so poor understanding of its spatial variation can lead to large and systematic errors in forest biomass estimates and carbon maps. The need to understand how and why wood density varies is especially critical in tropical America where forests have exceptional species diversity and spatial turnover in composition. As tree identity and forest composition are challenging to estimate remotely, ground surveys are essential to know the wood density of trees, whether measured directly or inferred from their identity. Here, we assemble an extensive dataset of variation in wood density across the most forested and tree-diverse continent, examine how it relates to spatial and environmental variables, and use these relationships to predict spatial variation in wood density over tropical and sub-tropical South America. Our analysis refines previously identified east-west Amazon gradients in wood density, improves them by revealing fine-scale variation, and extends predictions into Andean, dry, and Atlantic forests. The results halve biomass prediction errors compared to a naïve scenario with no knowledge of spatial variation in wood density. Our findings will help improve remote sensing-based estimates of aboveground biomass carbon stocks across tropical South America.

Several studies have documented dominance by few species in Amazonian forests. Dominant species tend to be either locally abundant (local dominants) or regionally frequent (widespread dominants) but rarely both (oligarchs). Here, we explore relationships between dominance and functional traits. We ask whether: (i) dominance is associated with specific functional profiles and (ii) dominance patterns (local vs. widespread dominants) are associated with different functional traits. We combined census data from 503 forest inventory plots across four lowland forest habitats in western Amazonia with trait information for ~2600 tree species, encompassing data collected in the focal plots and data from published sources. We considered traits that relate to leaf, wood, seed and whole‐plant strategies: specific leaf area (SLA), leaf area (LA), N content per unit leaf mass (LN), wood density (WD), seed mass (SM) and maximum diameter at breast height (DBH max ). Our results reveal that dominant species display different trait combinations depending on the habitat type. Taller dominant species exhibit higher regional frequency, associated with higher dispersal ability and lower local abundance, likely due to negative density dependence. Greater SM contributes to higher regional frequency of dominant species via greater dispersal by birds and mammals and seedling survival. Finally, traits related to resource conservation strategies, such as lower SLA, LA, LN and greater WD, favour higher local densities across most habitats, while the opposite pattern was linked to higher regional frequency. Synthesis . Our findings reveal that (i) dominance is associated with different functional traits depending on the habitat type, and (ii) different functional trait values define distinct dominance patterns. Our study exemplifies the potential of trait‐based approaches to illuminate the ecological mechanisms that may underlie dominance in tropical forests. Finally, accounting for both local abundance and regional frequency when studying dominance is likely to improve our understanding and forecasting of how different species will respond to global change drivers in western Amazonia.

Accurate detection of deforestation and logging activities is useful to monitor large scale damages in the Amazon forests. In this study, we focused on the use of deep learning based one dimensional convolutional neural network (1D-CNN) with Hyperspectral Precursor of the Application Mission (PRISMA) hyperspectral data for the detection of deforestation in the Amazon Forest. The PRISMA data was pre-processed to remove noisy bands, water absorption and some blue spectrum bands. Three main classes were identified and sampled as ground truth for classification: forest, deforestation and waterbodies. 1D-CNN were parameterised to obtain a classified map and then accuracy assessment was performed. Model achieved a very high overall accuracy of 98.92%, confirming that the method can be used for accurate mapping of deforestation.

Aim: We test the hypothesis that wind dispersal is more common among emergent tree species given that being tall increases the likelihood of effective seed dispersal. Location: Americas, Africa and the Asia-Pacific. Time period: 1970-2020. Major taxa studied: Gymnosperms and Angiosperms. Methods: We used a dataset consisting of tree inventories from 2821 plots across three biogeographic regions (Americas, Africa and Asia- Pacific), including dry and wet forests, to determine the maximum height and dispersal strategy of 5314 tree species. A web search was used to determine whether species were wind-dispersed. We compared differences in tree species maximum height between biogeographic regions and examined the relationship between species maximum height and wind dispersal using logistic regression. We also tested whether emergent tree species, that is species with at least one individual taller than the 95% height percentile in one or more plots, were disproportionally wind dispersed in dry and wet forests within each biogeographic region. Results: Our dataset provides maximum height values for 5314 tree species, of which more than half (2914) had no record of this trait in existing global databases. We found that, on average, tree species in the Americas have lower maximum heights compared to those in Africa and the Asia Pacific. The probability of wind dispersal increased significantly with tree species maximum height and was significantly higher among emergent than non-emergent tree species in both dry and wet forests in all three biogeographic regions. Main conclusion: Wind dispersal is more prevalent in tall, emergent tree species than in non-emergent species and may thus be an important factor in the evolution of tree species maximum height. By providing the most comprehensive dataset so far of tree species maximum height and wind dispersal strategies, this study paves the way for advancing our understanding of the eco-evolutionary drivers of tree size.

Dominance of neotropical tree communities by a few species is widely documented, but dominant trees show a variety of distributional patterns still poorly understood. Here, we used 503 forest inventory plots (93,719 individuals >= 2.5 cm diameter, 2609 species) to explore the relationships between local abundance, regional frequency and spatial aggregation of dominant species in four main habitat types in western Amazonia. Although the abundance-occupancy relationship is positive for the full dataset, we found that among dominant Amazonian tree species, there is a strong negative relationship between local abundance and regional frequency and/or spatial aggregation across habitat types. Our findings suggest an ecological trade-off whereby dominant species can be locally abundant (local dominants) or regionally widespread (widespread dominants), but rarely both (oligarchs). Given the importance of dominant species as drivers of diversity and ecosystem functioning, unravelling different dominance patterns is a research priority to direct conservation efforts in Amazonian forests. Dominance of neotropical tree communities by a few species is widely documented, but dominant trees show a variety of distributional patterns still poorly understood. Contrary to the widely supported positive abundance-occupancy relationship in ecology, we found that among dominant Amazonian tree species, there is a strong negative relationship between local abundance and regional frequency and/or spatial aggregation across habitat types. Our findings suggest an ecological trade-off whereby dominant species can be locally abundant (local dominants) or regionally widespread (widespread dominants), but rarely both (oligarchs).image

Data on the Peruvian Malaise trapping carried out in 1998, 2000, 2008 and 2011 by the University of Turku. Flying insects were collected by Malaise traps in Peruvian Amazonia by Ilari Sääksjärvi and Isrrael Gómez. The dataset contains data on what samples were collected, on the trap sites, on the vegetation around the trap sites, and on the weather. More importantly, it contains information on how this data was compiled. There was no complete list of Peruvian Malaise samples before this, and the sample data had to be combined from multiple sources. This dataset is also described in the associated paper.

ABSTRACT In Amazonian terra-firme non inundated forests, local floristic composition and species occurrence are explained by water availability as determined by topographic conditions. Topographic complexity can render these conditions quite variable across the landscape and the effects on plant ecological responses are difficult to document. We used a set of topographically defined hydrological metrics to evaluate community composition and single-species responses of four plant groups [pteridophytes (ferns and lycophytes), Melastomataceae, palms (Arecaceae) and Zingiberales] to topographic conditions in the middle Juruá River region, in western Brazilian Amazonia. The area spans two geological formations (Içá and Solimões) with contrasting topography. River terraces are also found along the main rivers in the area. Local topographic conditions were approximated by height above the nearest drainage (HAND), slope, and Strahler´s drainage order, all obtained from a SRTM digital elevation model (DEM). Data were analyzed using linear and generalized linear mixed models and regression trees. HAND was most successful in explaining floristic composition for all plant groups, except for Melastomataceae, and was more important in the hilly Içá formation than in the Solimões. Individual occurrences of 57% species were predicted by at least one of the topographic variables, suggesting a marked habitat specialization along topographic gradients. For these species, response models using SRTM-DEM-derived variables gave similar results than models using field-measured topography only. Our results suggest that topographical variables estimated from remote sensing can be used to predict local variation in the structure of plant communities in tropical forests.

Silica is the best-known component filling the spaces that form phytoliths in many plants, but phytoliths may also contain other elements. We used scanning electron microscopy with energy-dispersive X-ray spectroscopy (SEM–EDX) to map element distributions in the leaves of Guadua bamboo, which is a successful competitor in southwestern Amazonian forests. We emphasize immobile elements that can be mineralized (silicon and calcium) as well as potassium, an abundant mobile nutrient with many vital functions. We discovered high silicon (Si) content with little or no calcium (Ca) or potassium (K) in bulliform cells, bilobate shaped short cells and stomata, all of which can form phytoliths, and moderately high Si content in the bundle sheet, prickle tips and papillae. K often surrounded Si-loaded cells, Si and K had overlapping distributions in the intercostal areas near vein margins, and Ca showed abundant spotted distribution in the intercostal areas. The dark inside content of the costal zones of some samples in light microscopy showed no mineralization but prominently contained sulfur. Adjacent fusoid cells showed different Si, K and Ca combinations, which may suggest potentially variable functions of these cells. Widespread Si deposition strengthens Guadua bamboo leaves and may help it to outcompete tree species during periods of drought.

We examine the Indigenous uses, oral histories, and knowledge of native Guadua bamboo species in southwestern Amazonia. Two Guadua species form dense stands in which individual plants die en masse at regular intervals of about 28 years. Scholars suggested that pre-colonial earth builders took advantage of these die-off events as a natural aid in removing the forest to construct geometric earthworks. Our results show that Guadua species have a significant position in Indigenous socio-cosmologies, land use, and as a protector of diverse resources. Indigenous ontological understandings cannot be separated from discussions of the abundance and geographical distribution of Guadua as a critical controlling factor in the vegetation structure and function of southwestern Amazonian rain forests. Furthermore, oral histories point to the connection between land management and bamboo, as well as bamboo and the use of fire, conforming to the suggestion of opening ceremonial spaces in bamboo patches in pre-colonial earthwork societies.

In Finland, the distribution area of the taiga tick, Ixodes persulcatus (Schulze, 1930), is nested within a broader area of distribution of a congeneric species, the sheep tick, Ixodes ricinus (Linnaeus, 1758) (Acari: Ixodidae). We assess whether distinct environmental adaptations or dispersal history provides a more parsimonious explanation for the differences in the distributions of the two common and medically important ixodids in Finland. We used an innovative spatially constrained randomization procedure to analyze whether crowdsourced occurrence data points of the two tick species had statistically different associations with any of the 28 environmental variables. Using points of presence in a region of species co-occurrence, we built Maxent models to examine whether environmental factors or dispersal history could explain the absence of I. persulcatus in a part of the range of I. ricinus in Finland. Five environmental variables-number of inhabitants, road length, elevation above sea level, proportion of barren bedrock and boulders, and proportion of unsorted glacial deposits-were significant at p ≤ .05, indicating greater between-species difference in original than in the randomized data. Of these variables, only the optimum value for unsorted glacial deposits was higher for I. persulcatus than for I. ricinus. Maxent models also predicted high relative habitat suitability (suitability >80%) for I. persulcatus south of its current, sharply bounded distribution range, suggesting that the species has not fulfilled its distribution potential in Finland. The two most common and medically relevant ixodids in Finland may colonize habitats with different environmental conditions. On the contrary, the recent establishment and ongoing dispersion of I. persulcatus in Fennoscandia rather than environmental conditions cause the southernmost distribution limit of the species in Finland.

Remote sensing can provide useful explanatory variables for tree species distribution modeling, but only a few studies have explored this potential in Amazonia at local scales. Particularly for tropical forest management it would be useful to be able to predict the potential distribution of important tree taxa in areas where field data is as yet missing. Forest concessions produce valuable census data that cover large areas with high sampling effort and can be used as occurrence data in species distribution models (SDM). Nevertheless, these tree records are often spatially clumped and possibly only provide accurate predictions over areas close to where the training occurrence records are located. Here, we aim at investigating to what degree SDM performance and spatial predictions differ between models that have different spatial configurations of the occurrence data. For this, we divided the available occurrence data from a forest concession census in Peruvian Amazonia into different spatial configurations (narrow, elongated and compact), each of which contained approximately 20% of the full dataset. We then modelled the distributions of five tree taxa using Landsat data and elevation. More elongated configurations of the training data were more representative of the available environmental space, and also produced more robust SDMs. Average model performance (expressed as AUC) was 5% higher and variation in model performance 50% lower when elongated rather than compact configurations of training area were used. This confirms that covering only a small fraction of the environmental variability in the area of interest may lead to misleading SDM predictions, which needs to be taken into account when forest management decisions are based on SDMs.

Questions Does the floristic composition of trees differ between bamboo forests and adjacent non-bamboo forests? Can the degree of compositional differences be predicted from differences in canopy reflectance as measured by Landsat satellites? Are the results sensitive to different taxonomical data cleaning strategies, or to which tree-size class is considered? Are some tree taxa associated with either bamboo or non-bamboo forests? Location Peruvian Amazonia. Methods We used national forestry inventory data to characterise floristic composition of trees at 25 sites. Bamboo and non-bamboo forests were identified with a pixel-based time series analysis using the entire Landsat TM/ETM+ archive. To visualise floristic similarity among the plots, we used non-metric multidimensional scaling. Floristic differences between bamboo and non-bamboo forests were tested using an analysis of similarity (ANOSIM). Mantel tests were used to assess correlation between floristic turnover and differences in canopy reflectance. We tested the impact of applying three different taxonomic data cleaning strategies and of analysing different tree-size classes. Finally, we ran an indicator species analysis to identify taxa that were associated with either bamboo or non-bamboo forests. Results In floristic ordinations, bamboo-dominated forests appeared floristically distinct from non-bamboo forests regardless of the taxonomic cleaning strategy and tree-size class. This floristic separation was also confirmed through analysis of floristic similarity (ANOSIM). Turnover in floristic composition was strongly correlated with differences in the reflectance values of Landsat bands, especially when using genus-level rather than species-level data. Different palm taxa were associated with bamboo (Socratea) and non-bamboo forests (Iriartea). Conclusions Floristic differences of trees between bamboo-dominated and non-bamboo forests are consistent enough to be observable with different tree-size classes and taxonomic cleaning strategies. Although coarsening the taxonomic resolution to the genus level may hide ecological detail from the results, the fact that it allows keeping incompletely identified stems in the analyses seems to compensate for this.

The forests of Amazonia are among the most biodiverse plant communities on Earth. Given the immediate threats posed by climate and land-use change, an improved understanding of how this extraordinary biodiversity is spatially organized is urgently required to develop effective conservation strategies. Most Amazonian tree species are extremely rare but a few are common across the region. Indeed, just 227 ‘hyperdominant’ species account for >50% of all individuals >10 cm diameter at 1.3 m in height. Yet, the degree to which the phenomenon of hyperdominance is sensitive to tree size, the extent to which the composition of dominant species changes with size class and how evolutionary history constrains tree hyperdominance, all remain unknown. Here, we use a large floristic dataset to show that, while hyperdominance is a universal phenomenon across forest strata, different species dominate the forest understory, midstory and canopy. We further find that, although species belonging to a range of phylogenetically dispersed lineages have become hyperdominant in small size classes, hyperdominants in large size classes are restricted to a few lineages. Our results demonstrate that it is essential to consider all forest strata to understand regional patterns of dominance and composition in Amazonia. More generally, through the lens of 654 hyperdominant species, we outline a tractable pathway for understanding the functioning of half of Amazonian forests across vertical strata and geographical locations.

Questions Understanding spatial variation in floristic composition is crucial to quantify the extent, patchiness and connectivity of distinct habitats and their spatial relationships. Broad-scale variation in floristic composition and the degree of uniqueness of different regions remains poorly mapped and understood in several areas across the globe. We here aim to map vegetation heterogeneity in Amazonia. Location Middle Jurua river region, Amazonas State, Brazil. Methods We mapped four plant groups by applying machine learning to scale up locally observed community composition and using environmental and remotely sensed variables as predictors, which were obtained as GIS layers. To quantify how reliable our predictions were, we made an assessment of model transferability and spatial applicability. We also compared our floristic composition map to the official Brazilian national-level vegetation classification. Results The overall performance of our floristic models was high for all four plant groups, especially ferns, and the predictions were found to be spatially congruent and highly transferable in space. For some areas, the models were assessed not to be applicable, as the field sampling did not cover the spectral or environmental characteristics of those regions. Our maps show extensive habitat heterogeneity across the region. When compared to the Brazilian vegetation classification, floristic composition was relatively homogeneous within dense forests, while floristic heterogeneity in rainforests classified as open was high. Conclusion Our maps provide geoecological characterization of the regions and can be used to test biogeographical hypotheses, develop species distribution models and, ultimately, aid science-based conservation and land-use planning.